JP3480644B2 - Image encoding device and image decoding device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image coding apparatus and an image decoding apparatus, and more particularly to lossy predictive coding for a multivalued input image.

[0002]

2. Description of the Related Art Since an image generally contains a very large amount of data, it is generally compressed by encoding during storage or transmission. Image data encoding methods are roughly classified into two types: lossless encoding methods and lossy encoding methods.

In the case of lossless encoding, the image and the code have a one-to-one correspondence. Therefore, the image quality during decoding does not deteriorate,
It is difficult to control the code amount. For example, according to Shannon's information source coding theorem, the average code length cannot be made smaller than entropy (see, for example, "Introduction to Document Data Compression Algorithm" (Tomohiko Uematsu, CQ Publishing Co., p.32-33)). Therefore, when the code amount is limited due to some constraint, it is necessary to reduce the information amount itself of the image by distorting the image. This is lossy coding.

Next, predictive coding will be described from among reversible coding. Predictive coding is a technique of predicting a pixel of interest from the situation of processed peripheral pixels and using a prediction error as a code when the prediction is incorrect. In predictive coding, the size and distribution of prediction error influences the code amount. Therefore, for example, the code amount is small in a noise-free image drawn in PDL (Page Description Language), and conversely, the code amount tends to be large in an image with noise such as a scan-in image.

In the following, the inventors of the present invention filed Japanese Patent Application No. 8-31074.
The predictive coding proposed in No. 1 will be described as an example of predictive coding.

FIG. 17 is a block diagram showing a predictive coding apparatus according to this proposal. In FIG. 17, the predictive encoding device includes an image input unit 10, a first predictor 20, and a second predictor 2.
1, a prediction error calculation unit 30, a prediction error addition unit 31, a selection unit 40, an encoding unit 50, and a code output unit 60. The image input unit 10 receives an input image from the outside, and as the image data 100, the first prediction unit 20 and the second prediction unit 21.
And the prediction error calculation unit 30. First predictor 2
0 and the second predicting unit 21 predict the pixel value of the target pixel based on the image data 100 by a predetermined method, and send it to the selecting unit 40 as predicted value data 110. The prediction error calculation unit 30 predicts the pixel value of the target pixel based on the image data 100 by a predetermined method, calculates the difference between the predicted value and the actual pixel value of the target pixel, and selects the prediction error data 120 as the selection unit. 40 to 40. The selection unit 40 uses the image data 100.
And the prediction match / mismatch in the target pixel is detected from the prediction value data 110. As a result, if there is a prediction unit whose prediction is correct, the identification number of the prediction unit is converted. If none of the prediction units is correct, the prediction error data 120 is converted into the prediction situation data 130 and is sent to the encoding unit 50. The encoding unit 50 encodes the prediction situation data 130 using a predetermined encoding method,
The code data 140 is sent to the code output unit 60. The code output unit 60 outputs the code data 140 as an output code to the outside. In addition to the identification number, the rank of the predictive predictive ratio may be used as the identification information.

FIG. 18 is a flowchart showing the coding processing operation of the predictive coding apparatus shown in FIG. In FIG. 18, in step S1O, the image input unit 10 inputs an image to obtain image data 100. Step S20
Then, the first prediction unit 20 and the second prediction unit 21 predict the pixel value of the pixel of interest based on the image data 100, and at the same time, the prediction error calculation unit 30 calculates the prediction error of the prediction value by a predetermined method.

Further reference is made to FIG. 19 for step S30. In FIG. 19, in step S31, the selection unit 40 determines whether the prediction value of the first prediction unit and the pixel value of the target pixel match, and if they match, step S34.
If they do not match, the process proceeds to step S32. Similarly, in step S32, if the predicted value of the second predictor and the target pixel value match, the process proceeds to step S34, and if they do not match, the process proceeds to step S33. In step S33, the selection unit 40 converts the prediction error data 120 into the prediction situation data 130 and sends it to the encoding unit 50. In step S34, the selection unit 40 converts the identification number of the prediction unit whose prediction rule hits into the prediction situation data 130 and sends it to the encoding unit 50. Above, step S30 is completed.

In FIG. 18, in step S40, the encoding condition 50 is encoded by the encoding unit 50 by a predetermined encoding method to obtain encoded data 140. Step S
In 50, the code output unit 60 outputs the code data 140 to the outside. In step S60, it is determined whether or not all input image data has been processed, and if all have been processed, the encoding process ends. If unprocessed data remains, the process returns to step S10.

Next, a predictive code decoding apparatus for decoding the code coded by the predictive coding apparatus of FIG. 17 will be described. FIG. 20 shows this predictive code decoding apparatus. In this figure, the predictive code decoding apparatus includes a code input unit 70 and a decoding unit 5.
1, a selection unit 41, a first prediction unit 20, a second prediction unit 21, a prediction error addition unit 31, and an image output unit 80. Here, the same reference numerals are given to the same units as those of the encoding device, and detailed description thereof will be omitted.

The code input section 70 receives an input code from the outside and sends it to the decoding section 51 as code data 140.
The decoding unit 51 decodes the coded data 140 by a decoding method corresponding to the coding method used by the coding unit 50, and sends it to the selection unit 41 as prediction situation data 130. The selection unit 41 is
If the prediction situation data 130 indicates the identification number of the prediction unit, the control data 111 is sent to the corresponding prediction unit, and the prediction unit outputs the image data 100. If the content of the prediction situation data 130 is a prediction error, the prediction error data 1
It is sent to the prediction error addition unit 31 as 20. The prediction error addition unit 31 predicts the pixel value of the target pixel by the same method as the prediction error calculation unit 30, and the prediction value and the prediction error data 120
Then, the pixel values are restored and the image data 100 is sent to the image output unit 80. The image output unit 80 is the image data 100.
Is output to the outside as a decoded image.

Next, the decoding process will be described with reference to FIG. In step S110, the code input unit 70 inputs a code to obtain code data 140. In step S120, the decoding unit 51 uses the decoding method corresponding to the encoding method used in the encoding unit 50 to generate the encoded data 1
The prediction situation data 130 is obtained from 40.

Referring now to FIG. 22, step S130
Will be described in detail. 22, in step S131, the selection unit 41 determines whether or not the prediction situation data 130 matches the identification number of the first prediction unit. If they match, the process proceeds to step S134, and if they do not match, the step S132.
Go to. Similarly, in step S132, if the prediction situation data 130 matches the identification number of the second predictor, the process proceeds to step S134, and if they do not match, the process proceeds to step S133. In step S133, the selection unit 41 converts the prediction situation data 130 into prediction error data 120 and sends it to the prediction error addition unit 31. In step S134, the selection unit 40 sends the control data 111 to the prediction unit having the identification number that matches the prediction situation data 130.

In FIG. 21, in step S21, when there is a prediction unit that has received the control data 111, the predicted value of the pixel of interest predicted by the corresponding prediction unit is output as the image data 100. If not, the prediction error addition unit 31 restores the pixel value from the prediction value and the prediction error data 120 by the same method as the prediction error calculation unit 30,
The image data 100 is output. In step S140, the image output unit 80 outputs the image data 100 to the outside. In step S150, it is determined whether all input code data have been processed, and if all have been processed, the decoding process ends. If unprocessed data remains, the process returns to step S110.

According to the above proposal, it becomes possible to encode an image having a high predictive predictive ratio such as a computer-generated image at a high compression rate.

However, in the encoding / decoding device having the above-mentioned configuration, it is difficult to control the code amount because it is lossless encoding.
Since the content of the code is composed of a predictor or a prediction error in which the prediction has hit, the hit of the prediction is a condition for reducing the code amount. Therefore, especially when an image with noise such as a scan-in image is targeted, the code amount increases.

[0017]

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and is a lossy coding apparatus and decoding capable of controlling the code amount by using predictive coding which is a type of lossless coding. The purpose is to provide a device.

[0018]

According to the present invention, in order to achieve the above object, an image input device for inputting image data to an image encoding device, and an image input from the image input device, At least one pixel value prediction unit that predicts the pixel value of the pixel of interest, and prediction for calculating an error between the pixel value of the pixel of interest and the predicted value predicted by a predetermined prediction method for the image input from the image input unit For each of the error calculating means and the pixel value predicting means, a threshold processing means for determining that the prediction is correct when the difference between the predicted value of the pixel value predicting means and the actual pixel value is less than or equal to a predetermined threshold value. If the threshold value processing means determines that one of the pixel value predicting means has hit, the identification information of the pixel value predicting unit is used, and if not, the prediction information calculated by the prediction error calculating means is used. A selection means for selecting an error, a coding means for coding the data selected by the selection means by a predetermined method, and a code output means for outputting the code created by the coding means are provided. .

In this configuration, since the prediction result that falls within the prediction error equal to or less than the dark value is regarded as predictive, it is possible to code an image with a code amount smaller than the lossless coding.

Further, in this configuration, the image coding apparatus is further provided with threshold value determining means for determining the threshold value used in the threshold value processing means, and the threshold value determining means determines the peripheral pixel value in the input image or the decoded image. The threshold value may be generated depending on the situation. In this way, lossy encoding suitable for vision can be performed.

Further, the image coding apparatus is further provided with threshold value determining means for determining the threshold value used in the threshold value processing means, and the threshold value determining means sets the threshold value in accordance with the situation of the peripheral pixel value of the input image. The threshold may be generated and added to the code. By doing so, it is possible to send the information for performing the lossy encoding suitable for the visual sense and further for performing the more appropriate image processing to the decoding side.

Further, according to the present invention, in order to achieve the above object, an image input device for inputting an image to an image encoding device, and a pixel of a pixel of interest for the image input from the image input device A plurality of pixel value predicting means for predicting values,
Prediction error calculation means for calculating an error between the pixel value of the pixel of interest and the prediction value predicted by a predetermined prediction method with respect to the image input from the image input means, and the prediction error calculated by the prediction error calculation means With a predetermined method, and if the predicted value of the pixel value prediction means and the actual pixel value match, the identification information of the pixel value prediction unit, otherwise, the pre-quantization Selecting means for selecting the prediction error quantized by the means, encoding means for encoding the information selected by the selecting means by a predetermined method, and code output means for outputting the code created by the encoding means And is set.

In this configuration, since the prediction error is quantized, it is possible to code with a smaller code amount than the lossless coding.

Also in this configuration, the image coding apparatus is further provided with a quantized value deciding means for deciding a quantized value used in the quantized means, and the quantized value produced by the quantized value deciding means is further provided. The value may be generated according to the situation of the peripheral pixel value of the input image or the decoded image.

Further, the predictive coding apparatus is further provided with a quantized value deciding means for deciding a quantized value used in the quantizing means, and the quantizing value deciding means has a condition of a peripheral pixel value of the input image. The quantized value may be generated according to the above, and the quantized value may be added to the code.

Further, according to the present invention, in order to achieve the above object, code input means for inputting the above code to an image decoding apparatus for decoding the code encoded by the image encoding apparatus as described above. A pixel of interest based on a decoding unit that decodes the code input from the code input unit by a predetermined method, at least one pixel prediction unit that predicts the pixel value of interest, and a prediction error and a prediction value obtained by the predetermined method. A prediction error adding means for calculating a value and a selecting means for selecting either the pixel value predicting means or the prediction error adding means based on the information obtained by the decoding means are provided, and the pixel value is selected by the selecting means. When any one of the prediction means is selected, the target pixel value calculated by the selected pixel value prediction means is output, and when the prediction error addition means is selected by the selection means. , And outputs a target pixel value calculated by the prediction error addition means.

With this configuration, the code encoded by the above encoding device can be decoded.

In this structure, the image decoding apparatus further performs image processing on the pixel value of interest selected and output by the selecting means, and then performs image processing for clipping in a predetermined pixel value section. Means and image output means for outputting the image processed by the image processing means may be provided. With this configuration, the decoding device performs the image processing accompanied by clipping, so that the image quality of the decoded image can be improved.

Further, in this case, the image decoding apparatus is further provided with a deciding means for deciding the pixel value section used in the image processing means, and the deciding means has a threshold value or a quantized value in the peripheral pixel value of the decoded image. Alternatively, it may be determined from the auxiliary information generated in the decoded image along with the decoding. In this way, the pixel value section to be clipped in the image processing is obtained from the peripheral pixel values of the decoded image or the side information, so that it is possible to perform appropriate image processing on the decoded image.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, two examples of an embodiment of the present invention will be described, that is, a first embodiment in which an error is allowed depending on the presence or absence of a prediction hit, and a second embodiment in which a prediction error is quantized. [Embodiment 1] FIG. 1 is a block diagram showing an image coding apparatus according to Embodiment 1 of the present invention. In FIG. 1, the image coding apparatus includes an image input unit 10, a first prediction unit 20, and a second prediction unit 2.
1, a prediction error calculation unit 30, a selection unit 40, an encoding unit 50,
It comprises a code output unit 60 and a threshold processing unit 90.
Here, the portions corresponding to those in FIG. 17 are designated by the corresponding reference numerals, and detailed description thereof will be omitted.

If the difference between the predicted value data 110 and the image data 100 is less than or equal to a predetermined threshold value, the threshold value processing unit 90 performs threshold value processing for assuming that the prediction is correct even if the difference is not zero. Then, the result is predicted predictive data 150
To the selection unit 40. Note that the selection unit 40 generates a decoded signal decoded by the image decoding device (see FIG. 2) from the prediction error calculation unit 30 and the threshold processing unit 90 to generate the first prediction unit 20, the second prediction unit 21, and the prediction error calculation. It is supplied to the part 30. As a result, prediction can be performed with the same peripheral pixel value in the image encoding device and the image decoding device.

FIG. 2 is a flowchart showing the encoding process of the image encoding apparatus of the first embodiment. The encoding process of FIG. 2 is the same as the encoding process of FIG. 18 except for the selection of the prediction unit in step S30 ′. In FIG. 2, parts corresponding to those in FIG. 18 are designated by corresponding reference numerals and description thereof will be omitted.

Next, step S30 'will be described with reference to FIG. In FIG. 3, in step S35, the threshold processing unit 90 determines whether or not the difference between the prediction value of the first prediction unit 20 and the pixel value data 100 is less than or equal to a predetermined threshold value. If less than the threshold value, the process proceeds to step S34. If not, the process proceeds to step S36. In step S36, the threshold value processing section 90 determines whether or not the difference between the predicted value of the second prediction section 21 and the pixel value data 100 is less than or equal to a threshold value. If it is less than or equal to the threshold value, the process proceeds to step S34. Go to.

Next, the image decoding apparatus of the first embodiment will be described. FIG. 4 shows the configuration of this image coding apparatus, and FIGS. 5 and 6 show the decoding processing of the image decoding apparatus. The configuration and operation of this image decoding device are basically the same as those of the background image decoding device described in FIGS. 20 to 22 except that the decoded image data 101 does not match the image data 100. The description will not be repeated.

Next, the effect of the first embodiment will be described. When noise is added to a portion having the same pixel value, for example, a blank portion of the text image, the pixel value slightly changes depending on the pixel. An example of such a pixel value change is shown in FIG. This pixel value change is essentially meaningless information, but since the entropy is not 0, lossless coding causes an increase in code amount. Therefore, in the first embodiment, it is determined that the prediction is correct when the difference between the predicted value and the actual pixel value is less than or equal to the threshold value.

FIG. 8 shows an example of comparison between predictions when the threshold value is 3 and when reversible. It can be seen that the threshold value is improved about four times when the threshold value processing is performed. In the case of an 8-bit image, there is an average difference of about 8 bits between the predictor code and the prediction error code. In this case, the average code length is about 0.6 × 8 = about 4.
It can be reduced by about 8 [bit].

In this embodiment, the code amount can be reduced by increasing the threshold value, but on the other hand, the image quality may be deteriorated. Therefore, for example, the decoding side may perform image processing for improving the image quality. At this time, if the threshold on the encoding side is known on the decoding side, the original pixel value is within the threshold,
Image processing that does not exceed the range is preferable. Also,
In order to suppress the image quality deterioration more positively, it is possible to make the threshold variable according to the property of the image.

Next, the above modification will be specifically described. 9 shows an image coding apparatus, and FIG. 10 shows an image decoding apparatus. In FIGS. 9 and 10, parts corresponding to those in FIGS. 1 and 4 are denoted by corresponding reference numerals, and detailed description thereof will be omitted.

In FIG. 9, 91 is a threshold value determining unit, 92 is an image processing unit, 102 is processed image data, and 160 is threshold value data. The threshold value determination unit 91 analyzes the decoded image data 101, calculates threshold value data 160 from the result, and sends it to the threshold value processing unit 90. The image processing unit 92 uses the threshold data 16
The decoded image data 101 is subjected to image processing such as an average value filter using 0, and the result is sent to the image output unit 80 as processed image data 102.

Since the operation of the image decoding apparatus in FIG. 9 is almost the same as the operation of the image decoding apparatus in FIG. 1, only different points will be described. At the time of encoding, the threshold value determination unit 91 determines the threshold value before step S35 in FIG. At the time of decoding, the threshold value determination unit 91 determines the threshold value before step S140 in FIG. 2, and the image processing unit 92 performs image processing for improving image quality.

In the above description, the threshold value determined by the threshold value determining section 91 depends on the peripheral pixel values in the decoded image.
Noise is hard to be perceived in a portion where the pixel value changes drastically as a visual characteristic. Therefore, by utilizing this, the threshold Th is raised or lowered according to, for example, the dynamic range D of the peripheral pixels. In this case, the following formula can be considered as an example of the calculation formula. Th = D × a (1) where a is a positive constant. In addition to this, it may be calculated from the distribution or pattern of the peripheral pixels, or a more complicated expression than a linear expression such as the expression (1) may be used.

An example in which the threshold value is variable will be described with reference to FIG. As an example, A, B shown in FIG.
Consider a case in which a predictor that has hit from four predictors that predict values at four positions of C and D (x is a pixel position of interest) is encoded. As an image, an edge image having a sharp change in pixel value and a sweep image having a gentle change in pixel value are taken. When a = 1/7 in the equation (1), Th = (244-0) × (1/7) = 35 in the edge image of FIG. 11 (b) (2) In the sweep image, Th = (46−10) × (1/7) = 5 (3), and hence the encoding using these two threshold values is assumed below.

First, an example in which this embodiment is applied to an edge image (shown in (b)) using (b), (c), and (d) will be described. (C) and (d) each have a threshold value of 3
Pixels for which there is no predictor hit in the decoded image according to the present embodiment when the number is 5, 5 are indicated by hatching. Since the dynamic range is large in terms of image quality, neither is visually deteriorated. In addition, comparing the approximate code amounts, log ₂ 4 = 2 [bit] is required to indicate the encoder, and 8 [bit] is required to indicate the prediction error, and 50 [bi] in (c).
It becomes 80 [bit] in t] and (d). Since the difference in image quality deterioration is small, the threshold value of 35 is preferable to 5 in consideration of the code amount.

Next, an example in which the present embodiment is applied to a sweep image (shown in (e)) using (e), (f) and (g) will be described. As before, encoding was performed using two types of thresholds. The code amount is approximately 44 [bi in (f).
t] and (g) are 74 [bit]. In this case (f)
Then, 15 pixels out of 16 pixels have the same pixel value, and the image quality is severely deteriorated. Therefore, although the code amount increases, it is desirable to set the threshold value to 5.

From the above example, the effectiveness of the process of changing the threshold value is clear. If it is not desired to deteriorate the image quality, a in Formula (1) may be reduced, and conversely, a may be increased to reduce the code amount. Note that (1)
If a is set to 0 in the equation, lossless encoding that matches the decoded image can be realized.

Next, the image processing of the image processing unit 92 will be described. The image processing in the image processing unit 92 is performed with reference to the threshold data 160. Code image data 10
The pixel value of 1 generally does not match the pixel value of the image data 100, but the original pixel value is (pixel value)-(threshold value) ~
It is always included in the interval of (pixel value) + (threshold value). Therefore, it can be said that it is natural that the pixel value is included in this section even in the processed image data 102 subjected to the image processing. Therefore, in the image processing here, the processed pixel value is clipped using the above section. This example is shown in Figure 1.
2 shows. In this example, the clipping effect is remarkable in the pixels 4 and 7. It should be noted that the image processing mentioned here may be any processing that is generally used for improving image quality, such as an average value filter and an intermediate value filter.

Although an example of determining the threshold value from the decoded image has been described here, the threshold value may be determined from the image data 100 on the encoding side. When the decoding side does not perform image processing, the decoder can be realized by the configuration of FIG. When performing image processing, it is possible to send the threshold value data 160 to the decoding side as side information. FIG. 13 shows an example of a code format when the threshold value is changed for each image block. [Second Embodiment] FIG. 14 is a block diagram showing an image coding apparatus according to the second embodiment of the present invention. In the figure, the same parts as those in FIG. 93 is a quantizer,
Reference numeral 121 is quantized prediction error data. The quantization unit 92 quantizes the prediction error data 120 by a predetermined method, and sends it to the selection unit 40 as quantized prediction error data 121. The image decoding apparatus is similar to that of the first embodiment, so refer to FIG. 4 and its description.

The operation of the second embodiment is almost the same as that of the first embodiment. The different parts will be described below with reference to FIG. In step S37, the quantizing unit 93 quantizes the prediction error data 120 by a predetermined method. Quantizer 9
The quantization in 3 is performed so that the code amount of the code data 140 becomes small. For example, when the encoding unit 50 is adaptive encoding such as arithmetic encoding, the amount of code can be reduced by biasing the distribution of prediction errors. Therefore, the quantization process may be performed by quantizing a close prediction error to a specific representative value.
Further, when the encoding unit 50 is a fixed encoding such as Huffman encoding, only the code of the specific value is prepared from the beginning. At this time, the quantization process may round the prediction error to one of the specific values.

FIG. 16 shows an example of Huffman code using only specific values. However, the image is assumed to be an 8-bit image. The code length can be shortened by reducing the number of symbols to be encoded as shown in the figure. For example, as shown in FIG.
If collectively handled by one, the average code length log ₂ 5 = 2.3
A code amount reduction of about [bit] can be expected. In the case of adaptive coding, the entropy also decreases as the number of symbols decreases, so the same effect can be expected.

As a criterion for selecting the specific value, considering the above-mentioned visual characteristics, for example, it is conceivable that the error of around 0 is finely set, and conversely in the case of an 8-bit image, around ± 255 is roughly set.

In the present embodiment as well, similar to the first embodiment, the image quality can be improved by the image processing on the decoding side. Except that the threshold processing and the allowable error in the first embodiment correspond to the quantization processing and the quantization error in the present embodiment, the description is not repeated because they are substantially the same.

Since this embodiment and the first embodiment are not contradictory to each other, both embodiments may be carried out simultaneously.

[0053]

As is apparent from the above description, in the present invention, since the prediction value has a permissible error, the predictive coding can be made irreversible. Also, by quantizing the prediction error, lossy can be realized. Furthermore, the image quality of the decoded image can be improved by performing image processing according to the allowable error and the quantization error on the decoding side.

[Brief description of drawings]

FIG. 1 is a block diagram showing an image coding apparatus according to a first embodiment of the present invention.

FIG. 2 is a flowchart showing an operation of an encoding process in the image encoding device according to the first embodiment of the present invention.

FIG. 3 is a flowchart showing an operation of a predictor selection process in the image coding apparatus according to the first embodiment of the present invention.

FIG. 4 is a block diagram showing an image decoding apparatus according to the first embodiment of the present invention.

FIG. 5 is a flowchart showing an operation of a decoding process in the image decoding apparatus according to the first embodiment of the present invention.

FIG. 6 is a flowchart showing an operation of predictor selection processing in the image decoding device according to the first embodiment of the present invention.

FIG. 7 is a diagram showing an example of pixel value distribution of a scan image for explaining the first embodiment of the present invention.

FIG. 8 is a diagram showing an example of a prediction error distribution of a scan image for explaining the first embodiment of the present invention.

FIG. 9 is a block diagram showing a modified example of the image coding apparatus according to the first embodiment of the present invention.

FIG. 10 is a block diagram showing a modified example of the image coding apparatus according to the first embodiment of the present invention.

FIG. 11 is a diagram for explaining the modified example.

FIG. 12 is a diagram for explaining the modification example.

FIG. 13 is a diagram for explaining the modified example.

FIG. 14 is a block diagram showing an image encoding device according to a second embodiment of the present invention.

FIG. 15 is a flowchart showing an operation of a predictor selection process in the image encoding device according to the second embodiment of the present invention.

FIG. 16 is an explanatory diagram of an example of Huffman code that can be used in the image coding apparatus according to the second embodiment of the present invention.

[Fig. 17] Fig. 17 is a block diagram illustrating an example of an image encoding device as a background art.

FIG. 18 is a flowchart showing an operation of an encoding process in the image encoding device as the background art.

FIG. 19 is a flowchart showing the operation of a predictor selection process in the image coding device according to the background art.

[Fig. 20] Fig. 20 is a block diagram illustrating an example of an image decoding device according to the background art.

FIG. 21 is a flowchart showing an operation of a decoding process in the image decoding device according to the background art.

[Fig. 22] Fig. 22 is a flowchart showing an operation of a predictor selection process in the image decoding device according to the background art.

[Explanation of symbols]

10 Image input section 20 First Predictor 21 second predictor 30 Prediction error calculator 31 Prediction error adder 40, 41 Selector 50 encoding unit 51 Decryption unit 60 code output unit 70 code input section 80 Image output section 90 threshold processing unit 91 Threshold value determination unit 92 Image processing unit 93 Quantizer 100 image data 101 Decoded image data 102 Processed image data 110 predicted value data 111 Control data 120 prediction error data 121 Quantized prediction error data 130 Forecast situation data 140 coded data 150 predictive medium data 160 threshold data

Front page continued (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N ⁷ /24-7/68 H04N 1/41-1/419 G06T 1/00 G06T 9/00

Claims (9)

(57) [Claims] 1. An image input unit for inputting image data, at least one pixel value predicting unit for predicting a pixel value of a target pixel for an image input from the image input unit, and an image input from the image input unit. For each of the pixel value prediction means, a prediction error calculation means for calculating an error between the pixel value of the pixel of interest and the prediction value predicted by the predetermined prediction method, and the predicted value of the pixel value prediction means and the actual If the difference between the pixel value is less than or equal to a predetermined threshold value, the threshold value processing means determines that the prediction is correct, and if the threshold value processing means determines that one of the pixel value prediction means is correct. Is the identification information of the pixel value prediction unit, and if not, the selection means for selecting the prediction error calculated by the prediction error calculation means, and the data selected by the selection means are encoded by a predetermined method. The image encoding apparatus for encoding means, characterized in that it has a code output means for outputting the code created by the encoding means for reduction. 2. A threshold determining means for determining a threshold used in the threshold processing means, wherein the threshold determining means generates the threshold according to a situation of peripheral pixel values in the input image or the decoded image. The image encoding device described. 3. A threshold value determining means for determining a threshold value used in the threshold value processing means, wherein the threshold value determining means generates the threshold value according to a situation of peripheral pixel values of the input image,
The image coding apparatus according to claim 1, wherein the threshold value is added to the code. 4. An image input unit for inputting an image, at least one pixel value predicting unit for predicting a pixel value of a target pixel for the image input from the image input unit, and an image input from the image input unit. Prediction error calculating means for calculating an error between the pixel value of the pixel of interest and the prediction value predicted by a predetermined prediction method, and the prediction error calculated by the prediction error calculating means is quantized by a predetermined method. If the prediction value of the pixel value predicting unit and the actual pixel value match, the identification information of the pixel value predicting unit is used. If not, the prediction error quantized by the quantizing unit is used. An image characterized by having a selecting means for selecting, an encoding means for encoding the data selected by the selecting means by a predetermined method, and a code outputting means for outputting the code created by the encoding means. Sign Device. 5. A quantizing value determining means for determining a quantizing value used in the quantizing means, wherein the quantizing value generated by the quantizing value determining means is a peripheral pixel value of the input image or the decoded image. The image coding apparatus according to claim 4, wherein the image coding apparatus is generated according to the situation of. 6. A quantizing value determining means for determining a quantizing value used in the quantizing means, wherein the quantizing value determining means is generated according to a situation of peripheral pixel values of the input image. 5. The image coding apparatus according to claim 4, wherein the quantized value is added to the code. 7. An image decoding apparatus for decoding a code encoded by the image encoding apparatus according to claim 1 or 2, wherein a code input means for inputting the code and a code input from the code input means are predetermined. Decoding means for decoding by the method, at least one pixel predicting means for predicting the target pixel value, prediction error adding means for calculating the target pixel value from the prediction error and the predicted value obtained by the predetermined method, And a selecting means for selecting either the pixel value predicting means or the prediction error adding means based on the information obtained by the decoding means, and the selecting means selects one of the pixel value predicting means Output the pixel value of interest calculated by the selected pixel value prediction means, and when the prediction error addition means is selected by the selection means, the prediction error addition means An image decoding device, which outputs a pixel value of interest calculated by 8. An image processing means for performing image processing on the pixel value of interest selected and output by the selecting means, and then clipping the pixel value into a predetermined pixel value section, and processed by the image processing means. The image decoding apparatus according to claim 7, further comprising an image output unit that outputs an image. 9. A determining means for determining a pixel value section used in the image processing means, wherein the determining means generates a threshold value or a quantized value in a peripheral pixel value of the decoded image or in the decoded image according to the decoding. The image decoding device according to claim 8, wherein the image decoding device is determined from auxiliary information.