JP2013138395A - Image filtering device, image decoding device, image encoding device and data structure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accomplish an image filtering device and the like to which an adaptive offset filter reflected with characteristics of pixels is applied.SOLUTION: An image filtering device is configured to add an offset selected from among a plurality of offsets to each of pixel values of an input image comprising a plurality of unit regions. The image filtering device comprises: classification method selection means which selects from among a plurality of candidates a classification method as a method for determining one of a plurality of classes to which a pixel included in the unit region should be sorted; classification means which classifies the pixels in accordance with the classification method selected by the classification method selection means; offset determination method which determines for each unit region the offset to be added to a pixel value of the pixel classified by the classification means in accordance with the class to which the pixel is sorted; and filter means which adds the offset determined by the offset determination means to the pixel values of the pixels included in the unit region.

Description

  The present invention relates to an image filter device that performs image filtering. The present invention also relates to an image encoding device that encodes an image using an image filter and an image decoding device that decodes an image using an image filter. The present invention also relates to the data structure of encoded data.

  In order to efficiently transmit or record a moving image, a moving image encoding device (encoding device) that generates encoded data by encoding the moving image, and decoding by decoding the encoded data A video decoding device (decoding device) that generates an image is used. As a specific moving picture encoding method, for example, H.264 is used. H.264 / MPEG-4. A method adopted in AVC, a method adopted in KTA software, which is a codec for joint development in VCEG (Video Coding Expert Group), and a method adopted in TMuC (Test Model under Consideration) software, which is the successor codec And a method adopted in HM (HEVC TestModel) software.

  In such an encoding scheme, an image (picture) that constitutes a moving image is a slice obtained by dividing the image, a maximum coding unit (LCU: Large Coding Unit, tree block obtained by dividing the slice) Hierarchical structure consisting of coding units (CU: Coding Unit, also called coding nodes) obtained by dividing the maximum coding unit, and blocks and partitions obtained by dividing the coding unit In many cases, the block is encoded as a minimum unit.

  In such an encoding method, a prediction image is usually generated based on a local decoded image obtained by encoding and decoding an input image, and difference data between the prediction image and the input image is encoded. Is done. As methods for generating a predicted image, methods called inter-screen prediction (inter prediction) and intra-screen prediction (intra prediction) are known.

  In intra prediction, based on a locally decoded image in the same frame, predicted images in the frame are sequentially generated. Specifically, in intra prediction, usually, one prediction direction is selected from prediction directions included in a predetermined prediction direction (prediction mode) group for each prediction unit (for example, block), and A prediction pixel value on the prediction target region is generated by extrapolating the pixel value of the reference pixel in the locally decoded image in the selected prediction direction. Also, in inter prediction, by applying motion compensation using a motion vector to a reference image in a reference frame (decoded image) in which the entire frame is decoded, a predicted image in a prediction target frame is converted into a prediction unit ( For example, it is generated for each block).

  Non-Patent Document 1 describes an adaptive loop filter (also referred to as an “adaptive filter”) that is a subsequent stage of a deblocking filter that reduces block distortion of a decoded image and performs filter processing using adaptively determined filter coefficients. An adaptive offset filter (also referred to as “adaptive offset filter”) introduced in the preceding stage is disclosed. This adaptive offset filter adds an adaptively set offset to each pixel value of an image output from the deblocking filter.

  By providing such an adaptive offset filter, block distortion can be more effectively suppressed.

“WD4: Working Draft 4 of High-Efficiency Video Coding”, “JCTVC-F803-v5”, Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T SG16 WP3 and ISO / IEC JCT1 / SC29 / WG11, 6th Meeting: Torino, IT, 07/2011

  The above-described adaptive offset filter is applied using the same processing method regardless of whether the target pixel is a luminance pixel or a chrominance pixel. That is, the difference in characteristics between the luminance pixel and the color difference pixel is not reflected in the processing method. As a result, the encoding efficiency has been reduced. In other words, since the processing method does not correspond to the characteristics of the pixel, the encoding efficiency is reduced.

  The present invention has been made in view of the above problems, and an object thereof is to realize an image filter device and the like to which an adaptive offset filter reflecting pixel characteristics is applied.

  In order to solve the above problems, an image filter device according to the present invention is an image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image composed of a plurality of unit regions. A class classification method selecting means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are classified, and the class classification method According to the class classification method selected by the selection means, the class classification means for classifying the pixel, and the offset added to the pixel value of the pixel classified by the class classification means according to the class in which the pixel is classified The offset determining means for determining each unit area, and the offset determined by the offset determining means for the pixels included in the unit area Is characterized by comprising a filter means for adding a pixel value.

  According to the above configuration, the class of the target pixel is classified according to the class classification method selected by the class classification method selection unit. Then, an offset is added to the target pixel based on the classified class.

  Therefore, since the class classification method can be selected for each pixel, the class classification method can be selected according to the characteristics of the pixel. Therefore, classification can be performed by a classification method that reflects the characteristics of the pixels.

  In addition, the coding efficiency can be improved by using a class classification method that reflects pixel characteristics.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method from different candidate groups depending on whether a pixel value of a pixel included in the unit region indicates a color difference or luminance. You may do.

  According to the above configuration, the classification method candidates differ depending on whether the pixel value of the target pixel indicates a color difference (color difference pixel) or a luminance (luminance pixel).

  Thereby, the class classification method can be selected according to the characteristics of the color difference pixels and the luminance pixels. In addition, since the class classification method can be selected according to the characteristics of the color difference pixels and the luminance pixels, the encoding efficiency can be improved.

  In the image filter device according to the present invention, the different candidate group is a color difference candidate group selected by the class classification method selection unit when the pixel value of the pixel included in the unit region indicates a color difference. When the candidate group selected when the pixel value of the pixel included in the unit region indicates luminance is a luminance candidate group, the number of classes indicated by the class classification method included in the color difference candidate group is the luminance candidate group. It may be more than the number of classes indicated by the included class classification method.

  According to the above configuration, the number of classes included in the chrominance pixel class classification method is larger than the number of classes included in the luminance pixel class classification method.

  Thereby, in the case of a color difference pixel, the case where an offset is added rather than a luminance pixel can be increased.

  In the image filter device according to the present invention, the different candidate group is a color difference candidate group selected by the class classification method selection unit when the pixel value of the pixel included in the unit region indicates a color difference. When the candidate group to be selected when the pixel value of the pixel included in the unit region indicates luminance is a luminance candidate group, the number of types of class classification methods included in the color difference candidate group is the luminance candidate group. It may be less than the number of types of classification methods included.

  According to the above configuration, the number of types of chrominance pixel class classification methods is smaller than the number of types of luminance pixel class classification methods.

  Thereby, in the case of a chrominance pixel, it is possible to reduce the amount of processing involved in processing all types of class classification methods.

  For example, when an offset type is considered as a type of class classification method, according to the above configuration, the number of offset types included in the color difference pixel class classification method is smaller than the number of offset types included in the luminance pixel class classification method. Become.

  Thereby, in the case of a chrominance pixel, the processing amount accompanying processing all the band offset types can be reduced.

  In the image filter device according to the present invention, the filter means selects the class classification method for classifying the target pixel based on the state of the edge around the pixel, and the target pixel If the pixel value indicates a color difference, if the pixel value of the target pixel indicates luminance, the offset may be added to the pixel value of a pixel classified into a class to which no offset is added.

  According to the experiment of the present inventor, in the case of a class classification method for performing class classification based on the state of the edge around the target pixel, even if the target pixel is a luminance pixel, even if it is a class that does not add an offset, It was found that encoding efficiency is improved by adding an offset.

  Therefore, according to the above configuration, in the class classification method that performs class classification based on the state of the edge around the target pixel, when the pixel value of the target pixel indicates color difference, the pixel value of the target pixel indicates luminance. If there is a class for which no offset is added, the offset is added, so that the coding efficiency can be improved.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates. Two or more of the candidates may be overlapped in the range of pixel values classified into the class for which the offset addition process is performed.

  According to the above configuration, there is a high possibility that pixel values included in the overlapping range are classified into a class to which an offset is added. Therefore, the image quality and the like can be improved by including the pixel values that are highly likely to affect the image quality and the like in the overlapping range.

  In the image filter device according to the present invention, when the pixel value of the target pixel indicates a color difference, the class classification method selection unit selects a class classification method candidate in which classes in which pixels in a predetermined range are classified overlap. A classification method may be selected from among them.

  According to said structure, possibility that it will be classified into the class with which the pixel of a predetermined value range overlaps becomes high. Therefore, there is a high possibility that an offset is added to the pixels in the predetermined range. If a value range having a large influence on the subjective image quality is set as the predetermined value range, the possibility of adding an offset to the predetermined value range is increased, so that the subjective image quality can be improved efficiently.

  The predetermined value range may be a value range indicating an achromatic color or low saturation. According to the above configuration, there is a high possibility that an offset is added to an achromatic pixel or a pixel with low saturation. Since achromatic or low-saturation pixels have a large effect on subjective image quality, the possibility of adding an offset to achromatic or low-saturation pixels is increased, so that subjective image quality can be improved. it can.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes a plurality of conversion tables in which pixel values and classes are associated with each other. Among the plurality of conversion tables, pixel values included in a predetermined range of at least two conversion tables are subjected to an offset addition process. It may be associated with the class to be performed.

  According to said structure, possibility that an offset will be added with respect to the pixel value contained in a predetermined range becomes high. Therefore, if a value range including a pixel value to which an offset is preferably added is set to a predetermined range, the possibility that the offset can be added to the pixel value can be increased.

  As the predetermined range, for example, a value range such as a halftone, a dark part, or a bright part, or a low saturation area in the case of a color difference can be set.

  In the image filter device according to the present invention, in the conversion table, the pixel value is divided into a plurality of value ranges, and each divided value range is associated with a class, and the predetermined range is more than the other classes. It may be a range of value ranges associated with a class with a small number of associated value ranges.

  According to said structure, the range of the value range matched with the class with few value ranges matched with another class becomes a predetermined range. A value range that is associated with a class that has fewer value ranges than other classes can be said to be a value range in which the classes are finely associated. The value range is considered as a value range in which the subjective image quality can be improved by adding an offset. And according to said structure, possibility of adding an offset to the pixel value of this range can be improved. Thereby, the possibility of improving the subjective image quality can be increased.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes a plurality of conversion tables in which pixel values and classes are associated with each other, and the plurality of conversion tables have different pixel value ranges associated with classes on which offset addition processing is performed. It may be.

  According to the above configuration, a conversion table having different pixel value ranges associated with a class for which offset addition processing is performed is included in one class classification method. The conversion table can be selected according to the above.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the conversion tables has pixel values divided into a plurality of value ranges, and each divided value range and class are associated with each other. The plurality of conversion tables may be different from each other in the number of value ranges associated with each class.

  According to the above configuration, a conversion table with different number of value ranges associated with each class is included in one class classification method. Therefore, it is appropriate even when it is preferable to classify more finely. Conversion table can be provided.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes three or more conversion tables in which pixel values and classes are associated with each other, and the three or more conversion tables include a range of pixel values corresponding to a class on which offset addition processing is performed, A conversion table wider than other conversion tables, a conversion table whose pixel value range corresponding to the class for which offset addition processing is performed is narrower than the pixel value range in the wide conversion table, and a class for which offset addition processing is performed And the corresponding pixel value range includes both ends of the pixel value range and a conversion table close to the range. There.

  According to the above configuration, in one class classification method, a conversion table in which the range of pixel values corresponding to a class for which offset addition processing is performed is wider than other conversion tables, and a class for which offset addition processing is performed. The pixel value range corresponding to the conversion table is narrower than other conversion tables, and the pixel value range corresponding to the class for which the offset addition process is performed is the range of the pixel value range and the adjacent range. Conversion table.

  Thereby, it can respond to distribution of various pixel values. For example, an appropriate conversion table can be provided even when the pixel values are over the entire value range or are partially biased.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes three or more conversion tables in which pixel values and classes are associated with each other, and the three or more conversion tables include a range of pixel values corresponding to a class on which offset addition processing is performed, A first conversion table including the center of the range of pixel values, a second conversion table in which the range of pixel values corresponding to the class for which the offset addition processing is performed includes both ends of the range of pixel values and a range close thereto. One or more pixel value ranges corresponding to the class for which the offset addition processing is performed include ranges other than the pixel value ranges in the first conversion table and the second conversion table. And conversion table may be one that contains.

  According to the above configuration, in one class classification method, the range of pixel values corresponding to the class for which offset addition processing is performed includes the first conversion table including the center of the range of pixel values, and the offset addition processing includes The pixel value range corresponding to the class to be performed is one or more conversion tables not including both ends and the center of the pixel value range, and the pixel value range corresponding to the class to be subjected to the offset addition process is the pixel value range. And a second conversion table including both ends of the range and a range close to the range.

  Thus, an appropriate conversion table can be provided even when pixel values are not widely distributed and are likely to be concentrated in any range.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes three or more conversion tables in which pixel values and classes are associated with each other, and an offset addition process is performed on the code length of encoded data indicating one of the three or more conversion tables. The pixel value range corresponding to the class is the shortest when the pixel value range corresponding to the class indicates a conversion table including the center of the pixel value range, and the pixel value range corresponding to the class to which the offset addition processing is performed is at both ends of the pixel value range and its range. When a conversion table including the vicinity is shown, encoded data that is the longest may be used.

  According to the above configuration, the code length in the case of indicating any one of the three or more conversion tables in the encoded data is such that the pixel value range corresponding to the class for which the offset addition processing is performed is the center of the pixel value range. Is the shortest when a conversion table is included, and the pixel value range corresponding to the class for which the offset addition processing is performed is the longest when the conversion table includes both ends of the pixel value range and the vicinity thereof.

  Thereby, when selecting the conversion table corresponding to the band including the center of the range, the code amount indicating the conversion table can be reduced.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, The candidates include a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, and each of the divided value ranges and classes. And the class classification method selecting means classifies the pixel values into a larger number of value ranges than when the pixel values indicate luminance when the pixel values indicate color differences. The classification method including the converted table may be selected.

  According to the above configuration, when the pixel value indicates a color difference, the classification is performed using the conversion table divided into a larger number of value ranges than when the pixel value indicates the luminance. Done. Thereby, when the pixel value indicates a color difference, a finer classification is possible than when the pixel value indicates luminance. Since the pixel values are often concentrated in a narrower range than the luminance, the above configuration allows the pixel values of the color difference to be classified more accurately and efficiently. Efficiency can be improved.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, The candidates include a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, and each of the divided value ranges and classes. If the pixel value indicates color difference, the class classification method selection unit is configured such that the range of pixel values corresponding to the class for which the offset addition process is not performed is at both ends of the pixel value range. A class classification method including a conversion table that is a range close to this may be selected.

  According to the above configuration, when the pixel value indicates a color difference, the conversion table is used in which the pixel value range corresponding to the class for which the offset addition process is not performed is the both ends of the pixel value range and the value range close thereto. Classification. This is because the color difference pixel value is frequently generated near the center of the range. With the above configuration, pixel values of color differences can be classified more accurately and efficiently, and encoding efficiency can be improved.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, The candidates include a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, and each of the divided value ranges and classes. When the pixel value indicates a color difference, the class classification method selection unit has a pixel value range corresponding to the class for which the offset addition process is performed at the center of the pixel value range. A first conversion table including a third conversion table in which a range of pixel values corresponding to a class for which an offset addition process is performed does not include the center of the range of pixel values; In addition, a class classification method in which the range of pixel values corresponding to the class for which the offset addition processing is performed in the first conversion table is narrower than the range of pixel values corresponding to the class for which the offset addition processing is performed in the third conversion table. May be selected.

According to the above configuration, when the pixel value indicates a color difference, the offset is added to the range including the center of the pixel value range and the range not including the center of the pixel value range, and the center of the pixel value range is The range including it is narrower than the range not including the center of the pixel value range.
The pixel value of the color difference tends to occur frequently in the vicinity of the center of the value range representing low saturation. Therefore, with the above configuration, it is possible to add an offset to a relatively wide range of pixel values while finely classifying the vicinity of a value range where the color difference occurrence frequency is high.

  In the image filter device according to the present invention, the class classification method selection unit selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and each of the candidates Includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables includes pixel values divided into a plurality of value ranges, and each of the divided value ranges and classes includes The plurality of conversion tables that are associated with each other include a conversion table in which a plurality of non-contiguous ranges in the pixel value range become pixel value ranges corresponding to a class for which offset addition processing is performed. It may be what is.

  According to the above configuration, the class classification method includes a conversion table in which a plurality of non-contiguous ranges in the pixel value range are pixel value ranges corresponding to the class on which the offset addition processing is performed. It is. As a result, a large number of pixel values can be classified with a small number of classes, so that the amount of code required for class classification can be reduced and the coding efficiency can be improved.

  In order to solve the above problems, an image filter device according to the present invention is an image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image composed of a plurality of unit regions. A class classification method selecting means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are classified, and the class classification method According to the class classification method selected by the selection means, the class classification means for classifying the pixel, and the offset added to the pixel value of the pixel classified by the class classification means according to the class in which the pixel is classified The offset determining means for determining each unit area, the upper limit value and the lower limit value of the pixel value of the pixel, and the pixels having pixel values close to them. The offset correction means for correcting the offset to be added so that the pixel value after adding the offset determined by the offset determination means does not exceed the upper limit value or falls below the lower limit value, and the offset determination means And a filter unit that adds the determined offset or the offset corrected by the offset correction unit to the pixel value of the pixel included in the unit region.

  According to said structure, when adding the determined offset to the pixel value of the said pixel, it can prevent that it protrudes from the range of a pixel value. Further, by correcting the offset, it is possible to prevent the pixel value from protruding from the range, so that subsequent clip processing can be reduced.

  In order to solve the above-described problem, an image decoding device according to the present invention is an image decoding device that generates a decoded image by adding a generated prediction image to a prediction residual decoded from encoded data. An offset addition unit that adds an offset for each unit region to the addition image obtained by adding the prediction image to the residual, and an output unit that outputs an image obtained by adding the offset by the offset addition unit as a decoded image, The offset addition means includes a class classification method selection means for selecting a class classification method, which is a method for determining which of a plurality of classes a pixel included in the unit area is to be classified from a plurality of candidates, Class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means, and the class classification means An offset to be added to the pixel value of each pixel is determined for each unit region according to the class into which the pixel is classified, and the offset determined by the offset determination unit is included in the unit region. And a filter means for adding to the pixel value of the pixel.

  According to the above configuration, the class of the target pixel is classified according to the class classification method selected by the class classification method selection unit. Then, an offset is added to the target pixel based on the classified class.

  Therefore, since the class classification method can be selected for each pixel, the class classification method can be selected according to the characteristics of the pixel. Therefore, classification can be performed by a classification method that reflects the characteristics of the pixels.

  In addition, the coding efficiency can be improved by using a class classification method that reflects pixel characteristics.

  In order to solve the above problems, an image encoding device according to the present invention is an image encoding device that generates encoded data by encoding a prediction residual that is a difference between a generated predicted image and an original image. An offset adding unit that adds an offset for each unit region to a decoded image that is encoded and decoded from an original image, and a predicted image generating unit that generates the predicted image from an image obtained by adding the offset by the offset adding unit Encoding means for encoding a prediction residual that is a difference between the prediction image generated by the prediction image generation means and the original image, and the offset addition means includes pixels included in the unit region. Class classification method selection means for selecting a class classification method, which is a method for determining which of a plurality of classes to classify, from a plurality of candidates, and the class classification method selection unit Class classification means for classifying the pixel according to the selected class classification method, and an offset to be added to the pixel value of the pixel classified by the class classification means according to the class in which the pixel is classified. An offset determining unit that determines each unit region and a filter unit that adds the offset determined by the offset determining unit to the pixel values of the pixels included in the unit region are provided.

  According to the above configuration, the class of the target pixel is classified according to the class classification method selected by the class classification method selection unit. Then, an offset is added to the target pixel based on the classified class.

  Therefore, since the class classification method can be selected for each pixel, the class classification method can be selected according to the characteristics of the pixel. Therefore, classification can be performed by a classification method that reflects the characteristics of the pixels.

  In addition, the coding efficiency can be improved by using a class classification method that reflects pixel characteristics.

  In order to solve the above problem, the data structure of the encoded data according to the present invention is the encoded data of the encoded data in which the prediction residual, which is the difference between the predicted image and the original image, is encoded by the image decoding apparatus. A data structure, comprising: pixel value data; and class classification method selection data for selecting a class classification method for determining an offset to be added to a pixel value indicated by the pixel value data, and the image decoding device Is characterized by classifying the class of the target pixel according to the class classification method indicated by the class classification method selection data and adding the offset corresponding to the class to the target pixel to generate a decoded image.

  According to the above configuration, the class of the target pixel is classified according to the class classification method indicated by the class classification method selection data. Then, an offset is added to the target pixel based on the classified class.

  Therefore, since the class classification method can be selected for each pixel, the class classification method can be selected according to the characteristics of the pixel. Therefore, classification can be performed by a classification method that reflects the characteristics of the pixels.

  In addition, the coding efficiency can be improved by using a class classification method that reflects pixel characteristics.

  In order to solve the above problems, an image filter device according to the present invention is an image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image, and for each pixel value, An offset selection method for determining an offset selection method for selecting an offset to be added to the pixel value from the plurality of offsets is determined by an offset selection method determination unit for determining from among a plurality of candidates and the offset selection method determination unit. An offset selection unit that selects an offset from the plurality of offsets according to an offset selection method, and a filter unit that adds the offset selected by the offset selection unit to the pixel value. .

  According to said structure, the offset selection method which is a method of selecting the offset added to a pixel value can be determined from several candidates. Thereby, the offset selection method can be determined for each pixel without adding the offset uniformly by the same method.

  Therefore, the offset selection method can be determined according to the characteristics of the pixel, and the offset can be added by the offset selection method reflecting the characteristics of the pixel.

  In addition, encoding efficiency can be improved by using an offset selection method that reflects pixel characteristics.

  As described above, the image filter device according to the present invention selects, from among a plurality of candidates, a class classification method that is a method for determining which of a plurality of classes a pixel included in a unit area is classified into. Class classification method selection means, class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means, and an offset to be added to the pixel values of the pixels classified by the class classification means An offset determining unit that determines each unit region according to a class into which the pixel is classified, and a filter unit that adds the offset determined by the offset determining unit to the pixel value of the pixel included in the unit region. It is the structure provided with.

  Thereby, there is an effect that the classification can be performed by the classification method reflecting the characteristics of the pixels.

  In addition, by using a class classification method that reflects the characteristics of the pixels, the encoding efficiency can be improved.

  The image decoding apparatus according to the present invention includes an offset addition unit that adds an offset for each unit region to an addition image obtained by adding the prediction image to a prediction residual, and an image obtained by adding the offset by the offset addition unit. Output means for outputting as a decoded image, and the offset adding means includes a plurality of class classification methods for determining which of the plurality of classes the pixels included in the unit area are to be classified. Class classification method selection means for selecting from among candidates, class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means, and pixels of the pixels classified by the class classification means An offset determining unit that determines an offset to be added to the value for each unit region according to a class in which the pixel is classified; An offset offset determining means has determined, a configuration in which a filter means for adding the pixel values of pixels included in the unit area, is provided.

  Thereby, there is an effect that the classification can be performed by the classification method reflecting the characteristics of the pixels.

  In addition, by using a class classification method that reflects the characteristics of the pixels, the encoding efficiency can be improved.

  The image coding apparatus according to the present invention encodes an original image and adds an offset for each unit area to a decoded image, and an image obtained by adding the offset by the offset adding unit. A prediction image generation unit that generates a prediction image; and an encoding unit that encodes a prediction residual that is a difference between the prediction image generated by the prediction image generation unit and the original image, and the offset addition unit includes A class classification method selection means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are to be classified, and the class classification method selection According to the class classification method selected by the means, the class classification means for classifying the pixel, and the pixel value of the pixel classified by the class classification means is added to the pixel value An offset determining unit that determines a facet for each unit region according to a class in which the pixel is classified, and a filter that adds the offset determined by the offset determining unit to the pixel value of the pixel included in the unit region Means.

  Thereby, there is an effect that the classification can be performed by the classification method reflecting the characteristics of the pixels.

  In addition, by using a class classification method that reflects the characteristics of the pixels, the encoding efficiency can be improved.

  The data structure of the encoded data according to the present invention includes pixel value data, class classification method selection data for selecting a class classification method for determining an offset to be added to the pixel value indicated by the pixel value data, and The image decoding device classifies the class of the target pixel according to the class classification method indicated by the class classification method selection data, and adds the offset corresponding to the class to the target pixel to generate a decoded image It is a configuration.

  Thereby, there is an effect that the classification can be performed by the classification method reflecting the characteristics of the pixels.

  In addition, by using a class classification method that reflects the characteristics of the pixels, the encoding efficiency can be improved.

  In addition, the image filter device according to the present invention determines, for each pixel value, an offset selection method that is a method for selecting an offset to be added to the pixel value from the plurality of offsets, from among a plurality of candidates. An offset selection method determination unit, an offset selection unit that selects an offset from the plurality of offsets according to the offset selection method determined by the offset selection method determination unit, and the offset selected by the offset selection unit And a filter means for adding to.

  Thus, the offset selection method can be determined according to the characteristics of the pixel, and the offset can be added by the offset selection method reflecting the characteristics of the pixel.

  In addition, by using an offset selection method that reflects the characteristics of the pixel, the encoding efficiency can be improved.

It is a block diagram which shows the structure of the adaptive offset filter which concerns on embodiment of this invention. It is a figure which shows the data structure of the encoding data produced | generated by the moving image encoder which concerns on the said embodiment, and decoded by a moving image decoder, (a)-(d) is a picture layer and a slice layer, respectively. FIG. 4B is a diagram illustrating a configuration of QAOU information related to a QAOU of a leaf, and FIG. 5F is a diagram illustrating a configuration of QAOU information related to a QAOU of a leaf. It is a figure which shows the syntax of the offset information which concerns on the said embodiment, and QAOU information, (a) is a figure which shows the syntax of the whole adaptive offset filter which calls the syntax shown in (b) and (c). (B) is a figure which shows the syntax of QAOU information, (c) is a figure which shows the syntax of offset information. It is a figure which shows the aspect of the division | segmentation of the offset unit which concerns on the said embodiment, (a) is a figure which shows the case where sao_depth = 0, (b) is a figure which shows the case where sao_depth = 1, (c) (A) is a figure which shows the case where sao_depth = 2, (d) is a figure which shows the case where sao_depth = 3, (e) is a figure which shows the case where sao_depth = 4. It is a block diagram which shows the structure of the moving image decoding apparatus which concerns on the said embodiment. It is a figure which shows the example of the conversion table which the offset information decoding part which concerns on the said embodiment uses. It is a figure for demonstrating the said embodiment, Comprising: (a) is a figure which shows each QAOMU of the division | segmentation depth 3 which comprises the object process unit, and the QAOU index allocated to each QAOMU, (b) FIG. 10 is a diagram showing an offset type associated with each of QAOU indexes 0 to 9 and an offset for each class selectable in each offset type. It is a block diagram which shows the structure of the class classification | category part which concerns on the said embodiment. It is a figure for demonstrating the offset process by the adaptive offset filter which concerns on the said embodiment, Comprising: (a)-(d) is a figure which shows the pixel referred when sao_type_idx = 1-4, respectively. It is a figure for demonstrating the offset process by the adaptive offset filter which concerns on the said embodiment, Comprising: (a) is the magnitude relationship between the pixel value pix [x] of the process target pixel x, and the pixel value of the pixel a or b FIG. 5B is a schematic diagram showing the magnitude relationship between the pixel value of the pixel x to be processed and the pixel values of the pixels a and b. It is a figure which shows the value of EdgeClass according to a figure and its magnitude relationship, (c)-(d) is a figure which shows the conversion table showing the conversion from EdgeClass to class_idx. It is a figure for demonstrating the offset process by the adaptive offset filter which concerns on the said embodiment, Comprising: (a) is a figure which shows a class classification | category when sao_type_idx = 5 schematically, (b) sao_type_idx = 6 It is a figure which shows roughly the class classification | category when it is. In the said embodiment, it is a figure which shows an example of the class classification when a band offset is designated, (a) is a figure which shows an example of the class classification in case a component is a brightness | luminance, (b) is a component. It is a figure which shows an example of the class classification | category when is a color difference. (C) is a figure which shows another example of the class classification in case a component is color difference, (d) is a figure which shows another example of the class classification in case a component is color difference. It is a flowchart which shows the flow of a process by the adaptive offset filter process part which concerns on the said embodiment. It is a block diagram which shows the structure of the offset information decoding part which concerns on the said embodiment. It is a block diagram which shows the structure of the moving image encoder which concerns on the said embodiment. It is a block diagram which shows the structure of the adaptive offset filter with which the moving image encoder which concerns on the said embodiment is provided. It is a flowchart which shows the flow of a process by the offset calculation part of the adaptive offset filter with which the moving image encoder which concerns on the said embodiment is provided. In the said embodiment, it is a figure which shows an example of the QAOMU number attached | subjected to QAOMU contained in the process unit of object, Comprising: (a) is a figure which shows the QAOMU number attached | subjected to QAOMU whose division depth is 0. (B) is a diagram showing a QAOMU number assigned to a QAOMU having a division depth of 1, and (c) is a diagram showing a QAOMU number attached to a QAOMU having a division depth of 2. (D) is a figure which shows the QAOMU number attached | subjected to QAOMU with a division | segmentation depth of 3, (e) is a figure which shows the QAOMU number attached | subjected to QAOMU with a division | segmentation depth of 4. FIG. It is a flowchart which shows the flow of a process by the offset information selection part of the adaptive offset filter with which the moving image encoder which concerns on the said embodiment is provided. In the said embodiment, it is a figure which shows the outline | summary which calculates the square error about each offset type about QAOU whose QAOU index is "q". It is a figure for demonstrating the process by the offset information selection part of the adaptive offset filter with which the moving image encoder which concerns on the said embodiment is provided, Comprising: (a) is the aspect of a division | segmentation in case the division | segmentation depth is 0 and 1 (B) is a diagram showing an aspect of division when the division depth is 1, and (c) is a diagram showing an aspect of division when the division depth is 2. (D) is a figure which shows an example of the division | segmentation determined by the offset information selection part. In the said embodiment, it is a figure which shows the example of the adaptive clip process of a pixel value. It is a figure for demonstrating that a moving image decoding apparatus and a moving image coding apparatus can be utilized for transmission / reception of a moving image, (a) is the block diagram which showed the structure of the transmission device carrying a moving image coding apparatus (B) is a block diagram showing a configuration of a receiving apparatus equipped with a moving picture decoding apparatus. It is a figure for demonstrating that a moving image decoding apparatus and a moving image encoding apparatus can be utilized for recording and reproduction | regeneration of a moving image, (a) shows the structure of the recording device carrying the moving image encoding apparatus 2. FIG. 8B is a block diagram illustrating a configuration of a playback device equipped with a video decoding device. It is a block diagram which shows the structure of the offset addition part with which the moving image decoding apparatus which concerns on the said embodiment is provided. It is a figure which shows the structure of the QAOU information regarding QAOU of the leaf of another example of the data structure of the coding data produced | generated by the moving image encoder which concerns on the said embodiment, and decoded by a moving image decoder. It is a figure for demonstrating the offset process by the adaptive offset filter with which the moving image decoding apparatus which concerns on the said embodiment is provided, Comprising: (a) is a figure which shows the range of the value added as offset, (b) is an offset. It is a figure which shows the case where a calculation result is less than the minimum of a pixel value by addition, (c) is a figure which shows the case where a calculation result exceeds the upper limit of a pixel value by offset addition. In the said embodiment, it is a figure which shows the example of a setting of an offset restriction area. In the said embodiment, it is a figure which shows another example of the class classification | category when band offset is designated, (a) is a figure which shows an example in case an offset restriction area | region is made into an invalid class (class 0). (B) is a figure which shows the example which uses together a prior art and the case where an offset restriction | limiting area | region is used. In the said embodiment, it is a figure which shows another example of a setting of an offset restriction area | region. (A)-(c) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. In the said embodiment, it is a figure which shows another example of the class classification | category when a band offset is designated. In the said embodiment, it is a figure which shows another example of the class classification | category when a band offset is designated. (A), (b) is a figure which shows the method of calculating a band index by another quantization method with a brightness | luminance and a color difference in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, In a color difference class classification, the range with a small value and the range with a large value It is a figure which shows the example which classifies | categorizes into the class which does not add offset. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, In a color difference class classification, the range with a small value and the range with a large value It is a figure which shows the example which classify | categorizes the value of a wide range in the same class. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and the center value and the pixel value of the vicinity of it in a color difference class classification | category are shown. It is a figure which shows the example classified more finely than the classification of the pixel value away from the center. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and the center value and the pixel value of the vicinity of it in a color difference class classification | category are shown. It is another figure which shows another example which classify | categorizes finer than the classification of the pixel value away from the center. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, In the class classification of a color difference, a center value and the pixel value of the vicinity are shown. It is another figure which shows the example which uses the class classification classified into the class which adds an offset as a top class classification of several class classification. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is a figure which shows the example classified into a class. (A), (b) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is another figure which shows the example classified into a class. (A)-(c) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is another figure which shows the example classified into a class. (A)-(c) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is another figure which shows the example classified into a class. (A)-(c) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is another figure which shows the example classified into a class. (A)-(c) is a figure which shows another example of the class classification | category in case the band offset is designated in the said embodiment, and classify | categorizes the pixel value away from the center value and its vicinity pixel value In the class classification, the quantized pixel value X1, pixel value X2, and pixel value X3 (X1 <X2 <X3) are different from the class that classifies X2 as pixel values X1 and X3 that are not adjacent to each other. It is another figure which shows the example classified into a class.

[Embodiment 1]
(Encoded data # 1)
Prior to detailed description of the video encoding device 2 and the video decoding device 1 according to the present embodiment, the encoded data # 1 generated by the video encoding device 2 and decoded by the video decoding device 1 is described. The data structure will be described.

  FIG. 2 is a diagram illustrating a data structure of the encoded data # 1. The encoded data # 1 exemplarily includes a sequence and a plurality of pictures constituting the sequence. 2A to 2D are included in the picture layer that defines the picture PICT, the slice layer that defines the slice S, the tree block layer that defines the tree block TBLK, and the tree block TBLK, respectively. It is a figure which shows the CU layer which prescribes | regulates a coding unit (Coding Unit; CU).

(Picture layer)
In the picture layer, a set of data referred to by the video decoding device 1 for decoding a picture PICT to be processed (hereinafter also referred to as a target picture) is defined. As shown in FIG. 2A, the picture PICT includes a picture header PH and slices S1 to SNS (NS is the total number of slices included in the picture PICT).

  Hereinafter, when it is not necessary to distinguish each of the slices S1 to SNS, the subscripts may be omitted. The same applies to other data with subscripts included in encoded data # 1 described below.

  The picture header PH includes a coding parameter group that is referred to by the video decoding device 1 in order to determine a decoding method of the target picture. For example, the encoding mode information (entropy_coding_mode_flag) indicating the variable length encoding mode used in encoding by the moving image encoding device 2 is an example of an encoding parameter included in the picture header PH.

  When entropy_coding_mode_flag is 0, the picture PICT is encoded by CAVLC (Context-based Adaptive Variable Length Coding). When entropy_coding_mode_flag is 1, the picture PICT is encoded by CABAC (Context-based Adaptive Binary Arithmetic Coding).

  Note that the picture PICT may include a picture parameter set (PPS) in addition to the picture header PH.

(Slice layer)
In the slice layer, a set of data referred to by the video decoding device 1 for decoding the slice S to be processed (also referred to as a target slice) is defined. As shown in FIG. 2B, the slice S includes a slice header SH and a sequence of tree blocks TBLK1 to TBLKNC (NC is the total number of tree blocks included in the slice S).

  The slice header SH includes an encoding parameter group that is referred to by the video decoding device 1 in order to determine a decoding method of the target slice. Slice type designation information (slice_type) for designating a slice type is an example of an encoding parameter included in the slice header SH.

  As slice types that can be specified by the slice type specification information, (1) I slice using only intra prediction at the time of encoding, (2) P slice using unidirectional prediction or intra prediction at the time of encoding, (3) B-slice using unidirectional prediction, bidirectional prediction, or intra prediction at the time of encoding may be used.

  In addition, the slice header SH includes a filter parameter FP that is referred to by an adaptive filter included in the video decoding device 1. The filter parameter FP may be included in the picture header PH or the picture parameter set PPS.

(Tree block layer)
In the tree block layer, a set of data referred to by the video decoding device 1 for decoding a processing target tree block TBLK (hereinafter also referred to as a target tree block) is defined. Note that a tree block may be referred to as a maximum coding unit (LCU).

  The tree block TBLK includes a tree block header TBLKH and coding unit information CU1 to CUNL (NL is the total number of coding unit information included in the tree block TBLK). Here, first, a relationship between the tree block TBLK and the coding unit information CU will be described as follows.

  The tree block TBLK is divided into units for specifying a block size used in each process of intra prediction or inter prediction and conversion.

  The unit of the tree block TBLK is divided by recursive quadtree division. The tree structure obtained by this recursive quadtree partitioning is hereinafter referred to as a coding tree.

  Hereinafter, a unit corresponding to a leaf which is a node at the end of the coding tree is referred to as a coding node. In addition, since the encoding node is a basic unit of the encoding process, hereinafter, the encoding node is also referred to as an encoding unit (CU).

  That is, encoding unit information (hereinafter referred to as CU information) CU1 to CUNL is information corresponding to each encoding node (coding unit) obtained by recursively dividing the tree block TBLK into quadtrees.

  Also, the root of the coding tree is associated with the tree block TBLK. In other words, the tree block TBLK is associated with the highest node of the tree structure of the quadtree partition that recursively includes a plurality of encoding nodes.

  Note that the size of the image area corresponding to each coding node is half the size of the coding node to which the coding node directly belongs (that is, the unit of the node one level higher than the coding node). .

  The size that each coding node can take depends on the size designation information and the maximum hierarchical depth of the coding node included in the sequence parameter set SPS of the coded data # 1. For example, when the size of the tree block TBLK is 64 × 64 pixels and the maximum hierarchical depth is 3, the coding nodes in the hierarchy below the tree block TBLK have four sizes, that is, 64 × 64. Any of pixel, 32 × 32 pixel, 16 × 16 pixel, and 8 × 8 pixel can be taken.

  As a block structure, a slice S is divided to form a tree block TBLK, and the tree block TBLK is divided to form a coding unit CU.

(Tree block header)
The tree block header TBLKH includes an encoding parameter referred to by the video decoding device 1 in order to determine a decoding method of the target tree block. Specifically, as shown in FIG. 2C, tree block division information SP_TBLK that designates a division pattern of the target tree block into each CU, and a quantization parameter difference that designates the size of the quantization step. Δqp (qp_delta) is included.

  The tree block division information SP_TBLK is information representing a coding tree for dividing the tree block. Specifically, the shape and size of each CU included in the target tree block, and the position in the target tree block Is information to specify.

  Note that the tree block division information SP_TBLK may not explicitly include the shape or size of the CU. For example, the tree block division information SP_TBLK may be a set of flags (split_coding_unit_flag) indicating whether or not the entire target tree block or a partial area of the tree block is divided into four. In this case, the shape and size of each CU can be specified by using the shape and size of the tree block together.

  The quantization parameter difference Δqp is a difference qp−qp ′ between the quantization parameter qp in the target tree block and the quantization parameter qp ′ in the tree block encoded immediately before the target tree block.

(CU layer)
In the CU layer, a set of data referred to by the video decoding device 1 for decoding a CU to be processed (hereinafter also referred to as a target CU) is defined.

  Here, before describing specific contents of data included in the CU information CU, a tree structure of data included in the CU will be described. The encoding node is the root of a prediction tree (PT) and a transform tree (TT). The prediction tree and the conversion tree are described as follows.

  In the prediction tree, the encoding node is divided into one or a plurality of prediction blocks, and the position and size of each prediction block are defined. In other words, the prediction block is one or a plurality of non-overlapping areas constituting the encoding node. The prediction tree includes one or a plurality of prediction blocks obtained by the above division.

  Prediction processing is performed for each prediction block. Hereinafter, a prediction block that is a unit of prediction is also referred to as a prediction unit (PU).

  Broadly speaking, there are two types of division in the prediction tree: intra prediction and inter prediction.

  In the case of intra prediction, there are 2N × 2N (the same size as the encoding node) and N × N division methods.

  In the case of inter prediction, there are 2N × 2N (the same size as the encoding node), 2N × N, N × 2N, N × N, and the like.

  In the transform tree, the encoding node is divided into one or a plurality of transform blocks, and the position and size of each transform block are defined. In other words, the transform block is one or a plurality of non-overlapping areas constituting the encoding node. The conversion tree includes one or a plurality of conversion blocks obtained by the above division.

  There are two types of division in the transformation tree: one in which an area having the same size as that of a coding node is assigned as a transformation block, and the other in division by recursive quadtree division as in the above-described division of a tree block.

  The conversion process is performed for each conversion block. Hereinafter, a transform block that is a unit of transform is also referred to as a transform unit (TU).

(Data structure of CU information)
Next, specific contents of data included in the CU information CU will be described with reference to FIG. As shown in FIG. 2D, the CU information CU specifically includes a skip flag SKIP, PT information PTI, and TT information TTI.

  The skip flag SKIP is a flag indicating whether or not the skip mode is applied to the target CU. When the value of the skip flag SKIP is 1, that is, when the skip mode is applied to the target CU, PT information PTI and TT information TTI in the CU information CU are omitted. Note that the skip flag SKIP is omitted for the I slice.

  The PT information PTI is information related to the PT included in the CU. In other words, the PT information PTI is a set of information regarding each of one or a plurality of PUs included in the PT, and is referred to when the moving image decoding apparatus 1 generates a predicted image. As shown in FIG. 2D, the PT information PTI includes prediction type information PType and prediction information PInfo.

  The prediction type information PType is information that specifies whether intra prediction or inter prediction is used as a predicted image generation method for the target PU.

  The prediction information PInfo is configured from intra prediction information or inter prediction information depending on which prediction method is specified by the prediction type information PType. Hereinafter, a PU to which intra prediction is applied is also referred to as an intra PU, and a PU to which inter prediction is applied is also referred to as an inter PU.

  Further, the prediction information PInfo includes information specifying the shape, size, and position of the target PU. As described above, the generation of the predicted image is performed in units of PU. Details of the prediction information PInfo will be described later.

  The TT information TTI is information regarding the TT included in the CU. In other words, the TT information TTI is a set of information regarding each of one or a plurality of TUs included in the TT, and is referred to when the moving image decoding apparatus 1 decodes residual data. Hereinafter, a TU may be referred to as a block.

  As shown in FIG. 2 (d), the TT information TTI includes TT division information SP_TT that designates a division pattern for each transform block of the target CU, and quantized prediction residuals QD1 to QDNT (NT is a target CU). The total number of blocks contained in).

  Specifically, the TT division information SP_TT is information for determining the shape and size of each TU included in the target CU and the position within the target CU. For example, the TT division information SP_TT can be realized from information (split_transform_unit_flag) indicating whether or not the target node is divided and information (trafoDepth) indicating the division depth. The TT partition information SP_TT is basically encoded for each node of the quadtree, but depends on restrictions on the transform size (maximum transform size, minimum transform size, maximum hierarchy depth of the quadtree). May be omitted and estimated.

  Each quantized prediction residual QD is encoded data generated by the video encoding device 2 performing the following processes 1 to 3 on a target block that is a processing target block.

Process 1: The prediction residual obtained by subtracting the prediction image from the encoding target image is orthogonally transformed (DCT (Discrete Cosine Transform) or DST (Discrete Sine Transform)) into the frequency domain;
Process 2: Quantize the transform coefficient obtained in Process 1;
Process 3: Variable length coding is performed on the transform coefficient quantized in Process 2;
The quantization parameter qp described above represents the magnitude of the quantization step QP used when the moving image encoding apparatus 2 quantizes the transform coefficient (QP = 2qp / 6).

(Prediction information PInfo)
As described above, there are two types of prediction information PInfo: inter prediction information and intra prediction information.

  The inter prediction information includes a coding parameter that is referred to when the video decoding device 1 generates an inter prediction image by inter prediction. More specifically, the inter prediction information includes inter PU division information that specifies a division pattern of the target CU into each inter PU, and inter prediction parameters for each inter PU.

  The inter prediction parameters include a reference image index, an estimated motion vector index, and a motion vector residual.

  On the other hand, the intra prediction information includes an encoding parameter that is referred to when the video decoding device 1 generates an intra predicted image by intra prediction. More specifically, the intra prediction information includes intra PU division information that specifies a division pattern of the target CU into each intra PU, and intra prediction parameters for each intra PU. The intra prediction parameter is a parameter for designating an intra prediction method (prediction mode) for each intra PU.

(Offset unit)
In the present embodiment, each picture or each slice is recursively divided into a plurality of offset units (unit area, also referred to as QAOU: Quad Adaptive Offset Unit) by a quadtree structure. Here, QAOU is a processing unit of offset filter processing by the adaptive offset filter according to the present embodiment.

  As shown in FIGS. 2E and 2F, the QAOU information that is information about each QAOU includes sao_split_flag that indicates whether or not the own (own QAOU) is further divided. When sao_split_flag included in a certain QAOU indicates that the QAOU is further divided (that is, when the QAOU is not a leaf), as shown in FIG. 2 (e), the QAOU The QAOU information regarding QAOU information regarding each of a plurality of QAOUs included in the QAOU.

  On the other hand, when the sao_split_flag included in a certain QAOU indicates that the QAOU is not further divided (that is, when the QAOU is a leaf), as shown in FIG. The QAOU information related to the QAOU includes offset information OI related to the QAOU. Further, as shown in FIG. 2F, the offset information OI includes offset type designation information OTI for designating an offset type and an offset group determined according to the offset type. Furthermore, as shown in FIG. 2F, the offset group includes a plurality of offsets.

  Note that the offset in the encoded data is a quantized value. Further, the offset in the encoded data may be a prediction residual obtained by using some prediction, for example, linear prediction.

  Next, the syntax regarding an adaptive offset filter is demonstrated with reference to FIG.

  FIG. 3A shows the syntax of the entire adaptive offset filter, which is a syntax for calling the syntax shown in FIGS. 3B and 3C. The parameter sao_flag is a flag for selecting whether or not to apply the adaptive offset filter. Only when the flag is true (ie, “1”), the QAOU subsequent to the flag corresponding to the adaptive offset filter for the luminance pixel is set. Information and offset information OI are stored. When sao_flag is true, whether or not to apply an adaptive offset filter to each component of the pixel data using the flags sao_flag_cb and sao_flag_cr corresponding to the color difference (Cb) and the color difference (Cr), respectively. select. If not applied, QAOU information and offset information corresponding to the component are not stored.

  Only when the application of the adaptive offset filter is specified by the flags sao_flag, sao_flag_cb, and sao_flag_cr, the syntax sao_split_param () and sao_offset_param () are called with the highest hierarchy specified for the corresponding component. Here, sao_split_param () represents the QAOU split structure, and sao_offset_param () represents the offset in each leaf QAOU using the split structure.

  Arguments given to the syntax of sao_split_param () and sao_offset_param () are component, sao_depth, ys, and xs. sao_depth is a parameter representing the QAOU division depth, and ys and xs are parameters for representing the QAOU (or the position of the QAOMU described later in the y direction and the position in the x direction, respectively). , Sao_depthL = 0, ys = 0, and xs = 0, where component is a parameter representing luminance (Y), color difference (Cb), and color difference (Cr), 0 for luminance (Y), color difference ( In the case of Cb), it is called as 1 and in the case of color difference (Cr), it is called 2. The value of component may be another value as long as the component to be processed can be distinguished.

  FIG. 3B is a diagram showing the syntax of the QAOU information (corresponding to sao_split_param () in FIG. 3B) described in FIG. 2E. As shown in the syntax 301B of FIG. 3B, if the division depth sao_depth is smaller than the maximum value set by a predetermined saoMaxDepth, whether or not the QAOU is further divided is selected by the parameter sao_split_flag. More specifically, sao_split_flag is designated as a flag for which QAOU by the above-mentioned arguments component, sao_depth, ys, xs, and is also expressed as sao_split_flag [component] [sao_depth] [ys] [xs] .

  When QAOU is divided, sao_split_param () of the next hierarchical depth is recursively called. When the division depth reaches the maximum value (sao_depth is not smaller than saoMaxDepth), 0 is set in sao_split_flag [component] [sao_depth] [ys] [xs], and no further division is performed.

  FIG. 3C is a diagram illustrating the syntax of the offset information. The block indicated by (cx1) in the syntax 301C in FIG. 3C indicates the syntax that is called when the above-described sao_split_flag is true, and corresponds to the case in FIG.

  On the other hand, the block indicated by (cx2) in the syntax 301C is a syntax called when the above-described sao_split_flag is not true, and corresponds to the case of FIG. As shown in the syntax 301C in FIG. 3C, the offset information OI includes a parameter sao_type_idx [component] [sao_depth] [ys] [xs] described later. Further, when the parameter sao_type_idx [component] [sao_depth] [ys] [xs] is not 0, the offset information OI includes a parameter sao_offset [component] [sao_depth] [ys] [xs] [i] to be described later. Yes.

  Also, Descriptor (descriptor) ue (v) shown in FIG. 3 indicates that the syntax associated with this descriptor is an unsigned numerical value, and the value is variable-length encoded. se (v) indicates that the syntax associated with this descriptor is a signed numerical value, and is variable-length encoded into a sign and an absolute value. ae (v) indicates that the syntax associated with this descriptor is variable-length encoded using an arithmetic code. “|” Indicates that any one of the encoding methods is selectively used.

  Next, an aspect of QAOU division according to the value of sao_depth will be described with reference to FIG. FIG. 4 is a diagram illustrating a QAOU division mode according to the value of sao_depth. (A) illustrates a division mode when sao_depth = 0, and (b) illustrates a division mode when sao_depth = 1. (C) shows the mode of division when sao_depth = 2, (d) shows the mode of division when sao_depth = 3, and (e) shows the mode of division when sao_depth = 4 ing.

  As shown in FIGS. 4A to 4E, each QAOU is specified by sao_depth and (xs, ys). As shown in FIG. 4A, when sao_depth = 0, xs and ys are 0, respectively. As shown in FIG. 4B, when sao_depth = 1, xs and ys are 0 and 0, respectively. It can take any value of 1. Further, as shown in FIG. 4C, when sao_depth = 2, xs and ys can take values of 0, 1, 2, and 3, respectively. In general, for a given sao_depth, xs and ys can each take a value from 0 to 2 sao_depth-1. Such a division mode is common to luminance (Y), color difference (Cb), and color difference (Cr).

(Sao_type_idx)
sao_type_idx [component] [sao_depth] [ys] [xs] corresponds to the offset type designation information OTI in FIG. 2 (f) described above, and is a parameter for designating an offset type for each QAOU. In the present embodiment, sao_type_idx [component] [sao_depth] [ys] [xs] may be simply referred to as sao_type_idx or an offset type. Even when simply called sao_type_idx, necessary individual values can be referred to as sao_type_idx [component] [sao_depth] [ys] [xs].

  There are two types of offset types, called edge offset processing (EO) and band offset processing (BO). The edge offset process refers to a process of adding any of a plurality of offsets to the pixel value of the processing target pixel in accordance with the difference between the pixel value of the processing target pixel and the pixel values of the surrounding pixels. Further, the band offset process refers to a process of adding any of a plurality of offsets to the pixel value of the processing target pixel according to the size of each pixel value to be processed. Specific contents of the edge offset process and the band offset process will be described later.

In the present embodiment, sao_type_idx takes an integer value from 0 to 6 except for a modification described later, and what kind of filter is applied to the pre-offset filter image (for example, a deblocked decoded image P_DB described later). Indicates whether processing is performed or filtering processing is not performed. Specifically, it is as follows.
Sao_type_idx = 0 indicates that the offset filter processing is not performed on the pre-offset filter image in the target QAOU.
Sao_type_idx = 1 to 4 indicates that edge offset processing (EO) is performed on the pre-offset filter image in the target QAOU. Furthermore, the offset types indicated by sao_type_idx = 1 to 4 are sequentially called “EO_0”, “EO_1”, “EO_2”, and “EO_3”.
Sao_type_idx = 5 or 6 indicates that band offset processing (BO) is performed on the pre-offset filter image in the target QAOU. Furthermore, the offset types indicated by sao_type_idx = 5 or 6 are called “BO — 0” and “BO — 1” in order.

  In some modified examples described later, more offset types than those described above may be used for the edge offset and the band offset. For example, when three or more band offset types are used, the offset type names are distinguished by changing the subscripts so as to be “BO_2” and “BO_3” in order. Furthermore, when the number of offset types is different from that before the deformation as described above, the range of the value of sao_type_idx changes according to the total number of offset types. For example, when the total number of offset types is 9, sao_type_idx takes an integer value from 0 to 8.

(Sao_offset)
sao_offset [component] [sao_depth] [ys] [xs] [i] is a parameter representing a specific value of an offset added to each pixel included in the target QAOU in the adaptive offset filter processing according to the present embodiment. is there. In the present embodiment, sao_offset [component] [sao_depth] [ys] [xs] [i] may be simply referred to as sao_offset or offset. Even in the case of simply calling sao_offset, the necessary individual values can be referred to as sao_offset [component] [sao_depth] [ys] [xs] [i].

  sao_offset [component] [sao_depth] [ys] [xs] [i] is specified by an index “i” in addition to the above-described arguments component, sao_depth, ys, and xs. Here, the index i is an index for designating a class, and is also expressed as class_idx. The index i takes an integer value of 0 to Nec if the number of non-zero classes belonging to the type to be referenced is Nec in the case of an edge offset. For example, in the case of Nec = 5, i takes any integer value from 0 to 5. In the case of a band offset, if the number of non-zero classes belonging to the type to be referenced is Nbc, an integer value of 0 to Nbc is taken. For example, if the type is Nbc = 16, i takes any integer value from 0 to 16, and if Nbc = 8, i takes any integer value from 0 to 8. In any case, when i is 0, it represents a class to which no offset is added. For this reason, when i is 0, no offset is encoded. Hereinafter, the class with i = 0 is also referred to as an invalid class. A class to which an offset is added is also referred to as an effective class. The maximum value of i is determined by the number of effective classes belonging to each offset type, and is referred to from a predetermined value array saoNumClass common to the encoder and decoder, with (component, sao_type_idx) as a subscript.

  As will be described later, the adaptive offset filter according to the present embodiment classifies the target pixel included in the target QAOU into one of the plurality of classes, and classifies the target pixel with respect to the target pixel. Add the offset of.

(Moving picture decoding apparatus 1)
Next, the video decoding device 1 according to the present embodiment will be described with reference to FIG. 1 and FIGS. The moving picture decoding apparatus 1 includes H.264 as a part thereof. H.264 / MPEG-4. A method adopted in AVC, a method adopted in KTA software, which is a codec for joint development in VCEG (Video Coding Expert Group), and a method adopted in TMuC (Test Model under Consideration) software, which is the successor codec And the technology employed in HM (HEVC TestModel) software.

  FIG. 5 is a block diagram showing a configuration of the moving picture decoding apparatus 1. As shown in FIG. 5, the moving picture decoding apparatus 1 includes a variable length code decoding unit 13, a motion vector restoration unit 14, a buffer memory 15, an inter prediction image generation unit 16, an intra prediction image generation unit 17, and a prediction method determination unit 18. , An inverse quantization / inverse transform unit 19, an adder 20, a deblocking filter 41, an adaptive filter 50, and an adaptive offset filter 60. The moving picture decoding apparatus 1 is an apparatus for generating moving picture # 4 by decoding encoded data # 1.

  The variable length code decoding unit 13 decodes the prediction parameter PP for each partition from the encoded data # 1. That is, for the inter prediction partition, the reference image index RI, the estimated motion vector index PMVI, and the motion vector residual MVD are decoded from the encoded data # 1, and these are supplied to the motion vector restoration unit 14. On the other hand, with respect to the intra prediction partition, (1) size designation information for designating the partition size and (2) prediction index designation information for designating the prediction index are decoded from the encoded data # 1, and this is decoded into the intra prediction image. This is supplied to the generation unit 17. In addition, the variable length code decoding unit 13 decodes the CU information from the encoded data, and supplies this to the prediction method determination unit 18 (not shown). Furthermore, the variable length code decoding unit 13 decodes the quantization prediction residual QD for each block and the quantization parameter difference Δqp for the tree block including the block from the encoded data # 1, and dequantizes and decodes them. This is supplied to the inverse conversion unit 19. Further, the variable length code decoding unit 13 extracts an FP (filter parameter) from the encoded data # 1, and supplies the extracted FP to the adaptive filter 50. Further, the variable length code decoding unit 13 extracts QAOU information from the encoded data # 1 and supplies the extracted QAOU information to the adaptive offset filter 60.

  The motion vector restoration unit 14 restores the motion vector mv for each inter prediction partition from the motion vector residual MVD for the partition and the restored motion vector mv ′ for the other partitions. Specifically, (1) the estimated motion vector pmv is derived from the restored motion vector mv ′ according to the estimation method specified by the estimated motion vector index PMVI, and (2) the derived estimated motion vector pmv and the motion vector remaining are derived. The motion vector mv is obtained by adding the difference MVD. It should be noted that the restored motion vector mv ′ relating to other partitions can be read from the buffer memory 15. The motion vector restoration unit 14 supplies the restored motion vector mv together with the corresponding reference image index RI to the inter predicted image generation unit 16. For the inter prediction partition that performs bi-directional prediction (weighted prediction), the restored two motion vectors mv1 and mv2 are supplied to the inter predicted image generation unit 16 together with the corresponding reference image indexes RI1 and RI2.

  The inter prediction image production | generation part 16 produces | generates the motion compensation image mc regarding each inter prediction partition. Specifically, using the motion vector mv supplied from the motion vector restoration unit 14, the motion compensated image mc from the filtered decoded image P_FL ′ designated by the reference image index RI also supplied from the motion vector restoration unit 14 is used. Is generated. Here, the filtered decoded image P_FL ′ is obtained by performing deblocking processing by the deblocking filter 41, offset filtering processing by the adaptive offset filter 60, and adaptive filtering processing by the adaptive filter 50 on the local decoded image P. This is an obtained image, and the inter predicted image generation unit 16 can read out the pixel value of each pixel constituting the filtered decoded image P_FL ′ from the buffer memory 15. The motion compensation image mc generated by the inter prediction image generation unit 16 is supplied to the prediction method determination unit 18 as an inter prediction image Pred_Inter. For the inter prediction partition that performs bi-directional prediction (weighted prediction), (1) a motion compensated image mc1 is generated from the filtered decoded image P_FL1 ′ specified by the reference image index RI1 using the motion vector mv1. (2) A motion compensated image mc2 is generated from the filtered decoded image P_FL2 ′ specified by the reference image index RI2 using the motion vector mv2, and (3) a weighted average of the motion compensated image mc1 and the motion compensated image mc2 The inter prediction image Pred_Inter is generated by adding the offset value to the.

  The intra predicted image generation unit 17 generates a predicted image Pred_Intra related to each intra prediction partition. Specifically, first, a prediction mode is derived with reference to prediction index designation information and size designation information decoded from encoded data # 1, and the prediction mode is assigned to a target partition in, for example, raster scan order. . Subsequently, a predicted image Pred_Intra is generated from the local decoded image P (not shown) read from the buffer memory according to the prediction method indicated by the prediction mode. The intra predicted image Pred_Intra generated by the intra predicted image generation unit 17 is supplied to the prediction method determination unit 18.

  In addition, the intra predicted image generation unit 17 supplies intra coding mode information IEM, which is information indicating the size of the target partition and the prediction mode assigned to the target partition, to the adaptive filter 50 (not shown).

  Based on the CU information, the prediction method determination unit 18 determines whether each partition is an inter prediction partition that should perform inter prediction or an intra prediction partition that should perform intra prediction. In the former case, the inter predicted image Pred_Inter generated by the inter predicted image generation unit 16 is supplied to the adder 20 as the predicted image Pred. In the latter case, the inter predicted image generation unit 17 generates the inter predicted image Pred_Inter. The intra predicted image Pred_Intra that has been processed is supplied to the adder 20 as the predicted image Pred.

  The inverse quantization / inverse transform unit 19 (1) inversely quantizes the quantized prediction residual QD, and (2) uses a DCT (Discrete Cosine Transform) coefficient or a DST (Discrete Sine Transform) coefficient obtained by the inverse quantization. Inverse DCT transform or inverse DST transform is performed, and (3) prediction residual D obtained by inverse DCT transform or inverse DST transform is supplied to adder 20. When the quantization prediction residual QD is inversely quantized, the inverse quantization / inverse transform unit 19 derives the quantization step QP from the quantization parameter difference Δqp supplied from the variable length code decoding unit 13. The quantization parameter qp can be derived by adding the quantization parameter difference Δqp to the quantization parameter qp ′ relating to the tree block that has been subjected to the inverse quantization / inverse DCT (or DST) transformation immediately before, and the quantization step QP is performed by the quantization step QP. It can be derived from step qp by QP = 2pq / 6. The generation of the prediction residual D by the inverse quantization / inverse transform unit 19 is performed in units of blocks (transform units).

  The adder 20 generates the local decoded image P by adding the prediction image Pred supplied from the prediction method determination unit 18 and the prediction residual D supplied from the inverse quantization / inverse conversion unit 19.

  The deblocking filter 41 determines that there is block distortion when the difference between pixel values of pixels adjacent to each other via a block boundary or a CU boundary in the local decoded image P is within a predetermined range. By performing deblocking processing on the block boundary or the CU boundary in the decoded image P, the image near the block boundary or the CU boundary is smoothed. The image subjected to the deblocking process by the deblocking filter 41 is output to the adaptive offset filter 60 as a deblocked decoded image P_DB.

  The adaptive offset filter 60 performs an offset filter process using an offset decoded from the encoded data # 1 on the deblocked decoded image P_DB supplied from the deblocking filter 41, using QAOU as a processing unit. An offset filtered decoded image P_OF is generated. The generated offset filtered decoded image P_OF is supplied to the adaptive offset filter 60. Since a specific configuration of the adaptive offset filter 60 will be described later, description thereof is omitted here.

  The adaptive filter 50 applies the filter parameter FP (filter set number, filter coefficient group, region designation information, and on / off) decoded from the encoded data # 1 to the offset filtered decoded image P_OF supplied from the adaptive offset filter 60. The filtered decoded image P_FL is generated by performing a filtering process using (information). The filtered decoded image P_FL is output to the outside as the moving image # 4, and is stored in the buffer memory 15 in association with the POC designation information decoded from the encoded data by the variable length code decoding unit 13.

(Adaptive offset filter 60)
Next, details of the adaptive offset filter 60 will be described with reference to FIG. FIG. 1 is a block diagram showing the configuration of the adaptive offset filter 60. As shown in FIG. 1, the adaptive offset filter 60 includes an adaptive offset filter information decoding unit 61 and an adaptive offset filter processing unit 62.

(Adaptive offset filter information decoding unit 61)
As shown in FIG. 1, the adaptive offset filter information decoding unit 61 includes an offset information decoding unit 611 and a QAOU structure decoding unit 612.

  The offset information decoding unit 611 refers to the QAOU information included in the encoded data # 1, and decodes the offset information OI included in the QAOU information. Also, the values of sao_type_idx [component] [sao_depth] [ys] [xs] and sao_offset [component] [sao_depth] [ys] [xs] [i] obtained by decoding the offset information OI are The offset information is stored in the offset information storage unit 621 in association with the arguments (component, sao_depth, ys, xs) and (component, sao_depth, ys, xs, i).

  More specifically, the offset information decoding unit 611 decodes the code from the encoded data # 1, converts the decoded code into a sao_type_idx value, and supplies the value to the offset information storage unit 621 in association with the argument. Here, the offset information decoding unit 611 changes the code decoding method according to the pixel component of the QAOU to be processed. The condition such as the QAOU component is called a parameter condition. Decoding of offset information according to general parameter conditions will be described later with reference to FIG.

  The code decoding method may be different depending on whether the processing target is a luminance QAOU or a color difference QAOU. If they are different, for example, those using different binarization for luminance and color difference or those using different contexts may be used.

  Also, the offset information decoding unit 611 performs conversion from the decoded code to the value of sao_type_idx using a conversion table 801A as shown in FIG. The conversion table 801A in FIG. 6 shows an example of a conversion pattern from a code to an offset type. The offset information decoding unit 611 converts the decoded code into an offset type using the conversion table 801A. In the conversion table 801A of FIG. 6, the code “1” is associated with the offset type “EO — 0” (sao_type_idx = 1), and the code “2” is associated with the offset type “EO — 1” (sao_type_idx = 2). The offset type “EO_2” (sao_type_idx = 3) is associated with the code “3”. The same applies to the codes 4 to 6 below.

  The association between the code and the offset type is not limited to that shown in the conversion table 801A. As shown in the conversion table 801B in FIG. 6, an offset type having a higher appearance frequency is associated with a code having a shorter bit length, or an unused offset type is omitted as shown in the conversion table 801C. Or you may. Also, these conversion tables may be dynamically changed or switched.

  The QAOU structure decoding unit 612 determines the QAOU partition structure by decoding sao_split_flag [component] [sao_depth] [ys] [xs] included in the QAOU information, and represents the determined QAOU partition structure The QAOU structure information is supplied to the offset information storage unit 621.

(Adaptive offset filter processing unit 62)
As shown in FIG. 1, the adaptive offset filter processing unit 62 includes an offset information storage unit 621, a QAOU control unit 622, an offset type derivation unit 623, a class classification unit 624, an offset attribute setting unit 625, an offset derivation unit 626, and an offset. This is a configuration including an adder 627.

  The offset information storage unit 621 specifies each QAOU based on the QAOU structure information, sao_type_idx [component] [sao_depth] [ys] [xs], and sao_offset [component] [sao_depth] [ys] [xs] [i] And a specific value of an offset for each class selectable in the offset type, and is configured to include a map memory and a list memory.

  The map memory and list memory will be described with reference to FIG. FIG. 7 is a diagram illustrating examples stored in the map memory and the list memory, and FIG. 7A is a diagram for explaining an example of an identification number (QAOU index) of the QAOU stored in the map memory. FIG. 7B is a diagram for explaining an example of information stored in the list memory.

  The map memory 601 stores a QAOU index assigned to a minimum offset unit (also referred to as QAOMU: Quad Adaptive Offset Minimum Unit) with respect to the maximum division depth. In other words, the map memory 602 represents the QAOU division structure using the QAOMU for the maximum division depth.

  FIG. 7A shows each QAOMU having a division depth 3 constituting a target processing unit (for example, LCU) and a QAOU index including each QAOMU. In FIG. 7A, the indexes 0 to 9 are simply assigned to the QAOUs by serial numbers. However, in practice, it is preferable that the division depth can be derived from the QAOU index by an easy means such as table reference. In the example shown in FIG. 7A, QAOUs designated by QAOU index = I are represented as QAOU0 to QAOU9. In addition, the thin line in FIG. 7A indicates the QAOMU boundary, and the thick line indicates the QAOU boundary.

  The QAOU0 shown in FIG. 7A is composed of four QAOMUs with the maximum division depth, and 0 is assigned to these four QAOMUs as the QAOU index. On the other hand, QAOU3 is composed of one QAOMU, and 3 is assigned to this QAOMU as a QAOU index. In this way, the map memory stores the QAOU index assigned to each QAOMU.

  The list memory 602 stores, for each QAOU index, an offset type associated with the QAOU index and a specific value of an offset for each class selectable in the offset type in association with each other. .

  Specifically, this will be described with reference to FIG. FIG. 7B shows an offset type associated with each of the QAOU indexes 0 to 9 and an offset for each class selectable in each offset type. “Xxx” in FIG. 7B represents specific numerical values that may be different from each other.

  As shown in FIG. 7B, when the offset type is a band offset, the number of offsets stored in the list memory with respect to the offset type is that the number of effective classes of the band offset in the illustrated case is 16. Thus, a total of 16 offsets 1 to 16 are provided. Here, offset 1 to offset 16 are sao_offset [component] [sao_depth] [ys] [xs] [1] when the value of sao_type_idx [component] [sao_depth] [ys] [xs] is 5 or 6. ~ Sao_offset [component] [sao_depth] [ys] [xs] It refers to the value specified by [16].

  On the other hand, when the offset type is an edge offset, the number of offsets stored in the list memory with respect to the offset type is the same as the number of valid classes of the edge offset, which is a total of five offsets 1 to 5. Here, the offsets 1 to 5 indicate that sao_type_idx [component] [sao_depth] [ys] [xs] is any value from 1 to 4, and sao_offset [component] [sao_depth] [ys] [xs] [ 1] to sao_offset [component] [sao_depth] [ys] [xs] [5] Refers to the values specified by each. Note that nothing is stored in the memory of the offset that is not used. In addition, it is possible to omit securing an offset memory that is not always used, thereby reducing the amount of memory used.

  In the present embodiment, the number of offsets may be varied depending on the component represented by the QAOU to be processed, that is, whether the pixel is a luminance pixel or a color difference pixel. For example, for the luminance pixel, the number of effective classes of edge offsets, that is, the number of offsets is set to four offsets 1 to 4, and for the color difference pixels, a fifth offset described later is used, and five offsets 1 to 5 are used. Specifically, when sao_type_idx is 1 to 4 (edge offset), a value obtained by referring to saoNumClass in FIG. 3C is 4 for component = 0 (luminance), component = 1 or 5 for 2 (color difference).

  According to the inventor's experiment, encoding processing is not improved for luminance pixels by performing offset processing even on pixels classified as class 0 that are not subjected to offset processing in the prior art, but color difference pixels It was found that the coding efficiency is improved.

  Therefore, the coding efficiency can be improved by making the number of classes used for the edge offset different between the luminance pixel and the color difference pixel as described above. Therefore, as described above, it is preferable that at least the color difference is processed by classifying pixels that have been conventionally classified into class 0 into class 5 by using the fifth offset.

  The QAOU control unit 622 controls each unit included in the adaptive offset filter processing unit 62. Further, the QAOU control unit 622 refers to the QAOU structure information, divides the deblocked decoded image P_DB into one or a plurality of QAOUs, and scans each QAOU in a predetermined order. In addition, the QAOMU number representing the target QAOMU to be processed is supplied to the offset type deriving unit 623.

  The offset type deriving unit 623 refers to the map memory and the list memory of the offset information storage unit 621 and derives an offset type specified by the QAOMU number supplied from the QAOU control unit 622. Further, the derived offset type is supplied to the class classification unit 624.

(Class classification unit 624)
The class classification unit 624 classifies each pixel included in the target QAOU into any of a plurality of classes that can be selected in the offset type supplied from the offset type deriving unit 623. In addition, the offset type and a class index indicating a class into which each pixel is classified are supplied to the offset deriving unit 626.

(Configuration of class classification unit 624)
FIG. 8 is a block diagram illustrating a configuration of the class classification unit 624.

  As shown in FIG. 8, the class classification unit 624 includes an adaptive edge offset class classification unit 6241 and an adaptive band offset class classification unit 6244. The adaptive edge offset class classifying unit 6241 includes a used edge offset class selecting unit 6242 and an edge offset class classifying unit 6243. The adaptive band offset class classifying unit 6244 includes a used band offset class / bandwidth selecting unit. 6245 and a band offset class classifying unit 6246.

  The class classification unit 624 classifies each pixel into a class according to the parameter condition and the offset type. If the offset type indicates an edge offset, the adaptive edge offset class classifying unit 6241 classifies the pixel. If the offset type is a band offset, the adaptive band offset class classifying unit 6244 classifies the pixel.

  The adaptive edge offset class classification unit 6241 is a unit that adaptively classifies pixels into classes according to parameter conditions, and includes a use edge offset class selection unit 6242 and an edge offset class classification unit 6243.

  The used edge offset class selection unit 6242 selects the type of class to be used. The used edge offset class selection unit 6242 inputs parameters used for pixel classification to the edge offset class classification unit 6243. Specifically, the edge class conversion table selected according to the offset type and parameter conditions is input. One of the parameter conditions is the type of the component represented by the pixel. When the component is luminance and color difference, a conversion table corresponding to each is selected and input.

  The edge offset class classification unit 6243 classifies the pixels based on the reference pixel corresponding to the given offset type and the edge class conversion table. The used edge offset class selection unit 6242 may input the offset type using another parameter that can specify the position of the reference pixel, instead of inputting the offset type to the edge offset class classification unit 6243. For example, other indexes associated with the edge offsets EO_0 to EO_3 and the relative coordinates of the processing target pixel and the reference pixel can be used as input.

  The adaptive band offset class classifying unit 6244 is means for adaptively classifying pixels into classes according to parameter conditions, and includes a used band offset class / bandwidth selecting unit 6245 and a band offset class classifying unit 6246. Yes.

  The used band offset class / bandwidth selection unit 6245 inputs parameters used for pixel value classification to the band offset class classification unit 6246. Specifically, the bandwidth and the band class conversion table are input. One of the parameter conditions is the type of the component represented by the pixel. When the component is luminance and color difference, a band class conversion table corresponding to each is selected and input.

  The band offset class classification unit 6246 classifies the pixel values into classes according to the input bandwidth and the band class conversion table. The input bandwidth may be an integer corresponding to the bandwidth, not the bandwidth itself. For example, when the bandwidth is 2 ^ BWBit, the logarithm of base 2 is BWBit, or a value for obtaining the number of bits used for the bandwidth from the bit depth of the pixel, for example, BWBBit = PIC_DEPTH-5. 5 may be used instead of the bandwidth.

  Next, the classification process by the class classification unit 624 will be described in more detail with reference to FIGS. 9 and 10.

(For edge offset)
If the offset type supplied from the offset type deriving unit 623 is any one of 1 to 4, edge offset processing is performed. At this time, the class classification unit 624 determines whether or not there is an edge in the vicinity of the pixel to be processed, and if there is an edge, the type of the edge is determined, and according to the determination result, the pixel to be processed Are classified into any of a plurality of classes.

  More specifically, first, the class classification unit 624 derives EdgeClass representing the edge type of the processing target pixel x. EdgeClass is an intermediate value for deriving an offset class index class_idx. EdgeClass is a pixel value pix [x] of the processing target pixel x and pixel values pix [a] and pix [b] of two pixels a and b that are adjacent to the processing target pixel or share a vertex. And derived using Hereinafter, in order to derive EdgeClass, a pixel used together with the processing target pixel is referred to as a reference pixel. Which pixel is used as a reference pixel depends on the offset type. FIGS. 9A to 9D are diagrams for explaining the position of the reference pixel in each edge offset type. In any case, the pixel x indicates the processing target pixel, and the pixel a and the pixel b indicate the reference pixel. .

When the offset type is EO_0 (sao_type_idx = 1) As shown in FIG. 9A, the pixel adjacent to the left side of the processing target pixel x is defined as a pixel a, and the pixel adjacent to the right side of the processing target pixel is defined as a pixel b. .

When the offset type EO_1 (sao_type_idx = 2) As shown in FIG. 9B, the pixel adjacent to the upper side of the processing target pixel x is defined as a pixel a, and the pixel adjacent to the lower side of the processing target pixel is defined as a pixel b. To do.

When the offset type is EO_2 (sao_type_idx = 3) As shown in FIG. 9C, the pixel sharing the upper left vertex of the processing target pixel x is defined as a pixel a, and the pixel sharing the lower right vertex of the processing target pixel Is a pixel b.

When the offset type EO — 3 (sao_type_idx = 4) As shown in FIG. 9D, the pixel sharing the lower left vertex of the processing target pixel x is defined as a pixel a, and the pixel sharing the upper right vertex of the processing target pixel Let it be pixel b.

If the pixel a and the pixel b are determined, the sign of the difference between them and the processing target pixel x Sign (pix [x] −pix [a]), and Sign (pix [x] −pix [b])
The class classification unit 624 derives the edge type EdgeClass. Where Sign (z) is
Sign (z) = + 1 (when z> 0)
Sign (z) = 0 (when z = 0)
Sign (z) =-1 (when z <0)
It is a function that takes each value of. FIG. 10A is a schematic diagram showing the magnitude relationship between the pixel value of the processing target pixel x shown in FIG. 9 (white circle in the drawing) and the pixel value of the pixel a or b (black circle in the drawing) by the height of the circle. , And the value of the function Sign according to the magnitude relationship. Therefore, the relationship between the magnitude relationship between the pixel value pix [x] of the pixel x and the pixel value pix [p] of the pixel p (p is a or b) and the value of the function Sign is
When pix [x] <pix [p], Sign (pix [x] −pix [p]) = − 1,
When pix [x] = pix [p], Sign (pix [x] −pix [p]) = 0,
When pix [x]> pix [p], Sign (pix [x] −pix [p]) = + 1,
It becomes.

  Subsequently, the class classification unit 624 derives EdgeClass by the following expression based on Sign (pix [x] −pix [a]) and Sign (pix [x] −pix [b]).

EdgeClass = Sign (pix [x] −pix [a]) + Sign (pix [x] −pix [b]) + 2
FIG. 10B shows a schematic diagram of the type of magnitude relationship between the pixel value of the processing target pixel x and the pixel values of the reference pixels a and b, and the EdgeClass value corresponding to each magnitude relationship. As shown in FIG.
When both the pixel value pix [a] and the pixel value pix [b] are larger than the pixel value pix [x], EdgeClass = 0.
EdgeClass = 1 when one of the pixel value pix [a] and the pixel value pix [b] is larger than the pixel value pix [x] and the other pixel value is the same.
If one of the pixel value pix [a] and the pixel value pix [b] is smaller than the pixel value pix [x] and the other pixel value is the same, EdgeClass = 3.
When the pixel value pix [a] and the pixel value pix [b] are smaller than the pixel value pix [x], EdgeClass = 4.
The pixel value pix [x] has a pixel value pix [a] and a pixel value pix [b], one of which is small and the other pixel value is large, or the pixel value pix [x]. On the other hand, when both the pixel value pix [a] and the pixel value pix [b] are the same, EdgeClass = 2.

Subsequently, the class classification unit 624 derives the class index (class_idx) of the class to which the processing target pixel x should belong based on the derived EdgeClass as follows.
When the pixel to be processed is a luminance pixel, class_idx = EoTblY [EdgeClass]
When the pixel to be processed is a chrominance pixel, class_idx = EoTblC [EdgeClass]
Here, EoTblY and EoTblC are conversion tables for deriving class_idx from EdgeClass, and are hereinafter referred to as edge class conversion tables. EoTblY is used for luminance pixels, and EoTblC is used for color difference pixels. In this way, the class classification unit 624 uses the edge class conversion table properly according to the component represented by the processing target pixel x, that is, whether it is a luminance pixel or a pigment pixel, and adapts to the characteristics of each of the luminance pixel and the chrominance pixel. Classify pixels into classes.

  Next, FIGS. 10C and 10D show specific examples of EoTblY and EoTblC (class classification method). A conversion table 1001A in FIG. 10C is an edge class conversion table EoTblY used for luminance pixels. Pixels with EdgeClass = 2 are classified as class 0 and are not subjected to offset processing.

  A conversion table 1001B in FIG. 10D is an edge class conversion table EoTblC used for color difference pixels, and is used when offset processing is performed for all five edge types.

  Here, the edge class conversion table is switched using the component represented by the pixel as the parameter condition, but the edge class conversion table may be selected according to other parameter conditions. Other examples of parameter conditions include division depth (sao_depth) and QAOU size. As a specific example, the edge class conversion table 1001A is used for luminance pixels and the edge class conversion table 1001B is used for color difference pixels as described above only when the division depth is equal to or greater than the threshold value. 1001A can be used as the edge class conversion table. As another example, the edge class conversion table 1001A is used for luminance pixels and the edge class conversion table 1001B is used for chrominance pixels only when the QAOU size is less than the threshold value. In other cases, a common edge class is used. 1001A can be used as the conversion table.

(For band offset)
If the offset type supplied from the offset type deriving unit 623 is 5 or 6, band offset processing is performed. At this time, the class classification unit 624 classifies the pixel value of the processing target pixel into one of a plurality of classes according to the pixel value pix [x] of the processing target pixel x. In this embodiment, since the pixel value characteristics of the color difference pixels are used, different classifications are used for the luminance pixels and the color difference pixels. The number max used below represents the maximum value that can be taken by the pixel value of the processing target pixel x. For example, if the pixel is expressed by an 8-bit unsigned integer, the maximum pixel value max = 255. . For convenience, it is assumed that max1 = max + 1. Furthermore, the remainder of division by “/” is rounded down.

  The band offset process will be described with reference to FIGS.

FIG. 11 is a diagram for explaining offset processing by the adaptive offset filter 60. Each,
(A) Processing target pixel x is a luminance pixel and offset type BO_0 (sao_type_idx = 5)
(B) Processing target pixel x is a luminance pixel and offset type BO_1 (sao_type_idx = 6)
(C) Processing target pixel x is a color difference pixel and offset type BO_0 (sao_type_idx = 5)
(D) The processing target pixel x is a color difference pixel and the offset type BO_1 (sao_type_idx = 6)
The class classification when is is shown schematically. In both cases, a value range that is not hatched indicates that the pixel x is classified as class 0, and a value range that is shaded indicates that it is other than class 0. It will be as follows if each of (a)-(d) is demonstrated in detail.
(A) When the processing target pixel x is a luminance pixel and the offset type BO_0 (sao_type_idx = 5), the class classification unit 624 determines that the pixel value pix [x] of the processing target pixel x is
(Max1 × 1/4) ≦ pix [x] <(max1 × 3/4) (formula E1)
Is satisfied, the processing target pixel is classified into a class other than class 0. That is, when the pixel value of the processing target pixel is within the hatched range in FIG. 11A, the processing target pixel is classified into a class other than class 0.
(B) When the processing target pixel x is a luminance pixel and the offset type BO_1 (sao_type_idx = 6), the class classification unit 624 determines that the pixel value pix [x] of the processing target pixel x is
pix [x] <(max1 × 1/4) or (max1 × 3/4) ≦ pix [x]
(Formula E2)
Is satisfied, the processing target pixel is classified into a class other than class 0. That is, when the pixel value of the processing target pixel is within the hatched range in FIG. 11B, the processing target pixel is classified into a class other than class 0.
(C) When the processing target pixel x is a color difference pixel and the offset type BO_0 (sao_type_idx = 5), the class classification unit 624 determines that the pixel value pix [x] of the processing target pixel x is
(Max1 × 4/32) ≦ pix [x] <(max1 × 20/32) (formula E3)
Is satisfied, the processing target pixel is classified into a class other than class 0. That is, when the pixel value of the processing target pixel is within the hatched range in FIG. 11C, the processing target pixel is classified into a class other than class 0.
(D) When the processing target pixel x is the color difference pixel and the offset type BO_1 (sao_type_idx = 6), the class classification unit 624 determines that the pixel value pix [x] of the processing target pixel x is
(Max1 × 12/32) ≦ pix [x] <(max1 × 28/32) (formula E4)
Is satisfied, the processing target pixel is classified into a class other than class 0. That is, when the pixel value of the processing target pixel is within the hatched range in FIG. 11D, the processing target pixel is classified into a class other than class 0.

When max = 255, the above conditions (Equation E1) to (Equation E4) are such that pix [x] is quantized with a quantization width of 8, and (Equation E1) 8 ≦ (pix [x] / 8 ) ≦ 23
(Formula E2) (pix [x] / 8) ≦ 7 or 24 ≦ (pix [x] / 8)
(Formula E3) 4 ≦ (pix [x] / 8) ≦ 19
(Formula E4) 12 ≦ (pix [x] / 8) ≦ 27
Can also be expressed using the quantization width of 16, respectively.
(Formula E1) 4 ≦ (pix [x] / 16) ≦ 11
(Formula E2) (pix [x] / 16) ≦ 3 or 12 ≦ (pix [x] / 16)
(Formula E3) 2 ≦ (pix [x] / 16) ≦ 9
(Formula E4) 6 ≦ (pix [x] / 16) ≦ 13
It can also be expressed as

  The class classification process by the class classification unit 624 will be described in more detail as follows.

When the offset type is either BO_0 or BO_1, the class classification unit 624 indicates the class index (class_idx) of the class to which the processing target pixel x should belong, the pixel component represented by the processing target pixel x, and the range of pixel values ( It is derived as follows according to the band index (band_idx) representing the band) and the offset type (sao_type_idx).
When the pixel to be processed is a luminance pixel, class_idx
= BoTblY [sao_type_idx] [band_idx]
When the pixel to be processed is a color difference pixel, class_idx
= BoTblC [sao_type_idx] [band_idx]
Here, BoTblY and BoTblC are conversion tables used for deriving class_idx from band_idx and sao_type_idx, and are hereinafter referred to as band class conversion tables. BoTblY is used for luminance pixels, and BoTblC is used for color difference pixels. As described above, the class classification unit 624 uses the band class conversion table properly according to the component represented by the processing target pixel x, that is, the luminance pixel or the chrominance pixel, and adapts to the characteristics of the luminance pixel and the chrominance pixel. Classify pixel values into classes. That is, conversion from band_idx to class_idx is performed.

Here, band_idx is the following expression: band_idx = pix [x] >> BWBit
(">>" is a right shift operation)
Is derived by BWBit is a value derived from PIC_DEPTH-BWBit when the image bit depth is set to PIC_DEPTH, and the pixel value is quantized to a value of 2BWBit stage. Shifting to the right by BWBBit corresponds to dividing by (1 << BWBBit) ("<<" is a left shift operation). For example, if BWBit = 3, shifting right by BWBBit corresponds to division by 8 and means that the quantization width (bandwidth) of each pixel value is 8. If this time PIC_DEPTH = 8, the number of bands shown in band_idx becomes ^{2 8/2} 3 = 32 kinds.

  Next, a case where different quantization widths are used for luminance and color difference will be described. FIG. 38 is a diagram illustrating a method of calculating the band_idx of luminance and color difference. As shown in FIG. 38A, the luminance band index band_idx_Y is calculated by the following equation.

BWBitY = BitDepthY-5
band_idx_Y = pix [x] >> BWBitY
Here, BitDepthY is the luminance image bit depth PIC_DEPTH. The luminance shift value BWBitY is calculated by subtracting a predetermined constant NY (here, 5) from the luminance bit depth.

  As shown in FIG. 38B, the color difference band index band_idx_C is calculated by the following equation.

BWBitC = BitDepthC-6
band_idx_C = pix [x] >> BWBBitC
Here, BitDepthC is the color difference image bit depth PIC_DEPTH. The brightness shift value BWBitY is calculated by subtracting a predetermined constant NC (here, 6) from the brightness bit depth.

  According to the above calculation method, the color difference bandwidth is narrower than the luminance bandwidth. When observing a color difference histogram, pixel values are often concentrated in a narrow range of values as compared to a luminance histogram. Therefore, in order to efficiently classify color differences, it is effective to classify pixel values using a narrow bandwidth. At this time, the luminance is quantized to 32 levels and the color difference is quantized to 64 levels. That is, if the luminance is quantized to N1 level and the color difference is quantized to N2 level, the above processing is quantized so that N2> N1.

  As described above, in the calculation of the color difference shift value BWBitC and the luminance shift value BWBitY, the predetermined constant NC used for calculating the color difference shift value is more than the predetermined constant NY used for calculating the luminance shift value. By using a large value, a bandwidth having a narrower color difference than luminance is used. Thereby, the color difference is decomposed into more stages than the luminance, and the color difference concentrated in a narrow band can be suitably classified. That is, it can be decomposed into a relatively large number of bands. As a result, the encoding efficiency can be improved as compared with the case where the luminance quantization and the color difference quantization are performed by the same method.

  Next, specific examples of the band class conversion tables BoTblY and BoTblC (class classification method) are shown in FIG. In each of FIGS. 12A and 12B, “BO_0” indicates sao_type_idx = 5, and “BO_1” indicates sao_type_idx = 6. The band class conversion table 1301 in FIG. 12A is a conversion table used when the processing target pixel x is a luminance pixel, and the band class conversion table 1302 in FIG. It is a conversion table used when it is a pixel.

When the processing target is a luminance pixel and sao_type_idx = 5 (BO_0), as shown in the middle part of FIG. 12A, the class classification unit 624 has a band index band_idx of 8 ≦ band_idx ≦ 23.
The processing target pixel x belonging to the band_idx that satisfies the above condition is classified into one of the class indexes 1 to 16 according to the value of the band_idx.

When the processing target is a luminance pixel and sao_type_idx = 6 (BO_1), as shown in the lower part of FIG. 12A, the class classification unit 624 has a band index band_idx of band_idx ≦ 7 or 24 ≦ band_idx.
The processing target pixel x belonging to the band_idx that satisfies the above condition is classified into one of the class indexes 1 to 16 according to the value of the band_idx.

When the processing target is a color difference pixel and sao_type_idx = 5 (BO_0), as shown in the middle part of FIG. 12B, the class classification unit 624 has a band index band_idx of 4 ≦ band_idx ≦ 19.
The processing target pixel x belonging to the band_idx that satisfies the above condition is classified into one of the class indexes 1 to 16 according to the value of the band_idx.

  Further, when the processing target is a color difference pixel and sao_type_idx = 6 (BO_1), as shown in the lower part of FIG. 12B, the class classification unit 624 performs processing that the band index band_idx belongs to band_idx that satisfies 12 ≦ band_idx ≦ 27. The target pixel x is classified into one of class indexes 1 to 16 according to the value of band_idx.

  FIG. 12C and FIG. 12D are other examples of the band class conversion table BoTblC.

  In the band class conversion table 1302 shown in FIG. 12B and the band class conversion table 1303 shown in FIG. 12C, bands for which offset processing is performed in both BO_0 and BO_1 (bands overlapping in two types) are provided. Although existing, the band class conversion table 1302 and the band class conversion table 1303 have different numbers of overlapping bands.

  Further, in the band class conversion table 1302 shown in FIG. 12B and the band class conversion table 1303 shown in FIG. 12C, some bands near the pixel value 0 and the pixel value max perform the offset process. However, in the band class conversion table 1304 shown in FIG. 12D, all bands are subjected to offset processing with either BO_0 or BO_1.

  The band class conversion table 1304 shown in FIG. 12D is also an example in which a plurality of bands are classified into one class, and two bands are classified into one class for the intermediate gradation portion of each color of color difference. Here, an example in which two bands are classified into one class for the intermediate gradation portion of each color is shown, but a band near the pixel value 0 or the pixel value max may be classified into one class. Three or more bands may be classified into one class.

  In the color difference image, the appearance frequency of the band including the vicinity of the center of the pixel value, that is, the achromatic color and the pixel value close thereto is high. Therefore, offset correction of the band is effective for improving subjective image quality. Therefore, as shown in the band class conversion tables 1302, 1303, and 1304 shown in FIGS. 12B, 12C, and 12D, the band near the achromatic color is offset in any of a plurality of types. Regardless of which type is selected, the band can obtain the effect of offset processing. Thereby, encoding efficiency and subjective image quality can be improved. Note that this is not limited to the achromatic value range in the color difference image, and if the pixel value range that is important for subjective image quality or encoding efficiency can be determined in the same manner as in the achromatic color and the vicinity thereof, as in the case of the achromatic color described above. A configuration may be employed in which the value range is offset by a plurality of types.

  Here, the band class conversion table is switched using the component represented by the pixel as the parameter condition, but the band class conversion table may be selected according to other parameter conditions. Other examples of parameter conditions include division depth (sao_depth) and QAOU size. For example, only when the division depth is equal to or greater than the threshold value, different band class conversion tables can be used for the luminance pixels and the color difference pixels, and in other cases, a common band class conversion table can be used. In another example, only when the QAOU size is less than the threshold, different band class conversion tables can be used for luminance pixels and chrominance pixels, and in other cases, a common band class conversion table can be used.

  The offset attribute setting unit 625 determines a shift value (OFFSET_SHIFT) of the offset and outputs it to the offset deriving unit. The encoded data offset is encoded with an offset bit depth (also referred to as OFFSET_DEPTH) that is less accurate than the pixel bit depth (also referred to as PIC_DEPTH). That is, the offset in the encoded data is quantized, and the shift value indicates a bit shift amount necessary for performing inverse quantization. Also, the offset attribute setting unit 625 determines the offset bit depth and the offset value range. Here, the bit depth of the offset is determined from the pixel bit depth (not shown) input to the offset attribute setting unit 625. The pixel bit depth indicates a range of pixel values constituting the input image of the adaptive offset filter 60 by a bit width. When the pixel bit depth is N bits, the pixel value ranges from 0 to (2 ^ N) −1. Take a range of.

  The following equations are used to derive the SAO offset bit depth and shift value. Other values may be used based on the parameter conditions.

OFFSET_DEPTH = MIN (PIC_DEPTH, 10)
OFFSET_SHIFT = PIC_DEPTH-OFFSET_DEPTH
According to the above formula, for example, when PIC_DEPTH is 8 bits, OFFSET_DEPTH is 4 bits, and when PIC_DEPTH is 10 bits, OFFSET_DEPTH is 6 bits.

  The offset deriving unit 626 refers to the list memory of the offset information storage unit 621 and derives an offset specified by the offset type and class index supplied from the class classification unit 624 for each pixel included in the target QAOU. Furthermore, the offset deriving unit 626 includes an offset inverse quantization unit (not shown). The offset inverse quantization unit performs left bit shift, that is, inverse quantization, on the offset by the shift value set by the offset attribute setting unit 625 so that the bit depth of the offset matches the pixel bit depth. By performing such inverse quantization, it is possible to add the pixel value and the offset at the same bit depth in the addition processing of the offset addition unit 627 described later. The offset obtained by inverse quantization for each pixel is supplied to the offset adding unit 627.

  FIG. 25 is a block diagram illustrating a configuration of the offset addition unit 627. The offset addition unit 627 includes an addition unit 6271 and a clip unit 6772. The adding unit 6271 adds the offset supplied from the offset deriving unit 626 to each pixel of the deblocked decoded image P_DB in the target QAOU. The clip unit 6272 performs clip processing so that the result output from the adder 6271 falls within a value that can be expressed by the pixel bit depth. Note that the clipping process is a process of converting an arbitrary value into a value included in a predetermined value range. Here, the necessity of clip processing in offset addition will be described with reference to FIG. In FIG. 27, (a) represents a case where the offset value range is -Q0 to + Q1. The width of the offset range (also referred to as OFFSET_RANGE (range of offset value)) is determined based on OFFSET_DEPTH, and the value is expressed by the following equation. Note that D is a term representing a difference value for obtaining the offset value range from the offset accuracy, and D = 4 is used in the present embodiment.

OFFSET_RANGE = OFFSET_DEPTH-D
According to the above formula, for example, when OFFSET_DEPTH is 8 bits, OFFSET_RANGE is 4 bits. If the positive and negative ranges are symmetrical, the offset value range is −8 to 7. That is, Q0 = 8 and Q1 = 7.

FIG. 27B is a number line indicating the range of the pixel value from 0 to max, and shows that the value after the offset addition may be below the lower limit 0 of the pixel value. FIG. 27C is a number line that similarly shows the range of pixel values from 0 to max, and shows that the value after the offset addition may exceed the upper limit max of the pixel value. As shown in FIGS. 27B and 27C, since the pixel value may exceed a predetermined range due to the addition of the offset, the clipping process is required. For example, when the pixel bit depth is N bits, the pixel value range is 0 to (2 ^ N) -1, and it is necessary to perform clipping processing so that the value after offset addition falls within this range. . In FIG. 25, the addition unit and the clip unit are shown as separate configurations, but instead of these, a saturation calculation unit (not shown) using a saturation calculation in which calculation and clip processing are integrated is used. It is good also as a structure.
The offset addition unit 627 outputs an image obtained by processing all the QAOUs included in the deblocked decoded image P_DB as an offset filtered decoded image P_OF.

  Next, the flow of processing in the adaptive offset filter processing unit 62 will be described with reference to FIG. FIG. 13 is a flowchart showing the flow of processing by the adaptive offset filter processing unit 62.

(Step S101)
First, the QAOU control unit 622 acquires QAOU structure information from the offset information storage unit 621.

(Step S102)
Subsequently, the QAOU control unit 622 starts a first loop using the QAOMU number of the target QAOMU to be processed as a loop variable.

(Step S103)
The QAOU control unit 622 supplies the QAOMU number to the offset type deriving unit 623. The offset type deriving unit 623 reads the offset type specified by the QAOMU number supplied from the QAOU control unit 622 from the map memory and list memory of the offset information storage unit 621. Further, the offset type deriving unit 623 supplies the read offset type to the class classification unit 624.

(Step S104)
When the offset type read by the offset type deriving unit 623 is “0” (“sao_type_idx” = 0) (YES in step S104), the QAOU control unit 622 determines that all QAOMUs to be processed The offset adding unit 627 is controlled so as not to add an offset to the pixel, and the processing is continued from step S111.

(Step S105)
When the offset type read by the offset type deriving unit 623 is not “0” (“sao_type_idx” = 0) (NO in step S104), the QAOU control unit 622 continues with the pixels of each pixel included in the target QAOMU. Start a second loop with the number as the loop variable. Here, the pixel number is for identifying pixels included in the target QAOMU from each other. For example, the pixel number is assigned to each pixel included in the target QAOMU in a predetermined scan order. Further, instead of such a pixel number, the x coordinate and y coordinate of each pixel included in the target QAOMU may be used as a loop variable.

(Step S106)
Subsequently, the class classification unit 624 classifies the pixel to be processed into any of a plurality of classes selectable in the offset type supplied from the offset type deriving unit 623. Further, the offset type and the class index indicating the class into which the pixel to be processed is classified are supplied to the offset deriving unit 626.

(Step S107)
When the classification result of the target pixel is class 0 (“class_idx” = 0) (YES in step S107), the QAOU control unit 622 causes the offset addition unit 627 not to add an offset to the target pixel. Control is continued from step S110.

(Step S108)
When the target pixel classification result is not class 0 (“class_idx” = 0) (NO in step S107), the offset deriving unit 626 subsequently stores the offset to be added to the pixel to be processed as offset information. Read from the unit 621. That is, the offset specified by the offset type and class index supplied from the class classification unit 624 is read.

(Step S109)
Subsequently, the offset adding unit 627 adds the offset supplied from the offset deriving unit 626 to the pixel value of the processing target pixel of the deblocked decoded image P_DB.

(Step S110)
This step is the end of the second loop.

(Step S111)
This step is the end of the first loop.

  The above is the flow of processing in the adaptive offset filter processing unit 62.

(Other types of band offset configuration)
Here, another type configuration example of the band offset will be described with reference to FIGS.

  31 (a) to 31 (c), 32 (a), and 32 (b) are all band class conversion tables for configuring two band offset types each including eight classes excluding 0. This is a specific example. Each example is the same as the example shown in FIG. 12 in that the range of pixel values is divided into 32 bands and associated with several classes, but differs in the method of associating bands and classes.

  In the band class conversion table 3101 shown in FIG. 31A, in the type BO_0, each band is associated with eight valid classes. Type BO_1 associates a band near the center with 8 valid classes and 1 invalid class. Since QAOU with a small division depth has a large area, the distribution of pixel values tends to be wide. Even in this case, since many bands (in this example, all bands) are associated with any class in type BO_0, even if the distribution of pixel values is wide, offset processing is applied to many pixels. Has the advantage of being possible. On the other hand, the type BO_1 is highly effective when there are many pixel values in the vicinity of the center including the pixel value of the color difference.

  A band class conversion table 3102 shown in FIG. 31B is an example in which a pixel value closer to the center is more important than the band class conversion table 3101 shown in FIG. Type BO_0 is the same as (a), but type BO_1 is half the width of (a), and one band is associated with one class. Thereby, when the appearance frequency of the pixel value near the center is high, more accurate offset processing can be applied. Therefore, it is particularly effective for color differences.

  A band class conversion table 3103 shown in FIG. 31C is an example in which the offset processing is applied more intensively to the pixel value near the center. By using BO_0 and BO_1 with different corresponding bandwidths, even if pixel values are concentrated near the center, the type can be switched according to the degree of concentration of the pixel values, and offset processing can be applied appropriately .

  A band class conversion table 3201 shown in FIG. 32A is an example in which the same class is associated with bands of pixel values that are not adjacent to each other. In particular, in a color difference image, the distribution of pixel values tends to be concentrated, and this is more conspicuous as QAOU becomes smaller. At this time, the frequency with which greatly different pixel values simultaneously appear in the QAOU is low. Therefore, as shown in the band class conversion table 3201, even when a plurality of bands representing pixel values that are greatly different are associated with the same class, in many cases, only one of the bands is subjected to offset processing. Thus, by associating a plurality of bands representing greatly different pixel values with the same class, it is possible to adapt to a plurality of pixel value distributions despite a single offset type.

  The band class conversion table 3202 shown in FIG. 32B is an example in which a plurality of bands are associated with the same class as in the example shown in the band class conversion table 3201. However, when the accuracy with respect to the pixel value near the center is increased. It is an example.

  33 (a), (b), FIG. 34 (a), (b), FIG. 35 (a), (b), FIG. 36, and FIG. 37 all exclude the invalid class (class 0). It is a specific example of a band class conversion table for configuring four band offset types including eight effective classes. Each example is the same as the example shown in FIG. 12 in that the pixel value range is divided into 32 bands and associated with several classes, but the method for associating the band with the class is different. There are four types of offset types.

  In the band class conversion table 3301 shown in FIG. 33A, the four types BO_0 to BO_3 are sequentially arranged in the center wide band, the low pixel value range and the high pixel value range, the first halftone narrow band, and the second intermediate level. This is an example of assigning to a narrow band of keys. In the following, “broadband” means that the effective class included in each type includes more bands (for example, more than half of the bands and 16 or more bands in FIGS. 33 to 36). Refers to including fewer bands (eg, less than half of the bands and less than 16 bands in FIGS. 33-36). The “low pixel value range and high pixel value range” mean a pixel value of a dark part or a bright part in a luminance pixel, and means a pixel value having high saturation in a color difference pixel. According to the type configuration of the band class conversion table 3301 shown in (a), the type BO_0 corresponds to a wide range near the center, the type BO_1 corresponds to a dark part and a bright part or a high saturation area, and the type BO_2 uses a halftone low pixel. The value side and BO_3 correspond to the halftone high pixel value side, respectively. BO_2 and BO_3 are types adapted to the case where pixel values of color differences are concentrated on one color such as green or red.

  The band class conversion table 3302 shown in FIG. 33 (b) is similar to the band class conversion table 3301 shown in FIG. 33 (a), and the four types BO_0 to BO_3 are sequentially arranged in the center, wide band, low pixel value range, and high pixel value range. This is an example of assignment to the first halftone narrow band and the second halftone narrow band. The difference from the band class conversion table 3301 is only the type BO_0. According to the type configuration of the band class conversion table 3302 shown in (b), a pixel value having a wider bandwidth than the band class conversion table 3301 shown in (a) while maintaining offset correction performance near the center using type BO_0. It is possible to correspond to the distribution.

  In the band class conversion table 3401 shown in FIG. 34A, four types BO_0 to BO_3 are sequentially arranged in a wide band, a low pixel value range and a high pixel value range, a first halftone narrow band, and a second halftone narrow. It is an example assigned to a band. The difference between the band class conversion table 3301 shown in FIG. 33A and the band class conversion table 3302 shown in FIG. 33B is only type BO_0. According to the type configuration of the band class conversion table 3401 shown in FIG. 34 (a), it is possible to cope with a case where pixel values appear over the entire range due to a large QAOU or the like due to type BO_0. Further, in type BO_0, classes 2, 3, 28, and 29 are invalid classes. Depending on the standard of the video data, even if the pixel value range is 0 to 255 with 8 bits, the usable pixel value range may be defined as 16 to 235 for luminance and 16 to 240 for color difference. In such a case, due to a prediction error or the like, the correction effect by the offset addition is high for a pixel value that is within the range but is outside the range. In the type BO_0 of the band class conversion table 3401, if the classes 2, 3, 28, and 29 near the both ends in the range are invalid classes, the pixel value is prevented from being out of the range by offset addition, and the class is out of the range. By setting 0, 1, 30, and 31 as effective classes, there is an effect of being corrected to a value within the range by offset addition.

  The band class conversion table 3402 shown in FIG. 34 (b) is similar to the band class conversion table 3401 shown in FIG. 34 (a). The four types BO_0 to BO_3 are sequentially arranged in the entire band, the wide band at the center, and the low pixel. It is an example assigned to a value range and a high pixel value range. However, according to the type configuration of the band class conversion table 3402, the bands corresponding to the types BO_2 and BO_3 are wide, which is suitable for an image in which pixel values are more widely distributed.

  The band class conversion table 3501 shown in FIG. 35A is an example in which four types BO_0 to BO_3 are sequentially assigned to the entire band, the central narrow band, the low pixel value range, and the high pixel value range. It differs from the band class conversion table 3402 shown in FIG. 34 (b) only in the assignment of type BO_1. According to the type configuration of the band class conversion table 3501, the center portion can be corrected with higher accuracy by the type BO_1, which is suitable for an image in which pixel values are concentrated in the center.

  A band class conversion table 3502 shown in FIG. 35B is an example in which four types BO_0 to BO_3 are sequentially assigned to the wideband, low pixel value range, high pixel value range, low pixel value range, and high pixel value range in the center. . By adapting the type equally to each band, adaptation is easy regardless of which band the pixel value distribution is concentrated on.

  The band class conversion table 3601 shown in FIG. 36 is an example in which four types BO_0 to BO_3 are sequentially assigned to the entire band, the low pixel value range and the high pixel value range, the central broadband, and the central narrow band. It is possible to provide two types of types that are suitable for the central portion and to intensively correct the central portion.

  The offset types shown in the above specific examples can be used in addition to or in place of other offset types in the present embodiment. In that case, possible values of sao_type_idx for specifying the offset type are set according to the total number of offsets. The range of the index class_idx for designating a class is 0 to 7 because the number of classes is eight.

  The above specific examples may be selectively used only for luminance or color difference, or may be used in common for luminance and color difference. In particular, as described with reference to FIGS. 33 to 36, the type configuration using more than two offset types has both types suitable for the characteristics of luminance and color difference pixel values. The most effective offset type can be used without switching the type configuration by luminance and color difference.

  Further, in each of the above specific examples, a plurality of types may correspond to the offset processing of the same band. For example, in the example of the band class conversion table 3601 shown in FIG. 36, the center eight bands are offset by any of the types BO_0, BO_2, and BO_3. Further, such overlap is not limited to the center band, and in the example of the band class conversion table 3302 shown in FIG. 33B, the bands 4, 5, 6, 7 and the bands 24, 25, 26, 27 are typed. The offset processing is symmetrical between BO_0 and type BO_1. In this way, by making it possible to perform offset processing on a plurality of types of pixel values in a predetermined range, it is possible to increase the chance of performing offset processing on the pixel values in the predetermined range. Examples of the predetermined range include the vicinity of the center of the color difference, the vicinity of the upper limit and the lower limit of the luminance, and the like, but are not limited thereto. Pixel values having a high appearance frequency and pixel values having a high effect of offset correction are included in the predetermined range. As described above, encoding efficiency can be improved by enabling offset processing with a plurality of types. The plurality of types need not be all offset types of the band offset.

  In the band class conversion table 3701 shown in FIG. 37, four types BO_0 to BO_3 are sequentially arranged in a narrow band at the center, a halftone area on the low pixel value range side, a halftone area on the high pixel value range side, a low pixel value range, and a high range. It is an example assigned to the pixel value range. Each type includes a class corresponding to a narrow range, and highly accurate offset processing is possible. In the case of a small QAOU, for example, in the case of adaptive offset processing performed on a local block such as 64 × 64, the distribution of pixel values tends to be concentrated in a part of the range compared to a large QAOU. In the example of the band class conversion table 3701 shown in FIG. 37, bands do not overlap among a plurality of types. However, in adaptive offset filter processing using a relatively small block, a narrow band corresponding to a range of important pixel values Rather than preparing a plurality of types with a wide band, by preparing a plurality of types with high accuracy for a narrow band, it is possible to adapt even when pixel values are concentrated outside the center.

  The band table conversion table 3901 shown in FIG. 39A is a plurality of band tables and does not overlap. That is, for the pixel value X, there is no pixel value X in which X is classified into a plurality of classes. Each class classification is continuous. That is, in the quantized pixel values X1, X2, and X3, when the pixel value X1 and the pixel value X3 belong to the same class, the pixel value X2 always belongs to the same class as the pixel values X1 and X3. When using continuous and non-overlapping band conversion tables, the encoder can perform simple processing by selecting a class classification that includes representative values (for example, average value, median value, and mode value) of pixel values in a certain area. It is possible to select a suitable classification.

  The band conversion table 3902 shown in FIG. 39B is a diagram illustrating an example in which values in a wide range are classified into the same class in a small value range and a large value range. Thereby, the range which is not classified can be reduced.

  The band conversion tables 4001 and 4002 shown in FIG. 40 are tables used when the color difference is quantized into 64 levels as described with reference to FIG. It is the figure which divided and showed one table for classifying the band indexes from 0 to 63 into two, 4001 shows the first half (0-31), and 4002 shows the second half (32-63).

  FIG. 41 is a diagram illustrating an example in which a range with a small value and a range with a large value are classified into classes (invalid classes) to which no offset is added. As a result, it is possible to perform fine classification on pixel values near the center value where the frequency of occurrence of color differences is high, and high coding efficiency can be obtained.

  The band conversion tables 4101 and 4102 shown in FIG. 41 are examples in which values in a wide range are classified into the same class in a small value range and a large value range.

  The band conversion tables 4201 and 4202 illustrated in FIG. 42 are diagrams illustrating examples in which the classification of the center value and the pixel values in the vicinity thereof are classified more finely than the classification of the pixel values far from the center in the color difference class classification. According to the observation of the color difference histogram, the frequency of occurrence of pixel values tends to be higher at pixel values near the center representing low saturation. Therefore, by using the table shown in FIG. 42, it is possible to assign to a class (effective class) in which a relatively wide range of pixel values are offset added while finely classifying pixel values in the vicinity of the center value where the frequency of occurrence of color differences is high.

  The band conversion tables 4301 and 4302 shown in FIG. 43 are also examples in which the classification of the central value and the neighboring pixel values are classified more finely than the classification of the pixel values far from the center in the color difference class classification.

  In the band conversion tables 4401 and 4402 shown in FIG. 44, in the color difference class classification, the class classification that classifies the center value and the pixel values in the vicinity thereof into an effective class is set as the first (first) class classification of the plurality of class classifications. It is an example to use. That is, a smaller offset type value is assigned. As described above, in the color difference, the center value and the pixel values in the vicinity thereof have a high appearance frequency, and accordingly, the application frequency of the class classification for classifying the pixel value into the effective class tends to increase. By assigning a small offset type to a class classification having a high application frequency, it is possible to reduce the amount of code required for encoding the offset type.

  The band conversion tables 4501 and 4502 shown in FIG. 45 are classified into pixel values X1, pixel values X2, and pixel values X3 (quantized) in the class classification for classifying the pixel values apart from the center value and the neighboring pixel values. In this example, for X1 <X2 <X3), pixel values X1 and X3 that are not adjacent to each other are classified into the same class different from the class that classifies X2. By classifying pixel values with low probability of occurring simultaneously into the same class, it is possible to reduce the quantization width and cover a wide range of values even when using a band conversion table that performs finer classification. . That is, it is possible to narrow the range where the offset is not added and improve the encoding efficiency.

  The band conversion tables 4601 and 4602 shown in FIG. 46 are classified into pixel values X1, pixel values X2, and pixel values X3 (quantized) in the class classification for classifying the pixel values apart from the center value and the neighboring pixel values. In this example, for X1 <X2 <X3), pixel values X1 and X3 that are not adjacent to each other are classified into the same class different from the class that classifies X2. Furthermore, by using the fact that pixel values are close to each other because of the spatial correlation of the image, a wide range of values can be classified using fewer class classifications by using a periodic structure for the class classification pattern. Is possible. Here, it is possible to classify close values into different classes while covering all value ranges using two class classifications. That is, since the range can be efficiently classified with a small class and a small quantization step, high encoding efficiency can be obtained. Further, by using a small number of classes, it is possible to reduce the amount of flag codes necessary for class classification, and to obtain high coding efficiency. Furthermore, by using a small number of classes, it is possible to reduce the amount of processing when selecting an optimal class classification in the encoding apparatus.

  47 to 50 below are examples in which different class classifications are used for luminance and color difference.

  The band conversion tables 4701, 4702, and 4703 shown in FIG. 47 are another example of two class classifications having a periodic structure. Luminance classified into 32 levels uses a conversion table 4701, and color differences classified into 64 levels use conversion tables 4702 and 4703.

  Band conversion tables 4801, 4802, and 4803 shown in FIG. 48 are another example of two class classifications having a periodic structure. Luminance classified into 32 levels uses a conversion table 4801, and color differences classified into 64 levels use conversion tables 4802 and 4803.

  Band conversion tables 4901, 4902, and 4903 shown in FIG. 49 are another example of two class classifications having a periodic structure. Luminance classified into 32 levels uses a conversion table 4901 and color differences classified into 64 levels use conversion tables 4902 and 4903.

  Band conversion tables 5001, 5002, and 5003 shown in FIG. 50 are another example of two class classifications having a periodic structure. Luminance classified into 32 levels uses a conversion table 5001, and color differences classified into 64 levels use conversion tables 5002 and 5003.

(Configuration of Offset Information Decoding Unit 611)
Next, the configuration of the offset information decoding unit 611 will be described with reference to FIG. FIG. 14 shows an offset information decoding unit that changes the offset type to be used according to the type (parameter condition) of the adaptive offset parameter and / or changes the number of class classifications according to the parameter condition of the adaptive offset. FIG. 611 is a block diagram of FIG.

  The offset information decoding unit 611 includes an adaptive offset type decoding unit 6111 and an adaptive offset decoding unit 6114. The adaptive offset type decoding unit 6111 includes a use offset type selection unit 6112 and an offset type decoding unit 6113. The adaptive offset decoding unit 6114 includes a use offset number selection unit 6115 and an offset decoding unit 6116. Has been. The parameter condition is a parameter other than a value calculated from the pixel value, and includes a layer, an offset type, a block size (QAOU size), and a QP in addition to the pixel components (components) already described.

  The adaptive offset type decoding unit 6111 is a unit that adaptively decodes an offset type according to parameter conditions from QAOU information in the encoded data, and includes a use offset type selection unit 6112 and an offset type decoding unit 6113. Yes.

  The used offset type selection unit 6112 is means for selecting a set of offset types to be used according to the parameter condition. The used offset type selection unit 6112 selects the conversion table as described with reference to FIG. 6 according to the parameter condition, and inputs it to the offset type decoding unit 6113. Further, the maximum number of usable offset types is input to the offset type decoding unit 6113. The maximum number of offset types that can be used is the number of types of values that the offset type can take. In this embodiment, the conversion table of FIG. 6 is used. Therefore, the maximum number of offset types is 7. In some of the modifications described later, eight or more offset types may be used. In these examples, even when not particularly illustrated, a conversion table that assigns at least codes that do not overlap each offset type is used.

  The used offset type selection unit 6112 supplies a conversion table corresponding to the parameter condition. In this embodiment, the conversion table 801 is used as already described. However, as a configuration including a plurality of conversion tables, which conversion table is used may be switched according to the parameter condition. For example, by using a pixel component as a parameter condition, the edge offset type used when the component is a color difference may be different from the edge offset type used when the component is luminance. As a specific example, when the component is a color difference, as shown in the conversion table 801C of FIG. 6, an offset type that does not include EO_2 and EO_3 is used. In the case of the conversion table 801C, the maximum number of offset types is 5.

The offset type decoding unit 6113 decodes the offset type from the encoded data according to the input conversion table and the maximum number of offset types. If the maximum number of offset types is N, the range that the code can take is limited to N from 0 to N−1. Therefore, the number of bits required to code the code can be reduced. For example, m-bit fixed length coding can be used when the maximum number is greater than 2 ^m -1 and less than or equal to 2 ^m . Further, Truncated unary encoding with the maximum value and N-1 or Truncated Rice encoding can be used.

  The adaptive offset decoding unit 6114 is means for adaptively decoding an offset from QAOU information in the encoded data according to the parameter condition, and includes a used offset number selection unit 6115 and an offset decoding unit 6116.

  The used offset number selection unit 6115 is means for selecting the maximum value of the number of offsets to be used and the offset accuracy in accordance with the parameter conditions. One of the parameter conditions is a component represented by a pixel, and the number of edge offsets used when the component is luminance is different from the number of edge offsets used when the component is a color difference. For example, when the component is luminance, the offset number can be four, and when the component is color difference, the offset number can be five. For the edge class conversion table already described, if the edge class conversion table 1001B of FIG. 10D is used for luminance and the edge class conversion table 1001A of FIG. 10C is used for color difference, adaptive offset filter processing is performed. The number of classes excluding class 0, which is not broken, is 4 in FIG. 10 (d) and 5 in FIG. 10 (c). This indicates the required number of offsets.

  The offset decoding unit 6116 decodes the offset from the encoded data according to the number of offsets used for each offset type.

  As described above, the moving image decoding apparatus 1 adaptively decodes the encoded data according to the parameter condition and classifies the pixels.

  As described above, in this embodiment, the class classification is changed according to the parameter condition. More specifically, according to the component represented by the pixel to be processed, in the case of edge offset, the luminance pixel and the chrominance pixel are classified so that the class numbers are different, and in the case of band offset, the luminance pixel and the chrominance pixel are classified. And so that the correspondence between bands and classes is different. Thereby, it can adapt to the difference of the characteristic of a brightness | luminance and a color difference, and can improve encoding efficiency.

(Reduction of clip processing)
Hereinafter, a processing method of the adaptive offset filter processing unit 62 for the purpose of reducing clip processing in the offset addition unit 627 in the case of band offset processing will be described.

(Clip Processing Reduction Method 1) Clip Processing Reduction by Determination of Pixel Value Here, instead of step S109 in the flowchart showing the flow of processing by the adaptive offset filter processing unit 62 in FIG. As shown). Other steps are the same as those already described.
(Step S109A)
Subsequent to step S108, the offset adding unit 627 adds the offset supplied from the offset deriving unit 626 to the pixel value of the processing target pixel of the deblocked decoded image P_DB. However, when the pixel value is included in a value range represented by a predetermined “offset restriction region”, no offset is added.

  Here, the offset restriction region will be described with reference to FIG. FIG. 28 is a diagram illustrating an example of the offset restriction region. The offset limited region OLR0 is a range of pixel values from 0 to less than Lim0, and the offset limited region OLR1 is a range of pixel values from Lim1 to max. The values of Lim0 and Lim1 are determined so that Q0 ≦ Lim0 and Lim1 ≦ max−Q1 when the offset range is −Q0 to + Q1. As for the pixel values included in the offset limiting area OLR0 and the offset limiting area OLR1, the result of the offset process may exceed the pixel value range 0 to max. Therefore, if the offset is not added in step S109A, the clipping process can be omitted. For pixel values that are not included in the offset limit area OLR0 and the offset limit area OLR1, the value after offset addition does not deviate from the pixel value range 0 to max. Therefore, if the offset restriction area set as described above is used, the clipping process in the clipping unit 6272 can be omitted in the entire pixel value range, and the processing amount of the adaptive offset process can be reduced.

(Clip processing reduction method 2)
Clipping processing reduction by pre-branching in the used offset type selection unit 6112 Further, pixel values belonging to the offset restriction area may be classified into class 0 (invalid class). This will be described with reference to FIG. FIG. 29 is a diagram showing an example of a band class conversion table in consideration of the offset restriction area. The band class conversion table 2901 shown in (a) is different from the conventional BO_1 in the case of BO_1, and the bands with band indexes 0, 1, 30, and 31 are class 0. This can be expressed in terms of 8-bit pixel values when Lim0 = 16 and Lim1 = 240. Actually, the offset range in the case of the 8-bit pixel value is −8 to 7, that is, Q0 = 8 and Q1 = 7, so that Q0 ≦ Lim0 and Lim1 ≦ max−Q1 are satisfied. If this conversion table is used in the used offset type selection unit 6112, the pixel value that needs to be clipped becomes class 0 and is not subject to offset addition. Therefore, it is not necessary to use step 108A, and the original step S108 is changed. While using, clip processing in band offset addition processing can be omitted.

  The band class conversion table 2902 shown in FIG. 29 (b) has a conventional BO_1 with an offset type for classifying pixel values corresponding to the offset restriction region similar to the band class conversion table 2901 shown in (a) as class 0 as BO_2. The example of the band class conversion table in the case of using together with a type is shown. When this conversion table is used, there are three types of band offsets that can be used in the encoded data, and the total number of offset types is eight.

(Clip processing reduction method 3)
Reduction of Clip Processing by Limiting Range of Syntax Elements in Bitstream Furthermore, another method for reducing clip processing will be described. For example, in the encoded data # 1, if the offset range in the class corresponding to the offset limited area OLR0 shown in FIG. 28 is limited to 0 or a positive value, the pixel value after offset addition is clearly less than 0. Don't be. Similarly, if the range of the offset in the class corresponding to the offset restriction region OLR1 is restricted to 0 or a negative value, the pixel value after the addition of the offset is clearly not greater than max. In this way, if the offset range is restricted in consideration of the offset restriction area for a specific class of the band conversion table, clip processing after offset addition can be reduced, and the processing speed can be increased. is there. When this method is adopted, it is not necessary to change the band class conversion table as shown in FIG.

  Note that Lim0 and Lim1 representing the offset limited region may be fixed values according to the pixel bit depth, or may be set according to the band position when the offset type is BO. Alternatively, the encoded data # 1 may be encoded and transmitted. FIG. 26 shows an example of a data structure when a QAOU including offset limited area information RI is encoded and transmitted. When the offset type information OTI indicates a band offset, the offset limited area information RI is encoded. The offset restriction area information RI only needs to be able to represent the offset restriction area or the arrangement of classes corresponding to the offset restriction area. For example, the pixel values Lim0 and Lim1 serving as threshold values, or as another example, a threshold value LimC0 indicating a class index And LimCl. In the latter case, a class having a class index of LimC0 or lower or LimC1 or higher is handled as a class corresponding to the offset restriction area. Specifically, the classes corresponding to the offset restriction area OLR0 are class indexes 0 and 1, and the classes corresponding to the offset restriction area OLR1 are class indexes 30 and 31. Therefore, it is preferable to encode the offset limited region information RI with LimC0 = 1 and LimC1 = 30. The range of values is that when PIC_DEPTH is 8 bits and the bandwidth is 2 ^ BWBit, LimC0 is 0 to (2 ^ (8-BWBit) -2), and LimC1 is LimC0 + 1 to (2 ^ (8-BWBit) -1).

(Clip processing reduction method 4)
In the above, an example in which one set of offset restriction areas is set for each of the lower limit and upper limit of the range has been described. However, the setting of the offset restriction areas is not limited to this, and two or more sets of offset restriction areas as shown in FIG. An area may be set.

  FIG. 30 shows an example in which two offset restriction regions are set on the lower limit side and two upper limit sides of the range. An offset restriction area OLR00 and an offset restriction area OLR01 are arranged on the lower limit side, and an offset restriction area OLR10 and an offset restriction area OLR11 are arranged on the upper limit side. The range of each offset restriction area is that the offset restriction area OLR00 has a pixel value of 0 or more and less than Lim00, the offset restriction area OLR01 has a pixel value of Lim00 or more and less than Lim01, and the offset restriction area OLR11 has a pixel value of Lim11 or more and less than Lim10. The offset limiting region OLR10 is a pixel value Lim10 or more and max or less. Such a configuration of the offset restriction region is preferably used when the range of usable pixel values is set narrower than 0 to max. For example, when a pixel value range is defined as 16 to 235 in luminance and 16 to 240 in color difference when the pixel bit depth is 8 bits, in the offset limited area OLR00 and the offset limited area OLR10, after offset addition It is preferable to limit the offset range so that the value of the pixel value does not exceed the range of the pixel value and falls within or approaches the range. In the offset limit region OLR01 and the offset limit region OLR11, the value after the offset addition is within the range. In order not to exceed, it is preferable not to add an offset or limit the range of the offset. As a result, even if a range of usable pixel values is defined in the range of pixel values, the effect of correction using the offset value can be enhanced and the encoding efficiency can be improved.

  In any of the above methods, the clip processing after the addition is performed by setting an offset restriction region for restricting the offset value range so that the pixel value after the offset addition does not exceed the range that can be expressed by the pixel bit depth. It can be omitted and the processing can be speeded up.

(Moving picture encoding device 2)
Next, the moving image encoding apparatus 2 that generates encoded data # 1 by encoding the encoding target image will be described with reference to FIGS. The moving image encoding apparatus 2 includes H.264 as a part thereof. H.264 / MPEG-4. A method adopted in AVC, a method adopted in KTA software, which is a codec for joint development in VCEG (Video Coding Expert Group), and a method adopted in TMuC (Test Model under Consideration) software, which is the successor codec And the technology employed in HM (HEVC TestModel) software.

  FIG. 15 is a block diagram showing a configuration of the video encoding device 2 according to the present embodiment. As illustrated in FIG. 15, the moving image encoding device 2 includes a transform / quantization unit 21, a variable-length code encoding unit 22, an inverse quantization / inverse transform unit 23, a buffer memory 24, an intra-predicted image generation unit 25, The inter prediction image generation unit 26, motion vector detection unit 27, prediction method control unit 28, motion vector redundancy deletion unit 29, adder 31, subtractor 32, deblocking filter 33, adaptive filter 70, and adaptive offset filter 80 are provided. It is the composition which includes. The moving image encoding device 2 is a device that generates encoded data # 1 by encoding moving image # 10 (encoding target image).

  The transform / quantization unit 21 performs (1) DCT transform (Discrete Cosine Transform) or DST transform (Discrete Sine Transform) for each block on the prediction residual D obtained by subtracting the predicted image Pred from the encoding target image, and (2) The DCT coefficient obtained by the DCT transform or the DST coefficient obtained by the DST transform is quantized, and (3) the quantized prediction residual QD obtained by the quantization is converted into the variable length code encoder 22 and the inverse quantization / inverse This is supplied to the conversion unit 23. The transform / quantization unit 21 selects (1) a quantization step QP used for quantization for each tree block, and (2) a quantization parameter difference indicating the size of the selected quantization step QP. Δqp is supplied to the variable length code encoding unit 22, and (3) the selected quantization step QP is supplied to the inverse quantization / inverse conversion unit 23. Here, the quantization parameter difference Δqp refers to the DCT (or DST) conversion / quantization immediately before the DCT (or DST) conversion / quantization parameter qp (QP = 2pq / 6) related to the tree block to be quantized. This is a difference value obtained by subtracting the value of the quantization parameter qp ′ relating to the tree block.

  The variable-length code encoding unit 22 includes (1) a quantized prediction residual QD and Δqp supplied from the transform / quantization unit 21, (2) a prediction parameter PP supplied from a prediction scheme control unit 28 described later, and (3) The encoded data # 1 is generated by variable-length encoding the filter parameters FP (filter set number, filter coefficient group, region designation information, and on / off information) supplied from the adaptive filter 70 described later. . Further, the variable length code encoding unit 22 encodes the QAOU information supplied from the adaptive offset filter 80 and includes it in the encoded data # 1.

  The inverse quantization / inverse transform unit 23 (1) inversely quantizes the quantized prediction residual QD, and (2) performs inverse DCT (Discrete Cosine Transform) transform or inverse DCT coefficient or DST coefficient obtained by inverse quantization. DST (Discrete Sine Transform) conversion is performed, and (3) the prediction residual D obtained by inverse DCT transformation or inverse DST transformation is supplied to the adder 31. When the quantization prediction residual QD is inversely quantized, the quantization step QP supplied from the transform / quantization unit 21 is used. Note that the prediction residual D output from the inverse quantization / inverse transform unit 23 is obtained by adding a quantization error to the prediction residual D input to the transform / quantization unit 21. Common names are used for this purpose.

  The intra predicted image generation unit 25 generates a predicted image Pred_Intra related to each partition. Specifically, (1) a prediction mode used for intra prediction is selected for each partition, and (2) a prediction image Pred_Intra is generated from the local decoded image P using the selected prediction mode. The intra predicted image generation unit 25 supplies the generated intra predicted image Pred_Intra to the prediction method control unit 28.

  The intra-predicted image generation unit 25 identifies the prediction index PI for each partition from the prediction mode selected for each partition and the size of each partition, and supplies the prediction index PI to the prediction method control unit 28.

  In addition, the intra predicted image generation unit 25 supplies intra coding mode information IEM, which is information indicating the size of the target partition and the prediction mode assigned to the target partition, to the adaptive filter 70 (not shown).

  The motion vector detection unit 27 detects a motion vector mv regarding each partition. Specifically, (1) the filtered decoded image P_FL ′ to be used as a reference image is selected, and (2) the target partition is searched by searching for the region that best approximates the target partition in the selected filtered decoded image P_FL ′. Detects a motion vector mv. Here, the filtered decoded image P_FL ′ is subjected to deblocking processing by the deblocking filter 33, adaptive offset processing by the adaptive offset filter 80, and adaptive filtering processing by the adaptive filter 70 on the decoded decoded image. The motion vector detection unit 27 can read out the pixel value of each pixel constituting the filtered decoded image P_FL ′ from the buffer memory 24. The motion vector detection unit 27 supplies the detected motion vector mv to the inter predicted image generation unit 26 and the motion vector redundancy deletion unit 29 together with the reference image index RI that specifies the filtered decoded image P_FL ′ used as the reference image. To do. Note that for a partition that performs bi-directional prediction (weighted prediction), two filtered decoded images P_FL1 ′ and P_FL2 ′ are selected as reference images, and each of the two filtered decoded images P_FL1 ′ and P_FL2 ′ is selected. Corresponding motion vectors mv1 and mv2 and reference image indexes RI1 and RI2 are supplied to the inter prediction image generation unit 26 and the motion vector redundancy deletion unit 29.

  The inter prediction image production | generation part 26 produces | generates the motion compensation image mc regarding each inter prediction partition. Specifically, using the motion vector mv supplied from the motion vector detection unit 27, the motion compensated image mc is obtained from the filtered decoded image P_FL ′ designated by the reference image index RI supplied from the motion vector detection unit 27. Generate. Similar to the motion vector detection unit 27, the inter predicted image generation unit 26 can read out the pixel value of each pixel constituting the filtered decoded image P_FL ′ from the buffer memory 24. The inter prediction image generation unit 26 supplies the generated motion compensated image mc (inter prediction image Pred_Inter) together with the reference image index RI supplied from the motion vector detection unit 27 to the prediction method control unit 28. For the partition with bi-directional prediction (weighted prediction), (1) the motion compensated image mc1 is generated from the filtered decoded image P_FL1 ′ specified by the reference image index RI1 using the motion vector mv1, and (2 ) A motion compensated image mc2 is generated from the filtered reference image P_FL2 ′ designated by the reference image index RI2 using the motion vector mv2, and (3) an offset value is added to the weighted average of the motion compensated image mc1 and the motion compensated image mc2. Is added to generate the inter predicted image Pred_Inter.

  The prediction scheme control unit 28 compares the intra predicted image Pred_Intra and the inter predicted image Pred_Inter with the encoding target image, and selects whether to perform intra prediction or inter prediction. When the intra prediction is selected, the prediction scheme control unit 28 supplies the intra prediction image Pred_Intra as the prediction image Pred to the adder 31 and the subtractor 32 and also predicts the prediction index PI supplied from the intra prediction image generation unit 25. The parameter PP is supplied to the variable length code encoder 22. On the other hand, when the inter prediction is selected, the prediction scheme control unit 28 supplies the inter prediction image Pred_Inter as the prediction image Pred to the adder 31 and the subtractor 32, and the reference image index supplied from the inter prediction image generation unit 26. RI and an estimated motion vector index PMVI and a motion vector residual MVD supplied from a motion vector redundancy deleting unit 29 (described later) are supplied to the variable length code encoding unit 22 as prediction parameters PP.

  By subtracting the prediction image Pred selected by the prediction method control unit 28 from the encoding target image, a prediction residual D is generated by the subtractor 32. The prediction residual D generated by the subtractor 32 is DCT (or DST) transformed / quantized by the transform / quantization unit 21 as described above. On the other hand, by adding the prediction image Pred selected by the prediction method control unit 28 to the prediction residual D generated by the inverse quantization / inverse transformation unit 23, the adder 31 generates a local decoded image P. Generated. The local decoded image P generated by the adder 31 passes through the deblocking filter 33, the adaptive offset filter 80, and the adaptive filter 70, and is then stored in the buffer memory 24 as a filtered decoded image P_FL. Used as

  The motion vector redundancy deletion unit 29 deletes redundancy in the motion vector mv detected by the motion vector detection unit 27. Specifically, (1) an estimation method used for estimating the motion vector mv is selected, (2) an estimated motion vector pmv is derived according to the selected estimation method, and (3) the estimated motion vector pmv is subtracted from the motion vector mv. As a result, a motion vector residual MVD is generated. The motion vector redundancy deletion unit 29 supplies the generated motion vector residual MVD to the prediction method control unit 28 together with the estimated motion vector index PMVI indicating the selected estimation method.

  The deblocking filter 33 determines the block boundary in the local decoded image P when the difference between the pixel values of pixels adjacent to each other via the block boundary in the local decoded image P or the CU boundary is smaller than a predetermined threshold. Alternatively, smoothing of the block boundary or an image near the CU boundary is performed by performing deblocking processing on the CU boundary. The image subjected to the deblocking process by the deblocking filter 33 is output to the adaptive offset filter 80 as a deblocked decoded image P_DB.

  The adaptive offset filter 80 generates an offset filtered decoded image P_OF by performing an adaptive offset filter process on the deblocked decoded image P_DB supplied from the deblocking filter 33. The generated offset filtered decoded image P_OF is supplied to the adaptive filter 70. Since a specific configuration of the adaptive offset filter 80 will be described later, description thereof is omitted here.

  The adaptive filter 70 generates a filtered decoded image P_FL by performing an adaptive filter process on the offset filtered decoded image P_OF supplied from the adaptive offset filter 80. The filtered decoded image P_FL that has been filtered by the adaptive filter 70 is stored in the buffer memory 24. The filter coefficient used by the adaptive filter 70 is determined so that the error between the filtered decoded image P_FL and the encoding target image # 10 becomes smaller, and the filter coefficient thus determined is the filter parameter. It is encoded as FP and transmitted to the video decoding device 1.

(Adaptive offset filter 80)
Next, the adaptive offset filter 80 will be described with reference to FIG. FIG. 16 is a block diagram showing a configuration of the adaptive offset filter 80. As shown in FIG. 16, the adaptive offset filter 80 includes an adaptive offset filter information setting unit 81 and an adaptive offset filter processing unit 82.

  Also, as shown in FIG. 16, the adaptive offset filter information setting unit 81 includes an offset calculation unit 811, an offset clip unit 812, and an offset information selection unit 813.

(Offset calculation unit 811)
For all QAOUs up to a predetermined division depth included in the target processing unit (for example, LCU), the offset calculation unit 811 performs all offset types and all classes that exist according to the hierarchy of the processing target QAOU. The offset of is calculated. Here, the offset type and class are the same as those described in the description of the video decoding device 1.

  Next, the flow of processing in the offset calculation unit 811 will be described with reference to FIG. FIG. 17 is a flowchart showing the flow of processing by the offset calculation unit 811.

(Step S201)
First, the offset calculation unit 811 starts a first loop using the QAOMU number of the target QAOMU to be processed as a loop variable. The QAOMU number is a number for identifying each QAOMU, for example, as shown in FIG. The QAOMU number is assigned from the division depth 0 to the maximum division depth. When the maximum division depth is 4, as shown in FIG. 18, values of 0 to 340 are assigned as QAOMU numbers for all QAOMUs (1 + 4 + 16 + 64 + 256 = 341) of the division depth.

For example, in FIG.
(A) shows the QAOMU number assigned to the QAOMU whose division depth is 0,
(B) shows the QAOMU number assigned to the QAOMU whose division depth is 1.
(C) shows a QAOMU number assigned to a QAOMU having a division depth of 2.
(D) shows the QAOMU number assigned to the QAOMU whose division depth is 3.
In (e), a QAOMU number assigned to a QAOMU having a division depth of 4 is shown.

  Therefore, in the case of the example shown in FIGS. 18A to 18E, the first loop is a loop from QAOMU number = 0 to QAOMU number = 340.

(Step S202)
Subsequently, the offset calculation unit 811 starts a second loop in which an offset type that can be selected for the target QAOMU is a loop variable. In this embodiment, 0 to 6 are used as sao_type_idx, but when sao_type_idx = 0 (no offset processing), the offset is 0, so the second loop is a loop from offset type 1 to offset type 6 It is.

(Step S203)
Subsequently, the offset calculation unit 811 starts a third loop using the pixels included in the target QAOMU as a unit.

(Step S204)
Subsequently, the offset calculation unit 811 classifies the target pixel into one of a plurality of classes. More specifically, it is classified into one of selectable classes according to the component of the pixel to be processed. For example, when the component to be processed is luminance and the offset type that is the second loop variable is 1 to 4, the target pixel is classified into any of classes 1 to 4. The classification process in this step is the same as the classification process performed by the class classification unit 624 included in the adaptive offset filter 60 in the video decoding device 1.

  For the target QAOMU, the number of pixels classified into the class count [component] [part_idx] [sao_type_idx] [class_idx] is calculated for each class. Here, part_idx represents a QAOU index.

(Step S205)
Subsequently, the offset calculation unit 811 obtains the difference pixel value in the target pixel by taking the difference between the pixel value of the deblocked decoded image P_DB in the target pixel and the pixel value of the encoding target image # 10 in the target pixel. calculate. More specifically, P_DB (x, y) −Org (x, y) is calculated when the position of the target pixel is (x, y). Here, P_DB (x, y) represents the pixel value of the deblocked decoded image P_DB at the target pixel, and Org (x, y) represents the pixel value of the encoding target image # 10 at the position of the target pixel. Represents.

(Step S206)
This step is the end of the third loop. When this step is completed, the difference pixel values are calculated for all the pixels included in the target QAOU.

(Step S207)
Subsequently, the offset calculation unit 811 calculates an offset by dividing the sum of the difference pixel values for each pixel included in the target QAOMU for each class by the number of classifications of the class. More specifically, the offset calculation unit 811 calculates an offset offset [component] [part_idx] [sao_type_idx] [class_idx] for the target QAOMU, the target offset type, and the target class using the following expression.

offset [component] [part_idx] [sao_type_idx] [class_idx] =
Σ (P_DB (x, y) -Org (x, y)) / count [component] [part_idx] [sao_type_idx] [class_idx]
Here, the symbol Σ indicates that a difference sum is calculated for pixels classified into the target class specified by class_idx in the target QAOMU specified by part_idx and the target offset type specified by sao_type_idx.

(Step S208)
This step is the end of the second loop.

(Step S209)
This step is the end of the third loop.

  Through the above processing, the offset calculation unit 811 calculates offsets for all offset types and all classes for all QAOMUs up to a predetermined division depth included in the target LCU.

  The offset calculation unit 811 supplies offset information including the offset, offset type, class, and QAOU structure information representing the QAOU division structure calculated by the above processing to the offset clip unit 812.

(Offset clip part 812)
The offset clip unit 812 performs clip processing on the offset supplied from the offset calculation unit 811 by any one of clip processing 1 and clip processing 2 as described below.

(Clip processing 1)
The offset clip unit 812 performs clip processing using the maximum value and the minimum value in order to limit the values that each offset supplied from the offset calculation unit 811 can take. For example, each offset can be represented by 4 bits by clipping to a value from -8 to 7. Each clipped offset is supplied to the offset information selection unit 813. The bit width to be clipped is set according to the bit depth of the image and the bit depth of the offset, as in the video decoding device 1.

  Thus, by clipping each offset, the memory size of a memory (not shown) in which each offset is stored can be reduced. Also, by reducing the amount of offset code included in the encoded data # 1, it is possible to improve the encoding efficiency. Further, since excessive offset addition is suppressed, appropriate image quality is guaranteed.

(Clip processing 2)
The offset clip unit 812 may be configured to set the clip range of each offset supplied from the offset calculation unit 811 to a different range depending on the offset type.

  For example, when the offset type is an edge offset, the clip range is set so that the number of offset bits is 8 bits, and when the offset type is a band offset, the number of offset bits is 4 bits. More generally, when the offset type is an edge offset, the number of offset bits is B0 bits, and when the offset type is a band offset, the number of offset bits is B1 bits. If the offset is required, the clip range is set by determining the number of bits of the offset so that B0> B1 is satisfied.

  Thus, by making the number of offset bits different according to the offset type, it is possible to improve the encoding efficiency without requiring an excessive memory size for the memory for storing each offset.

Note that when the width of the offset range is 2 ^m −1 or more and less than 2 ^m , m-bit fixed-length encoding can be used as an encoding method for encoding the offset. More specifically, Trunked unary coding with a maximum value of th or Truncated Rice coding can be used.

  The clip processing obtained by combining the above clip processing 1 and 2 is also included in this embodiment. Further, the adaptive offset filter 80 may be configured not to include the offset clip unit 812.

  Further, the offset clip unit 812 may be configured to set the clip range of each offset supplied from the offset calculation unit 811 to a different range according to the component represented by the pixel value to be processed.

  For example, when the component is luminance, the clip range is set so that the offset bit number is 4 bits, and when the component is color difference, the offset bit number is 5 bits. If the image quality deterioration due to the color difference becomes large due to the fact that a larger value is used depending on the quantization width, it is preferable to perform the above.

(Offset information selection unit 813)
The offset information selection unit 813 determines an offset type, a class, a combination of offsets, and a corresponding QAOU partition structure with a smaller RD cost (Rate-Distortion cost), and determines the determined offset type, class, offset, and QAOU information indicating the corresponding QAOU division structure is supplied to the variable-length code encoding unit 22. Also, the offset information selection unit 813 supplies the determined offset to the adaptive offset filter processing unit 82 for each QAOU.

  The processing of the offset information selection unit 813 will be described in more detail with reference to FIGS. FIG. 19 is a flowchart showing the flow of processing by the offset information selection unit 813.

(Step S301)
First, the offset information selection unit 813 starts a first loop using the QAOMU number of the target QAOMU to be processed as a loop variable.

(Step S302)
Subsequently, the offset information selection unit 813 starts a second loop in which an offset type that can be selected for the target QAOMU is a loop variable. Here, since it may be selected not to perform the adaptive offset filter processing, the second loop is a loop from the offset type 0 to the offset type 6.

(Step S303)
Subsequently, the offset information selection unit 813 calculates a square error between the offset filtered decoded image P_OF and the encoding target image # 10 in the target QAOMU for the target offset type.

(Step S304)
This step is the end of the second loop.

(Step S305)
This step is the end of the first loop. At the end of the first and second loops, square errors for all offset types are calculated for each QAOMU.

(Step S306)
Subsequently, the offset information selection unit 813 determines a QAOU partition structure in which the RD cost is smaller among the QAOU partition structures that divide a target processing unit (for example, LCU) into QAOUs.

  A specific processing example of the offset information selection unit 813 in this step will be described as follows with reference to FIGS.

  FIG. 20 is a diagram showing an outline of calculating a square error for each offset type for a QAOU having a QAOU index of “q”. As shown in FIG. 20, the offset information selection unit 813 calculates a square error for each offset type for all QAOMUs. The offset type that minimizes the calculated square error is set as the offset type of the QAOU. Thereby, the offset type is determined for all QAOMUs (QAOMU numbers 0 to 340).

  Next, the offset information selection unit 813 calculates the RD cost J0 when the division depth is 0 and the RD costs J1 to J4 of each QAOMU when the division depth is 1 (FIGS. 21A and 21B). )). Here, the division depth 1 division is adopted assuming that the total RD cost (J1 + J2 + J3 + J4) of division depth 1 is smaller than the RD cost J0 of division depth 0 (FIG. 21C).

  Subsequently, the offset information selection unit 813 calculates the RD costs J5 to J20 of each QAOMU when the division depth is set to 2 (FIG. 21 (d)). The offset information selection unit 813 calculates the sum (J5 + J6 + J9 + J10, J7 + J8 + J11 + J12) of the RD cost (J1, J2, J3, J4) of the QAOMU at the division depth 1 and the RD cost of the QAOMU at the division depth 2 included in each QAOMU at the division depth 1. , J13 + J14 + J17 + J18, J15 + J16 + J19 + J20), and if the sum of the RD costs of the QAOMUs with the division depth 2 is smaller, the QAOMU with the division depth 1 is updated to the QAOMU with the division depth 2 (FIG. 21 (d)). ). In this way, the process of updating the RD cost of each QAOMU to a smaller one as compared with the RD cost in the case of further one-layer division is repeated until the maximum division depth is reached. As a result, a QAOU partition structure with a smaller RD cost is determined. FIG. 21 (e) shows a split state in which QAOMU1 and QAOMU4 are selected from QAOMUs with a division depth of 1, and QAOMUs with a division depth of 2 are selected in the other cases.

(Adaptive offset filter processing unit 82)
The adaptive offset filter processing unit 82 adds the offset supplied from the offset information selection unit 813 to each pixel of the deblocked decoded image P_DB in the target QAOU. The adaptive offset filter processing unit 82 outputs an image obtained by performing processing on all the QAOUs included in the deblocked decoded image P_DB as an offset filtered decoded image P_OF. Note that the configuration of the adaptive offset filter processing unit 82 is the same as that of the adaptive offset filter processing unit 62, and thus the description thereof is omitted here.

  Moreover, the structure using an adaptive clip type may be sufficient. That is, an adaptive clip (AC) type may be provided as one of the offset types. In this case, there are three offset types: EO, BO, and AC. In the adaptive clip type (AC), the pixel value is corrected so that the pixel value x satisfies c1 ≦ x ≦ c2 by using a clip of the lower limit value c1 and the upper limit value c2 without using an offset.

  FIG. 22 is a diagram for explaining the adaptive clip type. FIG. 22A shows an example in which clipping is performed using the lower limit value cl1 and the upper limit value cl2 when the bit depth is 8 bits. FIGS. 22B and 22C show the lower limit value cl1 and the upper limit value cl2. For example, cl1 = 16 and cl2 = 235 are used for luminance, and cl1 = 16 and cl2 = 240 are used for color differences. It may be used when the lower limit value and the upper limit value are determined in a range different from the range that can be expressed by the bit depth of the pixel value, such as a television broadcast video signal.

  By using the adaptive clip type, an offset encoding process and a memory for holding a large number of offsets are not required. When the lower limit value and upper limit value of a clip are adaptively given instead of fixed values, the lower limit value and upper limit value may be encoded. In this case, it is preferable to encode a difference from an appropriate fixed value, for example, an offset from 16 as the lower limit value and an offset from 235 as the upper limit value. In particular, since the upper limit value is a large value, the encoding efficiency is not lowered by doing so.

(Reduction of clip processing)
As already described with respect to the video decoding device 1, clip processing can be reduced in the band offset processing. Here, a processing method in the moving image encoding device 2 for generating encoded data # 1 corresponding to each clip processing reduction method in the moving image decoding device 1 will be described. Note that description of the already described contents is omitted as appropriate.

(Clip processing reduction method 1)
The processing of the video encoding device 2 corresponding to the method described in the section of the clip processing reduction method 1 in the video decoding device 1 will be described. The configuration of the adaptive offset filter processing unit 82 in the video encoding device 2 is the same as that of the adaptive offset filter processing unit 62 of the video decoding device 1, and a modification of the adaptive offset filter processing unit 62 in the description of the video decoding device 1. The example described above is applied as it is.

(Clip processing reduction method 2)
The processing of the video encoding device 2 corresponding to the method described in the section 2 on the clip processing reduction method 2 in the video decoding device 1 is also applied in the description of the video decoding device 1 in the same manner as the method 1 described above. The example described as a modification of the offset filter processing unit 62 is applied as it is. In addition, band class conversion tables 2901 and 2902 shown in (a) or (b) of FIG. 29 may be used in the adaptive offset filter information setting unit 81.

(Clip processing reduction method 3)
The processing of the video encoding device 2 corresponding to the method described in the section of the clip processing reduction method 3 in the video decoding device 1 will be described. As described above, in the adaptive offset filter information setting unit 81 of the moving image encoding device 2, the offset clip unit 812 clips the offset into a predetermined value range. Here, when the offset type is a band offset, clip processing can be reduced by limiting the offset value according to the class value range. Specifically, if the offset range in the class corresponding to the offset restriction area OLR0 shown in FIG. 28 is restricted to 0 or a positive value, the pixel value after the offset addition is clearly not less than 0. Similarly, if the range of the offset in the class corresponding to the offset restriction region OLR1 is restricted to 0 or a negative value, the pixel value after the addition of the offset is clearly not greater than max. However, these restrictions are performed in a range that does not exceed a predetermined offset range determined based on the accuracy of the offset, the division depth, the pixel type, and the like.

(Clip processing reduction method 4)
As for the processing of the video encoding device 2 corresponding to the method described in the section 4 on the clip processing reduction method 4 in the video decoding device 1, two or more sets are performed in the offset clip unit 812 similarly to the method 3 described above. It is preferable to apply a restriction according to the value range of the class corresponding to each of the set offset restriction areas. As a result, the pixel value can be brought close to a specified range, and clip processing can be reduced.

  In any of the above methods, if encoded data encoded using an offset limited region that does not cause clipping processing based on the offset value range and pixel bit depth is used, clipping processing in the video decoding device 1 is performed. Can be reduced and the processing speed can be increased.

(Appendix 1)
As described above, the method of switching the conversion table for class classification according to the parameter condition including the pixel value component can be combined with other parameter conditions. For example, processing or conversion according to the type of component as described above only when the QAOU hierarchy is equal to or greater than a threshold, when the size of the QAOU is equal to or smaller than a predetermined threshold, or when the QP value is equal to or greater than a predetermined threshold. The same process or conversion table as the luminance pixel may be used for the chrominance pixels without changing the table in other cases.

(Appendix 2)
In addition, only for a specific type of picture, for example, a B picture or a non-reference picture (IDR picture), processing according to the parameter condition as described above or conversion table switching may be performed. For example, since the I picture has a relatively smaller error than the P picture and the B picture, the conversion table is not switched between the luminance pixel and the chrominance pixel in the I picture, and the above-described conversion is performed only in the P picture and the B picture. It is good to switch the table.

(Appendix 3)
Further, as described above, processing corresponding to the parameter condition or switching of the conversion table may be performed according to the encoding parameter representing the encoding profile, level, and the like. For example, if priority is given to a small amount of processing and complexity, the conversion table switching as described above is not performed, and if not, the conversion table is switched.

(Application examples)
The moving picture decoding apparatus 1 and the moving picture encoding apparatus 2 described above can be used by being mounted on various apparatuses that perform moving picture transmission, reception, recording, and reproduction. Note that the moving image may be a natural moving image captured by a camera or the like, or an artificial moving image (including CG and GUI) generated by a computer or the like).

  First, the fact that the above-described moving picture decoding apparatus 1 and moving picture encoding apparatus 2 can be used for transmission and reception of moving pictures will be described with reference to FIG.

  FIG. 23A is a block diagram illustrating a configuration of the transmission apparatus A in which the moving image encoding apparatus 2 is mounted. As shown in FIG. 23 (a), the transmitting apparatus A encodes a moving image, obtains encoded data, and modulates a carrier wave with the encoded data obtained by the encoding unit A1. A modulation unit A2 that obtains a modulation signal by the transmission unit A2 and a transmission unit A3 that transmits the modulation signal obtained by the modulation unit A2. The moving image encoding device 2 described above is used as the encoding unit A1.

  The transmission apparatus A has a camera A4 that captures a moving image, a recording medium A5 that records the moving image, an input terminal A6 for inputting the moving image from the outside, as a supply source of the moving image that is input to the encoding unit A1. You may further provide image processing part A7 which produces | generates or processes an image. FIG. 23A illustrates a configuration in which the transmission apparatus A includes all of these, but some of them may be omitted.

  The recording medium A5 may be a recording of a non-encoded moving image, or a recording of a moving image encoded by a recording encoding scheme different from the transmission encoding scheme. It may be a thing. In the latter case, a decoding unit (not shown) for decoding the encoded data read from the recording medium A5 according to the recording encoding method may be interposed between the recording medium A5 and the encoding unit A1.

  FIG. 23B is a block diagram illustrating a configuration of the receiving device B on which the moving image decoding device 1 is mounted. As shown in FIG. 23B, the receiving device B includes a receiving unit B1 that receives a modulated signal, a demodulating unit B2 that obtains encoded data by demodulating the modulated signal received by the receiving unit B1, and a demodulating unit. A decoding unit B3 that obtains a moving image by decoding the encoded data obtained by B2. The moving picture decoding apparatus 1 described above is used as the decoding unit B3.

  The receiving device B has a display B4 for displaying a moving image, a recording medium B5 for recording the moving image, and an output terminal for outputting the moving image as a supply destination of the moving image output from the decoding unit B3. B6 may be further provided. FIG. 23B illustrates a configuration in which the receiving apparatus B includes all of these, but a part of the configuration may be omitted.

  Note that the recording medium B5 may be for recording an unencoded moving image, or is encoded by a recording encoding method different from the transmission encoding method. May be. In the latter case, an encoding unit (not shown) that encodes the moving image acquired from the decoding unit B3 in accordance with the recording encoding method may be interposed between the decoding unit B3 and the recording medium B5.

  Note that the transmission medium for transmitting the modulation signal may be wireless or wired. Further, the transmission mode for transmitting the modulated signal may be broadcasting (here, a transmission mode in which the transmission destination is not specified in advance) or communication (here, transmission in which the transmission destination is specified in advance). Refers to the embodiment). That is, the transmission of the modulation signal may be realized by any of wireless broadcasting, wired broadcasting, wireless communication, and wired communication.

  For example, a broadcasting station (such as broadcasting equipment) / receiving station (such as a television receiver) of terrestrial digital broadcasting is an example of a transmitting device A / receiving device B that transmits and receives a modulated signal by wireless broadcasting. A broadcasting station (such as broadcasting equipment) / receiving station (such as a television receiver) for cable television broadcasting is an example of a transmitting device A / receiving device B that transmits and receives a modulated signal by cable broadcasting.

  Also, a server (workstation or the like) / client (television receiver, personal computer, smartphone, etc.) such as a VOD (Video On Demand) service or a video sharing service using the Internet transmits and receives a modulated signal by communication. This is an example of the A / reception device B (usually, either a wireless or wired transmission medium is used in a LAN, and a wired transmission medium is used in a WAN). Here, the personal computer includes a desktop PC, a laptop PC, and a tablet PC. The smartphone also includes a multi-function mobile phone terminal.

  Note that the client of the video sharing service has a function of encoding a moving image captured by a camera and uploading it to the server in addition to a function of decoding the encoded data downloaded from the server and displaying it on the display. That is, the client of the video sharing service functions as both the transmission device A and the reception device B.

  Next, the fact that the above-described moving picture decoding apparatus 1 and moving picture encoding apparatus 2 can be used for recording and reproduction of moving pictures will be described with reference to FIG.

  FIG. 24A is a block diagram showing a configuration of a recording apparatus C in which the above-described moving picture decoding apparatus 1 is mounted. As shown in FIG. 24 (a), the recording device C encodes a moving image to obtain encoded data, and writes the encoded data obtained by the encoding unit C1 to the recording medium M. And a writing unit C2. The moving image encoding device 2 described above is used as the encoding unit C1.

  The recording medium M may be of a type built in the recording device C, such as (1) HDD (Hard Disk Drive) or SSD (Solid State Drive), or (2) SD memory. It may be of the type connected to the recording device C, such as a card or USB (Universal Serial Bus) flash memory, or (3) DVD (Digital Versatile Disc) or BD (Blu-ray Disc: registration) (Trademark) or the like may be mounted on a drive device (not shown) built in the recording apparatus C.

  The recording apparatus C also serves as a moving image supply source to be input to the encoding unit C1, a camera C3 that captures moving images, an input terminal C4 for inputting moving images from the outside, and reception for receiving moving images. A unit C5 and an image processing unit C6 that generates or processes an image may be further provided. FIG. 24A illustrates a configuration in which the recording apparatus C includes all of these, but some of them may be omitted.

  The receiving unit C5 may receive an unencoded moving image, or receives encoded data encoded by a transmission encoding method different from the recording encoding method. You may do. In the latter case, a transmission decoding unit (not shown) that decodes encoded data encoded by the transmission encoding method may be interposed between the reception unit C5 and the encoding unit C1.

  Examples of such a recording apparatus C include a DVD recorder, a BD recorder, and an HD (Hard Disk) recorder (in this case, the input terminal C4 or the receiving unit C5 is a main source of moving images). In addition, a camcorder (in this case, the camera C3 is a main source of moving images), a personal computer (in this case, the receiving unit C5 or the image processing unit C6 is a main source of moving images), a smartphone (this In this case, the camera C3 or the receiving unit C5 is a main source of moving images).

  FIG. 24B is a block diagram illustrating a configuration of the playback device D in which the above-described video decoding device 1 is mounted. As shown in FIG. 24 (b), the reproducing device D obtains a moving image by decoding the read data D1 read from the read data D and the read data read by the read unit D1. And a decoding unit D2. The moving picture decoding apparatus 1 described above is used as the decoding unit D2.

  The recording medium M may be of a type built in the playback device D such as (1) HDD or SSD, or (2) such as an SD memory card or USB flash memory. It may be of a type connected to the playback device D, or (3) may be loaded into a drive device (not shown) built in the playback device D, such as DVD or BD. Good.

  Further, the playback device D has a display D3 for displaying a moving image, an output terminal D4 for outputting the moving image to the outside, and a transmitting unit for transmitting the moving image as a supply destination of the moving image output by the decoding unit D2. D5 may be further provided. FIG. 24B illustrates a configuration in which the playback apparatus D includes all of these, but some of them may be omitted.

  The transmission unit D5 may transmit a non-encoded moving image, or transmits encoded data encoded by a transmission encoding method different from the recording encoding method. You may do. In the latter case, an encoding unit (not shown) that encodes a moving image with a transmission encoding method may be interposed between the decoding unit D2 and the transmission unit D5.

  Examples of such a playback device D include a DVD player, a BD player, and an HDD player (in this case, an output terminal D4 to which a television receiver or the like is connected is a main moving image supply destination). . In addition, a television receiver (in this case, the display D3 is a main destination of moving images), a desktop PC (in this case, the output terminal D4 or the transmission unit D5 is a main destination of moving images), Laptop type or tablet type PC (in this case, display D3 or transmission unit D5 is the main supply destination of moving images), smartphone (in this case, display D3 or transmission unit D5 is the main supply destination of moving images) ), Digital signage (also referred to as an electronic signboard or an electronic bulletin board, and the display D3 or the transmission unit D5 is the main supply destination of moving images) is an example of such a playback device D.

(Configuration by software)
Finally, each block of the moving picture decoding apparatus 1 and the moving picture encoding apparatus 2, in particular, the variable length code decoding unit 13, the motion vector restoration unit 14, the inter prediction image generation unit 16, the intra prediction image generation unit 17, and the prediction method determination Unit 18, inverse quantization / inverse transform unit 19, deblocking filter 41, adaptive filter 50, adaptive offset filter 60, transform / quantization unit 21, variable length code encoding unit 22, inverse quantization / inverse transform unit 23, The intra prediction image generation unit 25, the inter prediction image generation unit 26, the motion vector detection unit 27, the prediction scheme control unit 28, the motion vector redundancy deletion unit 29, the deblocking filter 33, the adaptive filter 70, and the adaptive offset filter 80 are integrated. It may be realized by hardware by a logic circuit formed on a circuit (IC chip), or a CPU (central proce It may be realized by software using a ssing unit).

  In the latter case, the moving picture decoding apparatus 1 and the moving picture encoding apparatus 2 include a CPU that executes instructions of a control program that realizes each function, a ROM (read only memory) that stores the program, and a RAM that expands the program. (Random access memory), a storage device (recording medium) such as a memory for storing the program and various data. An object of the present invention is to provide a computer with program codes (execution format program, intermediate code program, source program) of control programs for the video decoding device 1 and the video encoding device 2 which are software for realizing the functions described above. A readable recording medium is supplied to the moving picture decoding apparatus 1 and the moving picture encoding apparatus 2, and the computer (or CPU or MPU (micro processing unit)) is recorded on the recording medium. This can also be achieved by reading out and executing.

  Examples of the recording medium include tapes such as a magnetic tape and a cassette tape, a magnetic disk such as a floppy (registered trademark) disk / hard disk, a CD-ROM (compact disc read-only memory) / MO disk (magneto-optical disc). ) / MD (Mini Disc) / DVD (digital versatile disc) / CD-R (CD Recordable) and other optical disks, IC cards (including memory cards) / optical cards, mask ROM / EPROM (Erasable programmable read-only memory) / EEPROM (electrically erasable and programmable read-only memory) / semiconductor memories such as flash ROM, or logic circuits such as PLD (Programmable logic device) and FPGA (Field Programmable Gate Array) Can be used.

  Further, the moving picture decoding apparatus 1 and the moving picture encoding apparatus 2 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network. The communication network is not particularly limited as long as it can transmit the program code. For example, Internet, intranet, extranet, LAN (local area network), ISDN (integrated services digital network), VAN (value-added network), CATV (community antenna television / cable television) communication network, virtual private network (virtual private network) network), telephone line network, mobile communication network, satellite communication network, and the like. The transmission medium constituting the communication network may be any medium that can transmit the program code, and is not limited to a specific configuration or type. For example, infra red such as IrDA (infrared data association) or remote control even with wired such as IEEE (institute of electrical and electronic engineers) 1394, USB, power line carrier, cable TV line, telephone line, ADSL (asymmetric digital subscriber loop) line , Bluetooth (registered trademark), IEEE 802.11 wireless, HDR (high data rate), NFC (Near Field Communication), DLNA (Digital Living Network Alliance), mobile phone network, satellite line, terrestrial digital network, etc. Is possible. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the claims, and the embodiments can be obtained by appropriately combining technical means disclosed in different embodiments. The form is also included in the technical scope of the present invention.

  The present invention can be suitably used for an image filter that performs offset filtering on image data. Further, the present invention can be suitably applied to a decoding device that decodes encoded data and an encoding device that encodes encoded data.

1 video decoding device (image decoding device)
2 Video encoding device (image encoding device)
16 Inter prediction image generation unit (prediction image generation means)
17 Intra prediction image generation unit (prediction image generation means)
25 Intra prediction image generation unit (prediction image generation means)
26 Inter prediction image generation unit (prediction image generation means)
60 Adaptive offset filter (image filter device, offset addition means)
80 Adaptive offset filter (image filter device, offset addition means)
623 Offset type deriving unit (offset determining means)
624 class classification unit (class classification method selection means, class classification means, offset selection method determination means, offset selection means)
625 Offset attribute setting unit 626 Offset derivation unit (offset determination means)
627 Offset addition unit (filter means, offset correction means)

Claims (24)

An image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image composed of a plurality of unit regions,
Class classification method selection means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are to be classified, from a plurality of candidates;
Class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means;
An offset determining unit that determines an offset to be added to the pixel value of the pixel classified by the class classification unit for each unit region according to the class into which the pixel is classified;
An image filter device comprising: a filter unit that adds the offset determined by the offset determination unit to a pixel value of a pixel included in the unit region.   2. The class classification method selection unit selects a class classification method from different candidate groups depending on whether a pixel value of a pixel included in the unit region indicates a color difference or a luminance. The image filter device described in 1.   The different candidate groups are defined as a candidate group selected by the class classification method selection unit when the pixel value of the pixel included in the unit area indicates a color difference, and the pixel of the pixel included in the unit area. When the candidate group to be selected when the value indicates luminance is a luminance candidate group, the number of classes indicated by the class classification method included in the color difference candidate group is the number of classes indicated by the class classification method included in the luminance candidate group. The image filter device according to claim 2, wherein the number of the image filter devices is larger.   The different candidate groups are defined as a candidate group selected by the class classification method selection unit when the pixel value of the pixel included in the unit area indicates a color difference, and the pixel of the pixel included in the unit area. When the candidate group to be selected when the value indicates luminance is a luminance candidate group, the number of types of class classification methods included in the color difference candidate group is the number of types of class classification methods included in the luminance candidate group. The image filter device according to claim 2, wherein the number of the image filter devices is smaller.   The filter means, when the class classification method selection means selects a class classification method for classifying the target pixel based on the state of the edge around the pixel, and the pixel value of the target pixel indicates a color difference, The offset is also added to the pixel value of a pixel classified into a class to which no offset is added if the pixel value of the target pixel indicates luminance. Image filter device.   The class classification method selection means selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates, and two or more candidates among the plurality of candidates The image filter device according to claim 1, wherein there is an overlap in a range of pixel values classified into a class in which an offset addition process is performed.   The class classification method selection means, when the pixel value of the target pixel indicates a color difference, selects a class classification method from among candidate class classification methods in which classes in which pixels in a predetermined range are classified overlap The image filter device according to claim 5.   8. The image filter device according to claim 7, wherein the predetermined value range is a value range indicating an achromatic color or low saturation. The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes a plurality of conversion tables that associate pixel values with classes,
2. The image filter according to claim 1, wherein among the plurality of conversion tables, pixel values included in a predetermined range of at least two conversion tables are associated with a class on which an offset addition process is performed. apparatus. In the conversion table, the pixel value is divided into a plurality of value ranges, and each divided value range is associated with a class.
The image filter device according to claim 9, wherein the predetermined range is a range of a value range associated with a class having a smaller value range than other classes. The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes a plurality of conversion tables that associate pixel values with classes,
The image filter device according to claim 1, wherein the plurality of conversion tables have different pixel value ranges associated with classes for which offset addition processing is performed. The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the conversion tables includes pixel values divided into a plurality of value ranges, and each divided value range and class Are associated with each other,
The image filter device according to claim 1, wherein the plurality of conversion tables have different number of value ranges associated with each class. The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes three or more conversion tables in which pixel values are associated with classes,
The above three or more conversion tables include
A conversion table in which the range of pixel values corresponding to the class for which offset addition processing is performed is wider than other conversion tables,
A conversion table in which the pixel value range corresponding to the class for which the offset addition processing is performed is narrower than the pixel value range in the wide conversion table;
A conversion table in which the range of pixel values corresponding to the class for which the offset addition processing is performed is the both ends of the range of pixel values and a range close to the range,
The image filter device according to claim 1, further comprising: The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes three or more conversion tables in which pixel values are associated with classes,
The above three or more conversion tables include
A first conversion table in which the range of pixel values corresponding to the class for which the offset addition processing is performed includes the center of the range of pixel values;
A second conversion table in which the range of pixel values corresponding to the class to which the offset addition processing is performed includes both ends of the range of pixel values and a range close to the range;
One or more conversion tables in which the pixel value range corresponding to the class for which the offset addition processing is performed includes a range other than the pixel value range in the first conversion table and the second conversion table;
The image filter device according to claim 1, further comprising: The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the above candidates includes three or more conversion tables in which pixel values are associated with classes,
The code length of the encoded data indicating any of the three or more conversion tables is
The shortest when the pixel value range corresponding to the class for which the offset addition process is performed indicates a conversion table including the center of the pixel value range,
The range of pixel values corresponding to the class for which the offset addition processing is performed uses encoded data that is the longest when a conversion table including both ends of the pixel value range and the vicinity thereof is used. The image filter device according to claim 14. The class classification method selection means selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates,
Each of the candidates includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, Is associated with a class,
The class classification method selection means includes a class classification including a conversion table divided into a larger number of value ranges when the pixel values indicate color differences than when the pixel values indicate luminance. The image filter device according to claim 1, wherein a method is selected. The class classification method selection means selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates,
Each of the candidates includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, Is associated with a class,
In the class classification method selection unit, when the pixel value indicates a color difference, the pixel value range corresponding to the class for which the offset addition processing is not performed is the range of the pixel value range and the range close thereto. 2. The image filter apparatus according to claim 1, wherein a class classification method including a certain conversion table is selected. The class classification method selection means selects a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates,
Each of the candidates includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, Is associated with a class,
The class classification method selection means, when the pixel value indicates a color difference,
A first conversion table in which the range of pixel values corresponding to the class for which the offset addition processing is performed includes the center of the range of pixel values;
A third conversion table in which the range of pixel values corresponding to the class for which offset addition processing is performed does not include the center of the range of pixel values;
And a range of pixel values corresponding to a class for which offset addition processing is performed in the first conversion table is narrower than a range of pixel values corresponding to the class for which offset addition processing is performed in the third conversion table 2. The image filter device according to claim 1, wherein a classification method is selected. The class classification method selection means is for selecting a class classification method for performing class classification based on the size of the pixel value of the target pixel from a plurality of candidates.
Each of the candidates includes a plurality of conversion tables in which pixel values and classes are associated with each other, and each of the plurality of conversion tables has pixel values divided into a plurality of value ranges, Is associated with a class,
The plurality of conversion tables include
2. The conversion table according to claim 1, wherein a plurality of non-contiguous ranges in the pixel value range includes a conversion table in which the pixel value ranges corresponding to the class on which the offset addition processing is performed. Image filter device. An image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image composed of a plurality of unit regions,
Class classification method selection means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are to be classified, from a plurality of candidates;
Class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means;
An offset determining unit that determines an offset to be added to the pixel value of the pixel classified by the class classification unit for each unit region according to the class into which the pixel is classified;
The pixel value after adding the offset determined by the offset determination means for the upper limit value, the lower limit value, and the pixel value adjacent to the upper limit value, the lower limit value, or the lower limit value. Offset correcting means for correcting the offset to be added so as not to fall below
An image filter device comprising: a filter unit that adds the offset determined by the offset determination unit or the offset corrected by the offset correction unit to a pixel value of a pixel included in the unit region. An image decoding device that generates a decoded image by adding a generated prediction image to a prediction residual decoded from encoded data,
An offset addition means for adding an offset for each unit region to the addition image obtained by adding the prediction image to the prediction residual;
Output means for outputting an image obtained by adding the offset by the offset adding means as a decoded image;
The offset adding means includes
Class classification method selection means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are to be classified, from a plurality of candidates;
Class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means;
An offset determining unit that determines an offset to be added to the pixel value of the pixel classified by the class classification unit for each unit region according to the class into which the pixel is classified;
An image decoding apparatus, comprising: a filter unit that adds the offset determined by the offset determination unit to a pixel value of a pixel included in the unit region. An image encoding device that generates encoded data by encoding a prediction residual that is a difference between a generated predicted image and an original image,
Offset addition means for adding an offset for each unit region to the decoded image obtained by encoding the original image;
Predicted image generation means for generating the predicted image from an image obtained by adding the offset by the offset adding means;
Encoding means for encoding a prediction residual, which is a difference between the predicted image generated by the predicted image generation means and the original image,
The offset adding means includes
Class classification method selection means for selecting a class classification method, which is a method for determining which of the plurality of classes the pixels included in the unit area are to be classified, from a plurality of candidates;
Class classification means for classifying the pixels according to the class classification method selected by the class classification method selection means;
An offset determining unit that determines an offset to be added to the pixel value of the pixel classified by the class classification unit for each unit region according to the class into which the pixel is classified;
An image coding apparatus comprising: a filter unit that adds the offset determined by the offset determination unit to a pixel value of a pixel included in the unit region. A data structure of encoded data obtained by encoding a prediction residual, which is a difference between a predicted image and an original image, decoded by an image decoding device,
Pixel value data,
Class classification method selection data for selecting a class classification method for determining an offset to be added to the pixel value indicated by the pixel value data, and
The image decoding apparatus classifies a class of a target pixel according to a class classification method indicated by the class classification method selection data, and adds a offset corresponding to the class to the target pixel to generate a decoded image. Data structure of encoded data to be performed. An image filter device that adds an offset selected from a plurality of offsets to each pixel value of an input image,
An offset selection method determining means for determining an offset selection method, which is a method for selecting an offset to be added to the pixel value from the plurality of offsets, for each pixel value;
An offset selection means for selecting an offset from the plurality of offsets according to the offset selection method determined by the offset selection method determination means;
An image filter device comprising: a filter unit that adds the offset selected by the offset selection unit to the pixel value.

Images