JPH05176180A - Coding method for still natural picture with mixed existence of binary picture - Google Patents

Abstract

PURPOSE:To realize coding by which a white/black gradation picture and a color picture including a character area are less deteriorated by coding the character area through the use of the binary picture coding system and coding the remaining picture with the natural picture coding system. CONSTITUTION:A character area discrimination circuit 9 analyzes a received picture S' and gives a command to a picture element binarizing circuit 2 so as to apply binarizing processing to a picture element block only including a character. A binary picture coder 5 codes a picture B received from the circuit 2. A coder 9 codes mean values Sh, S1 of the picture received from the circuit 2 for each block, and a natural picture coder 7 codes a picture D received from a differential picture generating circuit 4. A signal output circuit 8 receives data Bc, Shc, S1c, Dc coded from the coders 5-7 and outputs a signal in a specific format. Thus, it is possible to realize the coding causing less deterioration of a white/black gradation picture and a color picture including the character area.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a high-efficiency coding method for black and white gradation images or color images.

[0002]

2. Description of the Related Art Conventionally, various encoding methods for color still images (including gradation images) have been proposed. In particular, CCITT SG VIII and ISO IEC / JTC1 / SC29 are jointly examined. Two international standard methods called the JPEG method and JBIG method, which are being advanced, are drawing attention.

The JPEG (Joint Photographic Expert Group) system proposed as a color still image encoding system is
Although it is not an information storage type, it is a method that can encode a gradation image with high efficiency. Orthogonal cosine transform (DCT), which is one of the orthogonal transform methods that is generally efficient for color still images, is adopted in this JPEG method. DCT
Is a method in which an image is divided into blocks, discrete cosine transform is applied to each block, and the output coefficient is encoded. The output coefficient is quantized independently for each frequency component, and the number of quantization bits is assigned to each transform coefficient so that the coding efficiency is optimized. Unlike the case of encoding a binary still image, it is not practical to use a coder without information loss because of the amount of information contained in the color still image. Quantization at the level is required, and generally, high coding efficiency can be achieved by roughing the quantization in the high frequency region.

On the other hand, JBIG (Joint Bi-level Im), which has been proposed as an international standard draft as a binary image coding method.
The age Expert Group) method is an information-preserving coding method that can basically code a binary-represented image with high efficiency without loss of information.

The image targeted by the color facsimile is
It is assumed that the document is not a color image alone, but a document in which black and white characters and color images are mixed or a multi-color document in which color characters, black and white characters, and color images are mixed. However, JP
Coding methods that use DCT, such as the EG method, have the drawback that the image quality of the character area is significantly degraded for documents with mixed characters and images. Therefore, some methods have been proposed which solve the problems of the conventional method and can efficiently encode a document in which binary characters and gradation images are mixed or a binary color document.

[0006]

As described above, when the orthogonal transform coding method (DCT) according to the above-mentioned conventional technique is applied to a color image, noise is generated around the edges of the characters included in the image and the quality of the image is deteriorated. Has the drawback that it deteriorates. The reason for this is that the quantization in DCT is also performed for the edge of the character, so if there is a pixel that changes sharply like the edge of the character in the image, the pixel value between Since the change is large, it contains a lot of high frequency components,
This is because efficient quantization of this means coarse quantization, and as a result, noise around the character cannot be ignored.

The present invention solves the above-mentioned problems of the prior art, and for a monochrome gradation image, a color image or a mixed document of multi-color and natural color, which includes a region having a sharp edge like a character, It is an object of the present invention to provide a coding method of a binary natural image mixed still natural image capable of obtaining a coded image with little deterioration.

[0008]

A feature of the present invention is that a character area (aside from black and white characters, color (Including characters) are separated, the character area is encoded using a binary image encoding method, and the remaining image is encoded using a natural image encoding method. On the decoding side, the images encoded by each encoding method are individually decoded,
An original image is obtained by combining these.

The details are as follows. On the encoding side, the target original image is divided into blocks of pixels, and it is described below whether or not there is an area that is not suitable for the conventional natural image encoding method, that is, the character area, in units of the divided blocks. Then, the character area is binarized.

The characteristics of the character area are detected as follows by the character area detection method (determination method) and extraction processing. The following methods may be used alone or in combination. (1) As a method based on the maximum and minimum values, in the character area, the difference between the values of adjacent pixels is larger than that in the natural image. Utilizing this feature, the maximum value and the minimum value of the pixel of the pixel block are checked, and it is determined whether the pixel block is a character area or not based on the difference. That is, for example, the maximum pixel value in the block is S _max ,
The smallest value as the S _{mi n,} F ≧ the threshold T which is the difference F
If it is T, it is a method of forming a character area including an edge. (2) As a method based on the distribution of pixel values, in general, the background of a region where characters are present is often a simple color scheme (or gradation), and the values of the pixels forming the character part and its background are
There is a tendency to concentrate on a specific value. Utilizing this, the distribution of the pixel values of the pixel block is examined, and if there is one or two peaks of the distribution in which the pixel values are concentrated, this pixel block is determined as the character area.

That is, for example, in order to investigate the distribution of pixel values that each pixel of the block can have, a histogram of the number H (x) of pixels in the block having a specific pixel value x is created as shown in FIG. .. This histogram is used to find the pixel value x where H (x) −H (x-1)> M and H (x) −H (x + 1)> M using a certain threshold (parameter) M, and It is regarded as the peak of the pixel value distribution. If one or two peak values x exist, the pixel block is determined to be a character area, and otherwise it is determined to be a non-character area.

The determination and extraction processing of the character area is intended to reduce the number of coding bits of the binary image generated by the binary processing described later, and the improvement of the image quality by the encoding is performed by this processing. Can be achieved without doing.

The binarization processing of the character area will be described below.
First, set each pixel value to 2 according to the density of each pixel in one block.
Calculate the threshold value (Sa) to be assigned to the value. As the threshold value, for example, a simple average value of the densities of the pixels in one block is calculated and set as Sa. In the above-mentioned character area extraction processing, in the block judged to have a character area, with this threshold (Sa) as a reference, a value higher than this is "1", and a lower value is "0". To generate a binarized image B. The binarization processing described above is not performed on the block determined to have no character area, and all "1" s are set on the binarized image B.

Subsequently, the density information of the binarized image in the block is generated as follows. In the block judged to have the character area, the area of "0" of the binarized image and "1"
Area exists. For each area, the average value of the pixel densities is calculated as follows, and this is generated as image density information. In the binarized image, the simple average value of the pixel values in the area of "1" is Sh, and the simple average value of the pixel values in the area of "0" is Sl. An image C in which the pixels of the binary image B are replaced with Sh if the pixel value is "1" and Sl if the pixel value is "0" from the binary image B and Sh and Sl (gradation information). To generate.

Next, each pixel value of the image C is subtracted from the pixel value of the original image S to generate an image D. The image B is binary coded, the gradation information Sh and Sl are coded, the image D is color image (natural image) coded, and each coded data is output to the decoding side.

Color image encoding uses conventional techniques,
For example, the image is divided into y, cb, and cr color components, and each color component is processed in the same manner as the binarized image to _obtain images D _y , D _cb ,
Generate and encode D _cr . Specific examples of the above-mentioned encoding steps are shown in FIGS. 7 (a) to 7 (d). In the figure (a), the original image
S is composed of three pixel blocks, the first block is a non-character area, and the second and third blocks are character areas. The actual character part (binary part) in the character area is a black key type shown by hatching. Binary image B from original image S
When is created, it becomes the same figure (b). Pixel value is the white part
In "1", the black part is the device value "0". Pixel block
1 is not the target of the binarization process, so all pixel values
Replace with "1". When the image C is created from the binarized image B, the same figure (c) is obtained. The binary pixel value "1" is replaced with the Sh value, and the binary pixel value "0" is replaced with Sl. When the image D is created from the original image S and the image C, it becomes the same figure (d). Subtracting the pixel value of the image C at the same position from each pixel value of the original image S results in a color image D in which the black key type portion as the character portion is omitted. Although it is represented only as the color image D here, it is actually created for three colors of y, cb, and cr. In the above description, three blocks are connected to each other, but in reality, processing is performed in pixel block units. When the image B, the gradation information Sh, Sl, and the image D created as described above are encoded and output to the decoding side, the decoding side performs decoding (B, Sh, Sl, D ′), respectively. Image C is restored with decoded image B and tone information Sh, Sl, and the restored image
The image S "is obtained by synthesizing (adding) C and the image D '.

Decoding is performed as follows. The coded data is converted into binary data by the binary decoder, Sh and Sl by the (gradation information) decoder, and D _y , D by the natural image decoder.
Decode _cb and D _cr respectively to synthesize and generate image B and image D '. Then, the pixel of the binarized image B is binarized by the binarized image B and the gradation information Sh, Sl, and if the pixel value is "1", then Sh
Then, if the pixel value is "0", the image C replaced with Sl is generated. The image C and the image D ′ are added to generate an image S ″ which is a decoded image of the image S 1.

Further, the image read by the scanner has the property that the edge portion of the character is blunted at the boundary between the character and the background in the image, that is, the unnecessary high frequency component is added to the image D. The following method can be used to correct this rounding. On the encoding side, when encoding the image D with the natural image encoder, the high-frequency component with a blurred edge in the object contour (edge) part in the image D is removed when the image D has a property different from general natural images. In order to do so, after smoothing the image D, it is encoded by the natural image encoder, and on the decoding side, the image restoration process is performed on the image D ′ decoded by the natural image decoder. is there.

As an example, the formula for generating the value D _{i, j} of the corresponding pixel of the image D to be generated is changed according to each pixel value B _{i, j in} the pixel block of the binary image B to change D _{i, j} . Make it smaller. Therefore, on the encoding side, D _{i, j} = S _{i, j} −C _{i, j} + P (B _{i, j} = 1) = C _{i, j} −S _{i, j} + P (B _{i, j} = 0) P = S _{i, j} maximum possible value × 0.5 On the decoding side, S ′ _{i, j} = D ′ _{i, j} −C ′ _{i, j} + P (B ′ _{i, j} = 1) = C ′ _{i, j} −D ′ _{i, j} + P (B ′ _{i, j} = 0) P = S _{i, j} is processed by the equation of the maximum value × 0.5.

[0020]

[Embodiment 1] FIG. 1 is a block diagram of an encoding system according to the present invention. 1 is an image block circuit, 2 is an image binarization circuit,
Reference numeral 3 is an image synthesizing circuit, and 4 is a difference image creating circuit. Reference numeral 5 is a binary image encoder, 6 is an encoder, 7 is a conventional natural image encoder, and 8 is a signal output circuit. The image blocking circuit 1 converts the original image S into, for example, 8 vertical pixels × 8 horizontal pixels.
= Divide into pixel blocks consisting of 64 pixels. The image S'composed of all the divided pixel blocks is sequentially output to the character area discrimination circuit 9. The character area discrimination circuit 9
The image S ′ received from the image blocking circuit 1 is analyzed for each pixel block, and the image binarizing circuit 2 is instructed to perform the binarization process only on the pixel block determined to include the character. It is given to the binarization circuit 2.

As an example of the character area discrimination circuit 9, the following discrimination circuit can be considered. First, the maximum and minimum values of the pixel values in the pixel block are examined, and the pixel block in which the difference between the maximum value and the minimum value is very large is determined to be the character area, and the difference between the maximum value and the minimum value is close to 0. The block is determined as a non-character area. In the case where it is not possible to determine whether or not it is a character area by the above determination method, in order to investigate the distribution of pixel values that pixels can take for a pixel block, as shown in Fig. 4, within a block that takes a certain pixel value x Create a histogram of the number of pixels in H (x). This histogram is used to find the pixel value x where H (x) −H (x-1)> M and H (x) −H (x + 1)> M using the threshold value (parameter) M Considered as the peak of the distribution of values. If one or two peak values x exist, the pixel block is determined to be a character area, and otherwise it is determined to be a non-character area.

The above-mentioned character area discriminating circuit is only one example, and other circuits based on only maximum and minimum values are applicable. In this embodiment, by reducing the number of images processed by the image binarization circuit 2, there is an effect of reducing the number of encoded bits when the image B and the values Sh and Sl are encoded.

The image binarization circuit 2 is an image blocking circuit.
A pixel block divided by 1 is received, and an average value Sa of pixel values of pixels included in the pixel block is calculated for each pixel block. Next, among the pixels of the image S ′, a pixel having a pixel value larger than Sa is converted into “1” and a pixel having a lower pixel value is converted into “0” in each block to generate an image B, and the image combining circuit is generated.
3 and binary image encoder 5. At the same time, the average value of the pixel values of pixels with a pixel value larger than Sa for each block
The average value Sl of the pixel values of the pixels having smaller pixel values than Sh and Sa is calculated and transferred to the image synthesizing circuit 3 and the difference value encoder 6. The image synthesizing circuit 3 uses the image B received from the image binarizing circuit 2 and the average values Sh and Sl for each block to target all the pixels of the image B, and
If the pixel value is "1", the pixel value is replaced with the average value Sh of the block containing the pixel. On the other hand, if the pixel value is "0", the pixel value is replaced with the average value Sl of the block containing the pixel. Generate image C. The image C is transferred to the difference image creating circuit 4. The differential image generation circuit 4 generates an image D having a pixel value obtained by subtracting the pixel value of the image C from the pixel value of the image S for all the pixels of the image S'blocked and the image C received from the image composition circuit 3. Then, it is transferred to the natural image encoder 7. The binary image encoder 5 encodes the image B 1 received from the image binarization circuit 2 and transfers the encoded data Bc to the signal output circuit 8. As an example of the binary image encoder 5, an existing encoder such as JBIG or MMR can be used. The encoder 6 receives from the image binarization circuit 2,
Encode the mean values Sh and Sl for each block of the image,
The encoded data Shc and Slc are transferred to the signal output circuit 8. As an example of the encoder 6, it is possible to use the DPCM method for encoding the difference from the average value of the preceding block that is continuous. The natural image encoder 7 encodes the image D 1 received from the difference image creating circuit 4 and transfers the encoded data Dc to the signal output circuit 8. As an example of the natural image encoder, the above-mentioned conventional DCT can be used. The signal output circuit 8 includes data Bc, Shc, Slc, encoded by the binary image encoder 5, the encoder 6, and the natural image encoder 7.
Receives Dc and outputs the signal in a specific format.

FIG. 2 shows a decoding device of an encoding system according to the present invention. 11 is a signal input circuit, 12 is a binary image decoder,
13 is a decoder, and 14 is a natural image decoder. Reference numeral 15 is an image composition circuit, and 16 is a difference image composition circuit. Signal input circuit
11 is the encoded data input, encoded data
Generates Bc, Shc, Slc, and Dc '. The data Bc is sent to the 2-image decoder 12, and the data Shc and data Slc are sent to the decoder 13
Then, the data Dc ′ is transferred to the natural image decoder 14, respectively. The binary image decoder 12 decodes the data Bc received from the signal input circuit 11 and transfers the decoded image B 1 to the image synthesizing circuit 15.

As an example of the binary image decoder 12, the binary image decoder 12 shown in FIG.
Decoders such as JBIG and MMR corresponding to the value image encoder 5 can be used. The decoder 13 decodes the data Shc and Slc received from the signal input circuit 11, and transfers the decoded values Sh and Sl to the image synthesizing circuit 15. As an example of the decoder 13, the DPCM system corresponding to the encoder 6 of FIG. 1 can be used. The natural image decoder 14 decodes the data Dc received from the signal input circuit 11 and transfers the decoded image D ′ to the difference image synthesizing circuit 16. As an example of the natural image decoder, the DCT corresponding to the natural image encoder 7 in FIG. 1 can be used. The image synthesizing circuit 15 synthesizes the image C from the images B and Sh and Sl received by the binary image decoder 12 and the decoder 13, and transfers the image C to the difference image synthesizing circuit 16. The composition method is exactly the same as the image composition circuit 3 in FIG. The difference image synthesizing circuit 16 generates an image S ″ having a pixel value obtained by adding the pixel value of the image C received from the image synthesizing circuit 15 and the pixel value of the image D ′ received from the natural image decoder 14 and the pixel value of the image C. 8 to 11 show a sample of the original image S. It is composed of 516 pixels in the vertical direction,
The signal of the original image S when the 100th pixel from the top is horizontally encoded is shown in Fig. 8, the signal of image B is shown in Fig. 9, and the signal of pixel C is shown.
Figure 10 shows the image D signal. According to the present invention, when S is an image with mixed characters, the character area
Image B is encoded separately using Sh and Sl, and image S is encoded using a direct orthogonal transform encoder by removing image C and image C corresponding to the image encoded with average values Sh and Sl from image S. It is possible to prevent the character area from deteriorating as compared with the above.

[0026]

[Embodiment 2] FIG. 3 shows a character area discrimination circuit from Embodiment 1.
It is a block diagram except 9. The image is displayed without performing the character area identification process in the character area identification circuit 9 in the first embodiment.
This is an example in which binary coding processing and natural image coding processing are performed on all pixels of S ′. 1 to 8 are the same as in Example 1.

[0027]

[Embodiment 3] FIG. 5 is a configuration diagram in which 10 image smoothing circuits are added to the embodiment 2. 1 to 8 are the same as in Example 1. The image smoothing circuit 10 smoothes the image D received from the difference image creating circuit 4 to enable efficient coding of a region having a high frequency component not suitable for the natural image encoder, and then the natural image coding. Transfer to device 7. As an example of the image smoothing circuit 7, for example, first, the method of generating the image D is changed according to the pixel value of the image B, and if the pixel value of the image B is "1".
Then, image D is generated as in the first embodiment, while image B is generated.
If the pixel value of is 0, a circuit that subtracts the image S from the image C to generate the image D can be used, contrary to the first embodiment. FIG. 6 shows a block diagram of a decoding device when encoding is performed using this image smoothing circuit 7. The processes from 11 to 16 are based on the decoding device of FIG. 2 of the first embodiment. 17
Is an image restoration device. The image restoration device 17 performs a process of undoing the smoothing process performed by the image smoothing circuit of FIG. 5 on the image D ′ received from the natural image decoder 14. Image smoothing circuit 7 and image restoration circuit described above
17 is only one example, and in addition to the above, it is possible to apply a circuit that blunts sharp changes in the image and restores them. The present embodiment has an effect of improving the coding efficiency in the natural image encoder, and may be applied to the first embodiment.

12 to 15 show images processed according to the present invention.
This is a concrete example of S, image C, and image D.

[0029]

As described above, according to the present invention, the image is binarized for each block of the image from the target black-and-white gradation image or color image, and the binary image is encoded without information loss. Encoding is performed by a code converter, and an image obtained by removing the above binary image from the original image is coded by an orthogonal transform encoder, thereby realizing encoding with less deterioration of a black and white gradation image and a color image including a character area. be able to. The effect S / N ratio and the reduction of the number of encoded bits according to the present invention are as shown in the following table.

[Table 1] JPEG JPEG method only NonMM No division / no determination of character area. MM Maximum / minimum method. T = 32. PEAK Pixel value distribution method. M = 6, W ₁
= 32, W ₂ = 128.

[Brief description of drawings]

FIG. 1 is a block diagram of a coding side device to which the present invention is applied.

FIG. 2 is a block diagram of a decoding side device to which the present invention is applied.

FIG. 3 is a block diagram of an encoding side device to which the present invention is applied (method excluding character area recognition).

FIG. 4 is an example diagram of a pixel value distribution diagram in a pixel block.

FIG. 5 is a block diagram of a coding side apparatus to which the present invention is applied (a method using image smoothing).

FIG. 6 is a configuration diagram of a decoding side device to which the present invention is applied (a method using image smoothing).

FIG. 7 is a schematic diagram showing a specific example of processing by the method of the present invention.

FIG. 8 is a waveform diagram showing a sample of an original image S 1 processed according to the present invention.

FIG. 9 is a diagram showing changes in pixel values of an image B 1 processed according to the present invention.

FIG. 10 is a diagram showing changes in pixel values of an image C 1 processed according to the present invention.

FIG. 11 is a diagram showing changes in pixel values of an image D 1 processed according to the present invention.

FIG. 12 is a diagram showing an example of an image S 1 processed according to the present invention as a monochrome image.

FIG. 13 is a diagram showing an example of an image B 1 processed according to the present invention as a monochrome image.

FIG. 14 is a diagram showing an example of an image C 1 processed according to the present invention as a monochrome image.

FIG. 15 is a diagram showing an example of an image D 1 processed according to the present invention as a monochrome image.

[Description of Codes] 1 Image Blocking Circuit 2 Image Binarization Circuit 3 Image Compositing Circuit 4 Difference Image Creating Circuit 5 Binary Image Encoder 6 Encoder 7 Natural Image Encoder 8 Signal Output Circuit 9 Character Area Discrimination Circuit 10 Image smoothing circuit 11 Signal input circuit 12 Binary image decoder 13 Decoder 14 Natural image decoder 15 Image synthesis circuit 16 Difference image synthesis circuit 17 Image restoration circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 7/13 Z 4228-5C

Claims (6)

[Claims] 1. A still color image or a black-and-white gradation image is divided into a plurality of pixel blocks (n × m pixels), and each pixel in each block is binarized by a certain threshold value (Sa) to produce a binary encoded output. Each pixel value of each pixel block of the original image that is output and binarized and encoded is replaced with a specific calculation value based on each pixel value,
The calculated value coded output obtained by coding the replaced calculated value itself is output, the replaced pixel block is output as a natural image coded output, and the coded output of the still color image or the black and white gradation image is output as the 2 A binary natural image mixed still natural image encoding method for extracting a combined output obtained by combining the value encoded output, the calculated value encoded output, and the natural image encoded output. 2. A still color image or a black-and-white gradation image is divided into a plurality of pixel blocks (n × m pixels), and each pixel in each block is detected as a binary image portion or a pixel block including a binary image portion. Then, each pixel in the detected block is binarized based on the threshold (Sa) and output as a binary encoded output, and each pixel value of each binarized and encoded pixel block of the original image is Replace with a specific calculated value based on the pixel value,
The calculated value coded output obtained by coding the replaced calculated value itself is output, the replaced pixel block is output as a natural image coded output, and the coded output of the still color image or the black and white gradation image is output as the 2 A binary natural image mixed still natural image encoding method for extracting a combined output obtained by combining the value encoded output, the calculated value encoded output, and the natural image encoded output. 3. Each pixel value in the block is a threshold area (Sa)
If it is greater than the average value (Sh) of the value of each pixel having a value greater than the threshold value, if it has a value less than the average value of the value of each pixel having a value less than the average value Average value (S
3. The coding method for a binary natural image mixed still natural image according to claim 1, wherein l) is used as a specific calculated value. 4. A block other than a block in which the difference between the maximum value and the minimum value of each pixel in the block is substantially zero is detected as a binary image partial block or a pixel block including a binary image part. The encoding method for a binary natural image-containing still natural image according to claim 2. 5. A distribution of pixel values in the pixel block is calculated, and a block in which the pixel values are concentrated on one or two specific values is a binary image partial block or a pixel block including a binary image part. 3. The method of encoding a still natural image containing binary pixels according to claim 2, wherein the detection is performed. 6. An unnecessary high frequency component is removed from a pixel value obtained by subtracting a pixel value replaced by gradation information from an original image pixel value in a pixel at an edge portion of the natural image at the time of encoding the natural image. By so doing, it is possible to prevent blunting of the image at the edge portion, and the method for encoding a still natural image containing binary pixels according to claim 1 or 2.