JP2011004285A - Image data encoding device, image data encoding method, image data code amount control method, and electronic information apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an image data encoding device for avoiding the fact that compression processing reducing a code amount of image encoding data concentrates in a specific block when an image data is encoded in each block, and for avoiding an overflow of an output buffer memory storing the image encoding data while suppressing a deterioration in image quality.SOLUTION: In an edge component detection processing section 110, an edge component in the horizontal direction of an input image and an edge component in the vertical direction of the input image are detected and then detection results by a horizontal edge addition buffer 111 and a vertical edge addition buffer 113 are integrated in each 8×8 pixel blocks of a JPEG processing unit. A future residual capacity of an output buffer memory is predicted from an addition result of edges and a residual capacity of a JPEG output buffer memory 60. In a quantization processing section 30a, a quantization table with a quantization coefficient changed based on a prediction result and the addition result of the edges is used.

Description

  The present invention relates to an image data encoding device, an image data encoding method, an image data code amount control method, and an electronic information device, and more particularly, to an image data encoding device and image data encoding method for encoding image data, and an image Image data code amount control method for controlling the code amount of image encoded data when data is encoded, and whether encoded data from this image data encoding device is stored in a storage means or is transmitted. The present invention also relates to an electronic information device such as a digital camera or a camera-equipped mobile phone device that performs display processing on a display unit.

  For example, when image data captured by a solid-state imaging device such as a CCD (Charge-Coupled Device) is recorded as digital data on a readable recording medium such as a memory card, a magnetic disk, or a magnetic tape, the amount of data is enormous. Become.

  In a digital still camera or the like, the captured image data is stored on a single memory card, magnetic disk, or one volume of magnetic tape, so that a prescribed number of images are recorded on these readable recording media. There is a need. Similarly, when recording a moving image in a digital VTR, a digital moving image file, etc., it is necessary to record image data for each frame regardless of the amount of image data per frame. In the case of 30 frames per second, 30 sheets of image data (30 frames) are stored per second.

  Therefore, it is necessary to compress the image data so that the amount of data falls within a specified capacity (image memory capacity, image communication capacity, etc.) regardless of still images and moving images. This is because if the data transmission device has a predetermined transmission capability, the data transmission by the data transmission device will not be in time if the data amount of the image data is larger, and if the image memory has a predetermined memory capacity, This is because if there is image data exceeding the memory capacity, data exceeding the memory capacity cannot be stored in the image memory.

  Conventionally, an image data encoding method combining orthogonal image encoding and variable length encoding has been used as a compression method for image data.

  In this image data encoding method, first, image data (pixel value corresponding to each pixel) is divided so as to correspond to a block of a predetermined size, for example, a block of 8 pixels × 8 pixels, and divided blocks. As each orthogonal transform, two-dimensional DCT (Discrete Cosine Translation) processing is performed on the image data. Thereby, image data (for example, luminance data) is converted from time-series data into a DC component and an AC component, which are frequency-sequence data (frequency components).

  For example, image data (64 pixel values) corresponding to an 8 × 8 pixel block is converted into 64 frequency components corresponding to an 8 × 8 matrix. In this matrix, data indicating the value of the DC component DC is at the origin position (0, 0), and data indicating the maximum frequency value of the AC component AC in the horizontal axis direction is at position (7, 7). , 0) generates data indicating the maximum frequency value of the AC component AC in the vertical axis direction, and generates data indicating the maximum frequency value of the AC component AC in the oblique direction at position (7, 7). Further, at intermediate positions other than the four positions of the matrix, the frequency components in the direction related by the respective coordinate positions are generated so that the ones having higher frequencies sequentially appear from the origin side.

  Also, linear quantization corresponding to each frequency component is performed by dividing the data (frequency component) at each coordinate position by the quantization width for each frequency component.

  Thereafter, the quantized value is subjected to Huffman coding as variable length coding. At this time, with respect to the direct current component DC, a difference value from the direct current component DC of a neighboring block is represented by a group number (additional bit number) and additional bits and is Huffman encoded. Encoded data is generated by combining the obtained codeword and additional bits.

  That is, since the DC component value is often close to the DC component value of the adjacent block, the difference from the DC component value of the immediately preceding block is Huffman encoded. For this reason, since the appearance frequency of the difference values close to 0 increases, grouping is first performed based on absolute values, and shorter Huffman codes are assigned to smaller absolute values. Next, an additional bit for distinguishing values in the group is added.

  As for the AC component AC, a scan from a low frequency component to a high frequency component called a zigzag scan is performed, and the number of consecutive invalid components (value is “0”) and the value of the subsequent valid component are calculated. Two-dimensional Huffman coding is performed from the group number. Encoded data is generated by combining the obtained codeword and additional bits.

  In this Huffman coding, in the data distribution of each of the direct current component DC and the alternating current component AC per frame image, the generation frequency of the plurality of levels of pixel values to be coded is centered on this level. The pixel value at the level closer to the central level has a smaller number of data bits when encoded, and the pixel value at a level farther from the central level has a higher number of encoded data bits. In addition, bit allocation is performed and data is encoded to obtain a code word.

  Conventionally, when encoding image data using such an image data encoding method, it has been necessary to control the code amount of the image data according to the capacity of the data recording medium and the capacity of the transmission path for transmitting the image data. It was.

  FIG. 2 is a block diagram illustrating an exemplary configuration of a main part of a conventional image data encoding device disclosed in Patent Document 1. In FIG.

  The image data encoding device 200 compresses and encodes image data and outputs image encoded data. The image data encoding device 200 temporarily stores input image data, and is output from the input storage unit 10. A compression encoding unit 200a that compresses and encodes the image data, and an output storage unit 60 that temporarily stores the encoded image data obtained by the compression encoding of the image data.

  Here, the input storage unit 10 is configured by a JPEG input buffer memory, and the output storage unit 60 is configured by a JPEG output buffer memory.

  Further, the compression encoding unit 200a quantizes the DCT processing unit 20 that is an orthogonal transformation circuit that performs orthogonal transformation on the image data from the JPEG input buffer memory 10 and the DCT data converted by the DCT processing unit 20 to quantize the data. A quantization processing unit 30 that outputs quantized data; a division processing unit 40 that divides the quantized data by a predetermined value; and the quantized data obtained by the division processing is converted into a variable length code such as a Huffman code. A variable-length encoding processing unit 50 that outputs variable-length encoded data to the JPEG output buffer 60.

  Further, the compression encoding unit 200a has a division control unit (not shown) that detects the remaining capacity of the output buffer memory 60 and controls the operation of the division processing unit 40 according to the remaining capacity. .

  The division control unit (not shown) may be provided in the output buffer memory 60 or the division processing unit 40, or may be provided separately from the output buffer memory 60 and the division processing unit 40. Good.

  Detection of the remaining capacity of the output buffer memory 60 by this division control unit (not shown) is performed by detecting the amount of encoded image data (input image data amount) input to the output buffer memory 60 and the output buffer memory 60. Can be performed by detecting a difference data amount with respect to the data amount (output image data amount) of the encoded image data output from. The input image data amount and the output image data amount may be managed by the storage control unit of the output buffer memory 60 to obtain the difference between the input and output.

  Next, the operation will be described.

  First, the image data input to the image data encoding device 200 is temporarily stored in the JPEG input buffer memory 10, and the image data is supplied from the input buffer memory 10 to the compression encoding unit 200a at a constant output rate. The

  The image data supplied to the compression encoding unit 200a is subjected to orthogonal transform processing (DCT processing) in the DCT processing unit 20 and converted into DCT data. Further, the DCT data is quantized by the quantization processing unit 30. Is done. The quantized DCT data (quantized data) is supplied to the variable length coding processing unit 50 via the division processing unit 40, and the variable length coding processing unit 50 performs variable length coding processing. It is supplied to the output buffer memory 60 as encoded image data.

  Here, when the remaining capacity of the output buffer memory 60 is larger than the predetermined remaining capacity, the control signal for operating the division processing unit 40 is not output from the division control unit, and the division process is not performed. Therefore, the data quantized by the quantization processing unit 30 is supplied to the variable length coding processing unit 50 as it is, and the amount of code supplied from the variable length coding processing unit 50 to the output buffer memory 60 is not reduced.

  On the other hand, when the remaining capacity of the output buffer memory 60 becomes smaller than the predetermined remaining capacity, a control signal for operating the division processing unit 40 is output from the division control unit, and division processing is performed. As a result, the variable length encoding processing unit 50 is supplied with data obtained by dividing the data quantized by the quantization processing unit 30 (quantized data) by a predetermined value (for example, 2 or 4). The amount of code supplied from the encoding processing unit 50 to the output buffer memory 60 is reduced. For example, when the quantized data is 8 bits, when the quantized data is divided by “2”, the quantized data becomes 7 bits with no final bit.

  The predetermined remaining capacity is a value preset in the division control unit, and can be, for example, 3/4 of the total capacity of the output buffer memory 60.

JP 2007-36745 A

  As described above, the JPEG image compression technique is a compression technique using variable-length coding, and the compressed image data is kept constant by using a memory that temporarily stores encoded data (compressed image data). Although output can be performed at a rate, if the amount of encoded image data input to the output buffer memory is larger than the remaining capacity of the output buffer memory and the output data transmission rate, an overflow occurs in the output buffer memory. There is a problem.

  Therefore, in the image data encoding device disclosed in Patent Document 1, the remaining capacity of the output buffer memory is monitored for this problem, and the remaining capacity of the output buffer memory for the quantized data to be subjected to variable length encoding. The code amount of the image encoded data is compressed by performing division according to the above.

  However, in the technique disclosed in Patent Document 1, since the code amount is compressed by the remaining capacity of the output buffer memory at the time of quantization, the code amount of the image encoded data output from the variable-length encoding processing unit In response to an increase in the code amount, the division processing of the quantized data input to the variable length coding processing unit for compressing the code amount of the encoded image data is delayed, and the code amount is encoded after the block to be compressed. There is a problem that code amount compression processing concentrates on the blocks to be converted.

  This problem can be suppressed by giving a margin to the threshold of the remaining capacity of the output buffer memory, which is a criterion for whether or not to compress the code amount. In this case, the output buffer memory Since the code amount is also compressed for blocks that do not affect the overflow of the image, another problem that the image quality deteriorates arises.

  The present invention can perform a compression process for reducing the amount of encoded image data while avoiding the compression process from being concentrated on a specific block, and thereby an output buffer memory for storing encoded image data. An object of the present invention is to obtain an image data encoding device, an image data encoding method, an image data code amount control method, and an electronic information device using the image data encoding device capable of avoiding an overflow while suppressing deterioration in image quality. And

  An image data encoding device according to the present invention is an image data encoding device that compresses and encodes image data and outputs image encoded data, and includes a compression encoding unit that compresses and encodes the image data, and the image A temporary storage unit that temporarily stores encoded image data obtained by compression encoding of data, and a storage remaining capacity prediction unit that predicts a storage remaining capacity in the temporary storage unit based on the image data, The compression encoding unit controls the compression encoding processing of the image data based on the predicted storage remaining capacity so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit. Thus, the above object is achieved.

  The present invention provides the image data encoding apparatus, wherein an edge component detection processing unit that detects an edge component from the image data, and an edge component detected by the edge component detection processing unit is a processing unit for encoding the image data. The remaining storage capacity prediction unit predicts the remaining storage capacity in the temporary storage unit based on the integrated value of the edge components integrated by the edge component integration unit. Is preferred.

  In the image data encoding device according to the present invention, the remaining storage capacity prediction unit includes a remaining memory capacity of the temporary storage unit and an integrated value obtained by integrating the edge component by the edge component integrating unit. It is preferable to predict the remaining memory capacity of the temporary storage unit.

  According to the present invention, in the image data encoding device, the edge component detection processing unit detects a horizontal edge component detection filter that detects a horizontal edge component from the image data, and detects a vertical edge component from the image data. And a vertical edge component detection filter.

  The present invention provides the image data encoding apparatus, wherein a horizontal edge integration buffer that integrates a horizontal edge component detected by the horizontal edge component detection filter for each block that is a processing unit of the image data encoding, and the vertical It is preferable to have a vertical edge integration buffer that integrates the vertical edge components detected by the edge component detection filter for each block that is a processing unit for encoding the image data.

  According to the present invention, in the image data encoding device, the storage remaining amount prediction unit includes an integrated value of each horizontal edge component obtained by the horizontal edge integrating buffer and a vertical value obtained by the vertical edge integrating buffer. It is preferable that the buffer remaining amount prediction processing unit predicts the memory remaining capacity of the temporary storage unit from the integrated value of the edge component for each block and the memory remaining capacity of the temporary storage unit.

  According to the present invention, in the image data encoding device, the storage remaining capacity prediction unit may control a future time point to be a remaining capacity prediction target according to the number used for prediction among the edge integrated values for each block. preferable.

  The present invention provides the image data encoding device, wherein the storage remaining capacity prediction unit subtracts a prediction value of a future code amount obtained from the edge integrated value from a current memory remaining capacity of the temporary storage unit, It is preferable to obtain a future buffer remaining capacity prediction result by adding a data transmission amount that depends on the transmission rate of the encoded image data output from the temporary storage unit.

  According to the present invention, in the image data encoding device, the storage remaining capacity prediction unit performs buffer remaining capacity prediction for one or more future time points, and obtains an overflow prediction result from one or more remaining capacity prediction results. It is preferable.

  According to the present invention, in the image data encoding apparatus, the predicted value of the code amount of a block to be compressed in the future includes a horizontal edge integrated value that is an output of a horizontal edge integrating buffer and a vertical edge that is an output of a vertical edge integrating buffer. It is preferable to obtain from an edge integrated value obtained by calculation using the integrated value.

  According to the present invention, in the image data encoding apparatus, the compression processing unit quantizes an orthogonal transform processing unit that converts the image data into frequency components for each block of a predetermined size, and a frequency component corresponding to each block. It is preferable to include a quantization processing unit and a variable length coding unit that performs variable length coding on the quantized frequency component and outputs encoded image data.

  The present invention preferably includes a quantization coefficient calculation processing unit that obtains a quantization coefficient used in the quantization processing unit from the predicted storage remaining capacity in the image data encoding device.

  According to the present invention, in the image data encoding device, the quantization coefficient calculation processing unit calculates a quantization coefficient used in the quantization processing unit from a horizontal edge integrated value and a vertical edge integrated value for each block. It is preferable to calculate separately for the quantization coefficient of the AC component in the axial direction and the quantization coefficient of the AC component in the vertical axis direction.

  The present invention is the above-described image data encoding device, wherein the orthogonal transform processing unit is a DCT processing unit that performs a discrete cosine transform process on the image data for each block to obtain a DCT coefficient corresponding to each block. It is preferable.

  In the image data encoding device according to the present invention, it is preferable that the quantization processing unit quantizes the DCT coefficient corresponding to each block using the quantization coefficient obtained by the quantization coefficient calculation processing unit. .

  In the image data encoding device according to the present invention, it is preferable that the variable length encoding unit converts the quantized data into the image encoded data by Huffman encoding.

  An image data encoding method according to the present invention is an image data encoding method for compressing and encoding image data and outputting image encoded data, the compression encoding step for compressing and encoding the image data, and the image A temporary storage step of temporarily storing the encoded image data obtained by the compression encoding of the data in the temporary storage unit, and a storage remaining capacity prediction step of predicting a storage remaining capacity in the temporary storage unit based on the image data; In the compression encoding step, the image data obtained by the compression encoding does not overflow in the temporary storage unit based on the predicted remaining storage capacity. Thus, the above object is achieved.

  An image data code amount control method according to the present invention is an image data code amount control method for controlling the code amount of image encoded data stored in a temporary storage unit when image data is encoded. An edge component integration step for integrating the edge components of the image data for each block, which is a unit of encoding, and a remaining storage capacity in the temporary storage unit are predicted based on an integration value obtained by integrating the edge components A storage remaining capacity prediction step, and based on the predicted remaining storage capacity, the image data is encoded so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit. The above-mentioned purpose is achieved.

  An electronic information device according to the present invention includes a storage process for storing image encoded data from the above-described image data encoding apparatus in a storage unit, a transmission process for transmitting the image encoded data, and the image encoded data. Based on the above, at least one of display processes for displaying image data on the display unit is performed, and thereby the above-described object is achieved.

  The operation of the present invention will be described below.

  In the present invention, the future buffer remaining capacity is predicted from the edge component of the input image and the current buffer remaining capacity, and the code amount is compressed by changing the coefficient of the quantization table only when an overflow is predicted to occur. Therefore, it is possible to compress the amount of code only when the future overflow is predicted for a block that causes a decrease in the remaining buffer capacity. Conversely, it is predicted that no overflow will occur. In this case, it is possible to avoid unnecessary compression of the code amount and to prevent the image quality from being lost.

  In the present invention, the edge component included in the block is detected separately in the vertical direction and the horizontal direction, so that the coefficient of the quantization table for the prediction of the occurrence of overflow is determined as the vertical direction according to the deviation in the edge direction of the block. Since the change is made in the horizontal direction, the code amount can be compressed while suppressing the deterioration of the image quality.

  As described above, according to the present invention, the storage remaining capacity in the temporary storage unit that temporarily stores the encoded image data obtained by the compression encoding of the image data is predicted based on the image data, and the prediction is performed. Based on the remaining storage capacity, the compression encoding process of the image data is adjusted so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit, so that the code amount of the encoded image data is reduced. The compression processing to be performed can be performed while avoiding the concentration of the compression processing on a specific block, and thereby overflow of the output buffer memory for storing the encoded image data can be avoided while suppressing deterioration in image quality. .

FIG. 1 is a block diagram illustrating an image data encoding apparatus according to Embodiment 1 of the present invention. FIG. 2 is a block diagram showing a conventional image data encoding apparatus having a buffer overflow prevention function.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1)
FIG. 1 is a block diagram illustrating an image data encoding apparatus according to Embodiment 1 of the present invention.

  The image data encoding device 100 according to the first embodiment has a function for preventing overflow of the output buffer memory, which is different from the conventional image data encoding device.

  That is, the image data encoding apparatus 100 according to the first embodiment predicts the remaining storage capacity of the temporary storage unit that temporarily stores the encoded image data obtained by the compression encoding of the image data based on the image data. Based on the predicted remaining storage capacity, the compression encoding of the image data is controlled so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit.

  Hereinafter, the image data encoding apparatus according to the first embodiment will be described in detail.

  The image data encoding apparatus 100 according to the first embodiment includes an edge detection processing unit 110 that performs edge detection on an input image based on the input image data, and an input storage unit that receives the input image data via the edge detection processing unit 110. As a JPEG input buffer memory 10, a compression encoding unit 100 a that performs compression encoding on the image data from the buffer memory 10, and an output storage unit that temporarily stores the encoded image data from the compression encoding unit 100 a JPEG output buffer memory 60.

  Here, the compression encoding unit 100a quantizes the DCT processing unit 20 which is an orthogonal transformation circuit that performs orthogonal transformation on the image data from the JPEG input buffer memory 10 and the DCT data converted by the DCT processing unit 20 to quantize the data. A quantization processing unit 30a that outputs the encoded data, and a variable length encoding processing unit 50 that converts the quantized data into a variable length code such as a Huffman code and outputs the image encoded data to the JPEG output buffer 60. is doing.

  The JPEG input buffer memory 10, the DCT processing unit 20, the variable length encoding processing unit 50, and the JPEG output buffer memory 60 in the image data encoding device 100 of the first embodiment are the same as those in the conventional image data encoding device 200. Is the same as The quantization processing unit 30a is different from that in the conventional image data encoding apparatus 200 in that the DCT coefficient is quantized based on a quantization table given from the outside.

  In the first embodiment, the edge detection processing unit 110 includes a horizontal edge detection filter (not shown) that extracts a horizontal edge component (horizontal edge component) of the input image based on the input image data, and the input image data. And a vertical edge detection filter (not shown) for extracting an edge component (vertical edge component) in the vertical direction of the input image.

  Further, the image data encoding apparatus 100 according to the first embodiment includes a horizontal edge integration buffer 111 that integrates the horizontal edge component from the edge detection processing unit 110 for each block of 8 × 8 pixels that is a JPEG processing unit, and an edge A vertical edge integration buffer 113 for integrating the vertical edge components from the detection processing unit 110 for each block of 8 × 8 pixels which is a JPEG processing unit. These horizontal edge integration buffer 111 and vertical edge integration buffer 113 each require a minimum capacity for the number of blocks in a horizontal row. Here, the capacity corresponding to the number of blocks in a horizontal row is, for example, for the horizontal edge integration buffer, the maximum output value for one block of the horizontal edge detection filter is the number of blocks composed of 8 × 8 pixels arranged in a horizontal row. Only the accumulated capacity. The minimum required capacity in the vertical edge accumulation buffer is the same as that in the horizontal edge accumulation buffer.

  Further, the compression encoding unit 100a includes the integrated value of the horizontal edge component integrated by the horizontal edge integrating buffer 111, the integrated value of the vertical edge component integrated by the vertical edge integrating buffer 113, and the remaining capacity of the JPEG output buffer. The remaining buffer capacity prediction processing unit 112 that predicts the remaining capacity of the output buffer in the future, the prediction result in the remaining buffer capacity prediction processing unit 112, the integrated value of the horizontal edge component, and the integrated value of the vertical edge component Based on the above, the coefficient of the quantization table used in the quantization processing unit 30a is calculated so as to avoid overflow in the output buffer memory, and the quantization table obtained by this calculation is supplied to the quantization processing unit 30 And a quantization coefficient calculation processing unit 114. In this quantization coefficient calculation processing unit 114, the coefficient (quantization coefficient) of the quantization table used in the quantization processing unit 30a is calculated from the horizontal edge integrated value and the vertical edge integrated value for each block on the horizontal axis. Alternatively, the quantization coefficient of the alternating current component (AC) in the direction and the quantization coefficient of the alternating current component (AC) in the vertical axis direction may be obtained separately.

  Next, the operation will be described.

  When input image data is input to the image data encoding apparatus 100 according to the first embodiment, the edge component detection processing unit 110 applies an edge detection filter to the input image data, thereby obtaining an edge component of the input image. Generate. Next, the detected edge components are integrated for each 8 × 8 pixel block which is a JPEG processing unit. At this time, the edge component detection processing unit 110 uses an edge detection filter that can detect a horizontal edge component and a vertical edge component, that is, a horizontal edge detection filter and a vertical edge detection filter.

  Specifically, the edge component detection processing unit 110 extracts a horizontal edge component of the image data using a horizontal edge detection filter, and extracts a vertical edge component of the image data using a vertical edge detection filter. Further, the horizontal edge integration buffer 111 and the vertical edge integration buffer 113 perform edge component integration for each block separately in the horizontal direction and the vertical direction. The integrated value of the horizontal edge component and the integrated value of the vertical edge component are stored in the edge integration buffers 111 and 113, respectively. The edge detection process does not change the information of the image data itself.

  On the other hand, the image data subjected to the edge component detection processing is subjected to compression coding processing after the DCT processing, and therefore has a capacity of at least 8 lines (that is, a capacity of image data of 8 pixels on the display screen). Stored in the input buffer memory 10.

  Further, the image data stored in the input buffer memory 10 is supplied from the input buffer memory 10 to the compression encoding unit 100a at a constant output rate.

  The image data supplied to the compression encoding unit 100a is subjected to orthogonal transform processing (DCT processing) by the DCT processing unit 20 and converted to DCT data. Further, the DCT data is quantized by the quantization processing unit 30a. Is done. The quantized DCT data (quantized data) is supplied to the variable-length encoding processing unit 50, subjected to variable-length encoding processing in this variable-length encoding processing unit 50, and output as image encoded data. It is supplied to the buffer memory 60.

  Here, in the quantization processing unit 30a, the coefficient of the quantization table calculated based on the prediction result in the buffer remaining amount prediction processing unit 112 and the integrated value of the vertical edge component and the integrated value of the horizontal edge component Is used to quantize the DCT data from the DCT processing unit 20.

  Next, the prediction process of the remaining buffer capacity will be described in detail.

  In the prediction process of the remaining buffer capacity, information on the remaining buffer capacity of the current output buffer memory and the edge integrated value of one or more blocks on which future encoding processing is performed are used. Here, as the edge integrated value of one or more blocks, the integrated value of blocks for one line on the screen can be used at the maximum. Of the integrated values for each block integrated in the edge integration buffer, the number of integrated values used for buffer remaining capacity prediction is used to predict the number of blocks after the number of blocks counted from the block that reflects the prediction result. Whether to do is controlled. For example, in order to predict the buffer remaining capacity immediately after quantization for a certain block, the edge integrated value of only that block is used.

  Since there is a correlation between the amount of the edge component included in each block and the code amount after JPEG compression, the predicted value of the code amount of the block to be compressed in the future is determined for each block depending on the size of the edge integrated value. Therefore, the buffer remaining amount prediction processing unit 112 subtracts the predicted value of the future code amount from the current buffer remaining capacity and adds the data transmission amount depending on the transmission rate of the data output from the output buffer memory to the future. This is the buffer remaining capacity prediction result. This prediction result is compared with a reference value set for the output buffer memory to determine the possibility of overflow. Here, the amount of data transmission is, for example, from the output buffer memory to the above transmission rate from the current time until the encoded image data corresponding to the block subjected to edge detection at this time is input to the output buffer memory. The total code amount of the encoded image data to be output can be set.

  By using the edge integration values of more blocks for prediction, overflow of the output buffer memory 60 can be predicted at an early stage. As a result, since the compression of the code amount is distributed over many blocks, local image quality degradation can be suppressed.

  Further, in predicting the remaining buffer capacity, the prediction accuracy of overflow is improved by obtaining a plurality of prediction results at different future times (for example, one or more times). For example, after the encoding of one block, the remaining buffer capacity is predicted in order from the elapse of the maximum number of blocks used for prediction, and the minimum value of the predicted remaining buffer capacity is obtained, so that there is more possibility of overflow. Can be predicted.

  In addition, the predicted value of the code amount of the block to be compressed in the future uses, specifically, the horizontal edge integrated value that is the output of the horizontal edge integrating buffer and the vertical edge integrated value that is the output of the vertical edge integrating buffer. It is obtained from the integrated edge value obtained by the predetermined calculation.

  Next, coefficient control of the quantization table based on the buffer remaining capacity prediction result will be described.

  As a result of the prediction of the remaining buffer capacity, when it is predicted that an overflow of the output buffer memory 60 will occur, the quantization processing unit 30a avoids the overflow by changing the coefficient of the quantization table and compressing the code amount.

  Here, the change of the coefficient is applied to the position (change start position) where the change of the quantization coefficient is started on the quantization table used for the quantization of the DCT data and the subsequent quantization coefficients in the quantization table. Change amount (quantization coefficient increase rate). The change start position and the increase rate of the quantization coefficient are determined according to the difference between the predicted value of the remaining buffer capacity and the reference value.

  Further, the change start position and the change amount can be set independently for the quantization coefficients arranged in the horizontal direction and the vertical direction on the 8 × 8 matrix quantization table, for example, respectively, the integrated value of the edge component in the horizontal direction, It can also be obtained from the integrated value of the edge components in the vertical direction.

  For example, the independent quantization coefficient setting for the quantization coefficients arranged in the horizontal direction is performed for the 64th quantization coefficient arranged in a matrix of 8 rows × 8 columns on the quantization table in the Kth column ( (1 column, 1, 2, 3..., 8 columns) and subsequent columns) are reduced by a predetermined ratio, and independent quantizers for the quantized coefficients arranged in the vertical direction. The quantization coefficient is set to any one of the Lth rows (1, 2, 3,..., 8 rows) with respect to 64 quantization coefficients arranged in a matrix of 8 rows × 8 columns on the quantization table. That is, the quantization coefficient in the subsequent rows is decreased by a predetermined rate. Here, the quantization coefficient located at the intersection of the first row and the first column is a coefficient for quantizing the DC component, and the quantization coefficient located at the intersection of the eighth row and the eighth column is horizontal. It is a coefficient for quantizing the AC component having the maximum frequency in both the direction and the vertical direction.

  As described above, in the image data encoding apparatus 100 according to the first embodiment, the edge component detection processing unit 110 detects the horizontal edge component of the input image and the vertical edge component of the input image, and then detects the horizontal edge. The integration of the detection results in the integration buffer 111 and the vertical edge integration buffer 113 is performed for each 8 × 8 pixel block of the JPEG processing unit, and the future output buffer memory remaining is determined from the edge integration result and the remaining capacity of the JPEG output buffer memory 60. The capacity is predicted, and the quantization processing unit 30a uses the quantization table in which the quantization coefficient is changed based on the prediction result and the edge integration result, thereby reducing the code amount of the encoded image data. Compression processing can be performed while avoiding the compression processing from being concentrated on a specific block. The overflow of the output buffer memory to pay can be avoided while suppressing image quality degradation.

  That is, in the conventional image data encoding device 200, when the image data is variable-length encoded in the JPEG format, the code length of the output image encoded data is compared with the transmission capability on the output side of the image data encoding device. Is large, the code length of the output encoded image data is compressed by changing the quantization coefficient only by the remaining capacity of the output buffer memory. Therefore, there is a problem that the code amount is not compressed with respect to a block that causes a significant decrease in the remaining buffer capacity, and the compression processing is concentrated on the subsequent blocks, and the image quality is greatly degraded locally.

  On the other hand, the image data encoding apparatus 100 according to the first embodiment of the present invention predicts the future buffer remaining capacity from the edge component of the input image and the current buffer remaining capacity, and calculates the predicted value of the buffer remaining capacity, Only when it is predicted that an overflow will occur as a result of comparison with the reference value, the code amount is compressed by changing the coefficient of the quantization table according to the difference between the predicted value of the remaining buffer capacity and the reference value. Do. For this reason, it is possible to compress the code amount only when a future overflow is predicted for a block that causes a decrease in the remaining buffer capacity. Conversely, if it is predicted that overflow will not occur as a result of comparison between the predicted value of the remaining buffer capacity and the reference value, unnecessary code amount compression can be avoided and image quality can be prevented from being impaired. be able to.

  Furthermore, by detecting the edge components included in a block separately in the vertical and horizontal directions, the quantization table coefficients for the prediction of overflow occurrence can be calculated in the vertical and horizontal directions according to the deviation of the edge direction of the block. The change can be made separately, and there is an effect that the code amount can be compressed while suppressing the deterioration of the image quality.

  In the first embodiment, as the image data encoding device, the detection and integration of the edge components are separately performed in the vertical direction and the horizontal direction of the image. However, the integration of the edge components is performed in the vertical direction. The component and the edge component in the horizontal direction may be integrated together.

  In this case, each quantization coefficient in the quantization table is set by, for example, setting the change start position of the quantization coefficient in the order of zigzag scanning performed at the time of variable length encoding processing, and further after this position. It is conceivable to set a change amount that changes the quantization coefficient.

  In addition, when the signal processing apparatus that implements the present invention has a signal processing unit (for example, an edge enhancement processing unit) that performs edge detection in addition to the image data encoding processing unit, the encoding processing unit The edge components of the image may be integrated using edge components detected by a signal processing unit such as the edge enhancement processing unit. In this case, it is not necessary to provide an edge component detection processing unit in the image data encoding device constituting the encoding processing unit.

  Further, although not particularly described in the first embodiment, the image data encoding device of the first embodiment can be favorably applied to an electronic information device such as a digital camera or a camera-equipped mobile phone device.

(Embodiment 2)
Hereinafter, such an electronic information device will be described as Embodiment 2 of the present invention.

  The electronic information device according to the second embodiment includes a data processing unit that processes encoded image data output from the image data encoding device 100 according to the first embodiment.

  The data processing unit stores the image encoded data in a storage unit, a transmission process for transmitting the image encoded data, and displays the image data on the display unit based on the image encoded data. At least one of display processes is performed.

  In the electronic information device having such a configuration, the encoded image data from the image data encoding device is stored in the storage unit, is transmitted properly, or is displayed on the display screen of the display unit. can do.

  As mentioned above, although this invention has been illustrated using preferable embodiment of this invention, this invention should not be limited and limited to this embodiment. It is understood that the scope of the present invention should be construed only by the claims. It is understood that those skilled in the art can implement an equivalent range based on the description of the present invention and the common general technical knowledge from the description of specific preferred embodiments of the present invention. Patents, patent applications, and documents cited herein should be incorporated by reference in their entirety, as if the contents themselves were specifically described herein. Understood.

  The present invention relates to an image data encoding device, an image data encoding method, an image data code amount control method, and an electronic information device, and the compression process specifies a compression process for reducing the code amount of image encoded data. Image data encoding apparatus and image data encoding that can be performed while avoiding concentrating on blocks, and thereby avoiding an overflow of an output buffer memory for storing image encoded data while suppressing deterioration in image quality To provide a method, an image data code amount control method, and an electronic information device that stores encoded data from the image data encoding device in a storage unit, performs transmission processing, or performs display processing on a display unit Can do.

DESCRIPTION OF SYMBOLS 10 JPEG input buffer memory 20 DCT process part 30a Quantization process part 50 Variable length encoding process part 60 JPEG output buffer memory 100 Image data encoding apparatus 100a Compression encoding part 110 Edge detection process part 111 Horizontal edge accumulation | storage buffer memory 112 Buffer Remaining amount prediction processing unit 113 Vertical edge integration buffer memory 114 Quantization coefficient calculation processing unit

Claims (19)

An image data encoding device that compresses and encodes image data and outputs image encoded data,
A compression encoding unit for compressing and encoding the image data;
A temporary storage unit that temporarily stores image encoded data obtained by compression encoding of the image data;
A remaining storage capacity prediction unit that predicts a remaining storage capacity in the temporary storage unit based on the image data;
The compression encoding unit controls the compression encoding process of the image data based on the predicted storage remaining capacity so that the image encoded data obtained by the compression encoding does not overflow in the temporary storage unit. Image data encoding device. The image data encoding device according to claim 1,
An edge component detection processing unit for detecting an edge component from the image data;
An edge component integration unit that integrates the edge component detected by the edge component detection processing unit in a processing unit for encoding the image data;
The said memory remaining capacity prediction part is an image data encoding apparatus which estimates the memory remaining capacity in the said temporary memory part based on the integration value of the edge component integrated | accumulated by this edge component integration part. The image data encoding device according to claim 2,
The storage remaining capacity prediction unit predicts the memory remaining capacity of the temporary storage unit from the memory remaining capacity of the temporary storage unit and the integrated value obtained by integrating the edge components in the edge component integration unit. Image data encoding device. The image data encoding device according to claim 2,
The edge component detection processing unit
A horizontal edge component detection filter for detecting a horizontal edge component from the image data;
An image data encoding device comprising: a vertical edge component detection filter that detects an edge component in a vertical direction from the image data. The image data encoding device according to claim 4, wherein
A horizontal edge integration buffer that integrates horizontal edge components detected by the horizontal edge component detection filter for each block that is a processing unit of encoding the image data;
An image data encoding apparatus, comprising: a vertical edge integration buffer that integrates the vertical edge component detected by the vertical edge component detection filter for each block that is a processing unit for encoding the image data. The image data encoding device according to claim 5, wherein
The storage remaining amount prediction unit includes an integrated value for each block of horizontal edge components obtained by the horizontal edge integrating buffer, an integrated value for each block of vertical edge components obtained by the vertical edge integrating buffer, and the temporary value An image data encoding device, which is a remaining buffer capacity prediction processing unit that predicts a remaining memory capacity of a temporary storage unit from a remaining memory capacity of a storage unit. The image data encoding device according to claim 6, wherein
The storage remaining capacity prediction unit is an image data encoding device that controls a future time point to be a remaining capacity prediction target according to the number used for prediction among edge integration values for each block. The image data encoding device according to claim 7,
The storage remaining capacity prediction unit subtracts a predicted value of a future code amount obtained from the edge integrated value from the current memory remaining capacity of the temporary storage unit, and stores the encoded image data output from the temporary storage unit. An image data encoding apparatus that obtains a future buffer remaining capacity prediction result by adding a data transmission amount depending on a transmission rate. The image data encoding device according to claim 8,
The storage remaining capacity prediction unit is an image data encoding device that performs buffer remaining capacity prediction for one or more future time points, and obtains an overflow prediction result from one or more remaining capacity prediction results. The image data encoding device according to claim 9, wherein
The predicted value of the code amount of the block to be compressed in the future is the edge integration obtained by calculation using the horizontal edge integration value that is the output of the horizontal edge integration buffer and the vertical edge integration value that is the output of the vertical edge integration buffer. An image data encoding device obtained from a value. The image data encoding device according to claim 1,
The compression processing unit
An orthogonal transform processing unit that transforms the image data into frequency components for each block of a predetermined size;
A quantization processing unit that quantizes the frequency component corresponding to each block;
An image data encoding apparatus comprising: a variable length encoding unit that performs variable length encoding on the quantized frequency component and outputs image encoded data. The image data encoding device according to claim 11, wherein
An image data encoding apparatus comprising: a quantization coefficient calculation processing unit that obtains a quantization coefficient used in the quantization processing unit from the predicted storage remaining capacity. The image data encoding device according to claim 12,
The quantization calculation processing unit uses a quantization coefficient used in the quantization processing unit from a horizontal edge integrated value and a vertical edge integrated value for each block, an AC component quantization coefficient in a horizontal axis direction, and a vertical axis An image data encoding device that is obtained separately for quantization coefficients of alternating current components in a direction. The image data encoding device according to claim 12,
The orthogonal transform processing unit is a DCT processing unit that is a DCT processing unit that performs a discrete cosine transform process on the image data for each block to obtain a DCT coefficient corresponding to each block. The image data encoding device according to claim 14,
The said quantization process part is an image data encoding apparatus which quantizes the DCT coefficient corresponding to each block using the quantization coefficient calculated | required by the said quantization coefficient calculation process part. The image data encoding device according to claim 15,
The variable length coding unit is an image data coding device that converts the quantized data into the image coded data by Huffman coding. An image data encoding method for compressing and encoding image data and outputting the image encoded data,
A compression encoding step for compression encoding the image data;
A temporary storage step of temporarily storing image encoded data obtained by compression encoding of the image data in a temporary storage unit;
A storage remaining capacity prediction step of predicting a storage remaining capacity in the temporary storage unit based on the image data,
In the compression encoding step, the compression encoding process of the image data is controlled based on the predicted remaining storage capacity so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit. Image data encoding method. An image data code amount control method for controlling the code amount of image encoded data stored in a temporary storage unit when performing image data encoding processing,
An edge component integration step of integrating edge components of the image data for each block which is a unit of encoding;
Based on an integrated value obtained by integrating the edge components, a storage remaining capacity prediction step for predicting a storage remaining capacity in the temporary storage unit, and
An image data code amount control method that adjusts the encoding process of the image data based on the predicted remaining storage capacity so that the encoded image data obtained by the compression encoding does not overflow in the temporary storage unit. An electronic information device having a data processing unit for processing image encoded data from the image data encoding device according to claim 1,
The data processing unit stores the image encoded data in a storage unit, a transmission process for transmitting the image encoded data, and displays the image data on the display unit based on the image encoded data. An electronic information device that performs at least one of display processing.

Images