JPH08186717A - Communication equipment - Google Patents

Abstract

PURPOSE: To provide a communication equipment provided with a function for promptly deciding the position of an adaptive template capable of efficiently encoding picture data. CONSTITUTION: A facsimile equipment binarizes the picture data read in a read part 1 in a picture processing part 2, encodes them in an encoding/decoding device 6, outputs them through a modem 7 and an NCU 8 to a telephone line and transmits them to a reception side. Prior to encoding, a control part 3 calculates an autocorrelation coefficient for the picture element block of a prescribed size or the prescribed number of the picture element lines of the binarized picture data, finds a picture element interval for indicating high correlation and defines it as the relative position of an adaptive template picture element and an encoding object picture element. The encoding/decoding device 6 refers to the adaptive template picture element and performs the encoding.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a communication device such as a facsimile, an electronic copying machine having a data transmission function and an image scanner, and more particularly to encoding of transmitted image data.

[0002]

2. Description of the Related Art In facsimiles, which are now widely used, binary (0/1, L level / H level, etc.) image data is transmitted. Further, binarized image data is also handled in an electronic copying machine, and recently binarized image data is stored in a storage device such as a magnetic disk and can be read out to reproduce an image as needed. It is supposed to be. In recent years, along with the diversification of image data, there has been an increasing demand for higher resolution of images, that is, higher density of the number of pixels, and the amount of data to be transmitted or stored is inevitably huge.

In order to efficiently transmit image data, 2
The value-coded data is encoded and compressed, and as its method, the modified Huffman (MH), modified read (MR), and modified / modified read (MMR) methods have been standardized and put into practical use. In these encoding methods, sequential buildup is performed in which binary data is encoded in the order in which the original image is scanned and read. In sequential build-up, like a facsimile, the sending device scans the paper surface of the document from left to right from top to bottom and transmits, and the receiving device decodes in the order received and reproduces the image on the paper surface. This is convenient because it is not necessary to store the received data in the receiving side device.

On the other hand, a high resolution image is not always required, and it may be necessary to quickly transmit and reproduce the entire image even with a rough quality. However, it is not easy to sufficiently deal with this in the above sequential buildup. For this reason, progressive build-up, in which coarse image data is first transmitted, and additional data is added as necessary to gradually increase the resolution of a reproduced image, has been receiving attention. Furthermore, there is an increasing demand for high-speed and faithful reproduction of halftone images.

Under such circumstances, JBIG (Joint Bi-l
evel Image Coding Experts Group) was established to study efficient coding of binary data. In JBIG, the main issues are parallelization of progressive build-up and sequential build-up, high data compression effect, information storability, and high-speed processing.
The base system of value data encoding is adopted.

First, a progressive build-up hierarchical image (images having different resolutions) is created by image reduction. Next, a pixel whose value is inevitably determined, that is, a pixel which does not need to be encoded is found from pixels whose values are known. This is performed in two steps, for example, a line in which all pixels are blank, in which the pixel value is determined irrespective of the image reduction method, and a line in which the pixel value is determined depending on the image reduction method. The former is a typical prediction and the latter is a definitive prediction. This reduces the number of pixels to be encoded. For the pixels remaining as the encoding target, arithmetic encoding is performed with reference to a pixel group called a template. A method called a QM coder is adopted for arithmetic coding.

A model template and an adaptive template (AT) are used as templates to be referred to when encoding a pixel to be encoded (hereinafter also simply referred to as an encoded pixel).
There is. The model template is a set of coded pixels located in the vicinity of the coded pixels, and the number of pixels forming the model template and the positional relationship with the coded pixels are defined. For example, the image of the lowest resolution layer in the progressive build-up has two lines of 9 pixels shown by M in FIG. 7 and 9 lines shown by M in FIG.
The one with three lines of pixels is used. In both figures, C is a coded pixel. For higher resolution layers,
Other model templates having different numbers of constituent pixels are prepared. In sequential build-up, there is only one hierarchy and the model templates in Figures 7 and 8 are used.

On the other hand, the adaptive template does not have a fixed positional relationship with respect to the coded pixels, but the position can be set according to the image. The default position of the adaptive template in the lowest resolution layer image is shown in FIGS. 7 and 8A. The adaptive template aims to improve the coding efficiency for an image having a periodic correlation between pixels, and is selected in consideration of the correlation of pixels. In particular, it is used for an image having a strong correlation in a pixel value at a constant cycle, such as a dither image obtained by binarizing an image of a halftone with a predetermined threshold value, to improve coding efficiency.

In FIGS. 7 and 8, x and y are coordinate axes when the coded pixel C is the origin, the arrow x indicates the front in the main scanning direction, and the arrow y indicates the rear in the sub scanning direction. As the encoding progresses, the encoded pixel moves by one pixel in the main scanning direction, and the pixel positions of the model template and the adaptive template also move by one pixel accordingly.
The relative position of the template with respect to the coded pixel is kept constant except for the peripheral area of the image. For the coded pixels existing in the peripheral area of the image data, the relative positions of the model template and the adaptive template cannot be kept constant, but an exception is defined separately for this.

The selection of the model template and the determination of the position of the adaptive template with respect to the coded pixel are
It is performed according to the image transmitted by the transmitting side. The selected model template and adaptive template position are described in the header along with other information used for encoding,
It is sent before the image data. The receiving side decodes the encoded image data based on the information written in the header. Therefore, the image data obtained by scanning and reading the original on the transmitting side is surely reproduced on the receiving side.

The compression rate of image data can be expressed by various methods, but the most general expression is expressed by the equation (1).
Expression (1) represents the size of the image data after encoding with respect to the size of the image data before encoding, and the smaller the value, the higher the compression rate.

[0012]

[Equation 1]

CCITT (International Telegraph and Telephone Consultative Committee) No. When one original image is encoded, MH, MR, MM
In the R and JBIG systems, 0.14 and 0.1, respectively.
Compression ratios of 3, 0.035 and 0.028 have been obtained. The original document image is image data composed of characters, tables, graphs, etc., and is compressed well by any method. However, the former three methods are not intended for the image data of the intermediate gradation, so that the compression efficiency is significantly reduced for such image data, and the compression rate becomes 1 or more and the reverse It is known that the amount of data increases. On the other hand, the JBIG method is expected to achieve good compression with a compression rate of about 0.1 even for image data of halftone.

[0014]

As described above, the adaptive template is referred to in order to efficiently encode and compress an image having a periodic correlation between pixels. In JBIG, the adaptive template pixel exists. Although it defines the range, it does not give instructions on how to determine the position of the adaptive template pixel. Further, there is no conventional technique showing a method of determining the adaptive template pixel position.

For the purpose of the adaptive template, it is important to use a pixel having a high correlation with the coded pixel. When the adaptive template is a pixel having a low correlation with the coded pixel, the coding efficiency is lowered and the compression rate is also lowered. On the other hand, when it takes a long time to determine the position of the adaptive template, even if the coding efficiency is improved, the efficiency of the entire transmission of the image data may be lowered.

It is an object of the present invention to provide a communication device having a function of rapidly determining the position of an adaptive template pixel which can efficiently encode image data.

[0017]

In order to achieve the above object, according to the present invention, in a communication device for encoding image data according to the JBIG encoding method, a plurality of pixels lined up in a row are defined as pixel lines, A means for performing an algorithm for calculating a pixel interval that gives a maximum correlation coefficient for each of a number of pixel lines, and based on the pixel interval having the maximum appearance frequency among the pixel intervals. Coding means for deciding positions of adaptive template pixels and coding image data.

In addition, in a communication device that encodes and transmits image data composed of binarized pixels, the autocorrelation coefficient of the pixel value of each of a plurality of pixel lines in which the pixels are continuously arranged in a line. To obtain a pixel interval that gives the maximum correlation coefficient, and a pixel located at a position separated from the encoding target pixel by the reference interval with the one that appears most frequently among the pixel intervals as the reference interval. And a coding unit that codes the pixel to be coded with reference.

Further, in a communication device which encodes and transmits image data composed of binarized pixels, a two-dimensional autocorrelation coefficient of pixel values of a pixel block in which pixels are continuously arranged in a vertical direction and a horizontal direction. For obtaining the maximum correlation coefficient to obtain the vertical pixel interval and the horizontal pixel interval, and the vertical pixel interval and the horizontal pixel from the encoding target pixel in the vertical direction and the horizontal direction, respectively. Coding means for coding a pixel to be coded with reference to pixels located at a distance.

Specifically, the above configuration is realized in a facsimile machine.

Further, a document reading unit, an image processing unit for processing the image data read by the document reading unit and giving the image data to the encoding unit, and the image data encoded by the encoding unit to the storage device. The above configuration may be realized in a copying machine and an image scanner that include a transmitting unit that transmits.

[0022]

In a communication device that encodes image data according to the JBIG encoding method, a plurality of pixels arranged in a row in a row is used as a pixel line, and an autocorrelation coefficient of pixel values is calculated for each of a predetermined number of pixel lines. Means for executing an algorithm for obtaining a pixel interval giving the maximum correlation coefficient, and encoding means for deciding the position of the adaptive template pixel based on the pixel interval having the maximum appearance frequency among the pixel intervals and encoding the image data. When the configuration is provided with, the pixel interval having the maximum correlation is obtained for each pixel line. Although this pixel interval differs for each pixel line, it can be expected that a pixel that frequently appears in a predetermined number of pixel lines has a high appearance rate over the entire image. An adaptive template pixel whose position is determined based on the pixel interval of the maximum appearance frequency has a very high probability of correlation with the pixel to be encoded.

Further, in a communication device for encoding and transmitting image data composed of binarized pixels, an autocorrelation coefficient of pixel values is obtained for each of a plurality of pixel lines in which pixels are continuously arranged in a line. To obtain a pixel interval that gives the maximum correlation coefficient, and a pixel located at a position separated from the encoding target pixel by the reference interval with the one that appears most frequently among the pixel intervals as the reference interval. When a coding unit for coding the pixel to be coded by reference is provided, the pixel interval having the maximum correlation is obtained for each pixel line. Of the pixel intervals, the reference interval that appears at the highest frequency in a plurality of pixel lines has a high probability of indicating the interval between pixels having a strong correlation even in the pixels of the image data for which the correlation is not actually calculated. Therefore, the pixel located at the reference distance from the pixel to be encoded correlates with the pixel to be encoded with high probability.

Further, in a communication device which encodes and transmits image data composed of binarized pixels, a two-dimensional autocorrelation coefficient of pixel values of a pixel block in which pixels are continuously arranged in a vertical direction and a horizontal direction. For obtaining the maximum correlation coefficient to obtain the vertical pixel interval and the horizontal pixel interval, and the vertical pixel interval and the horizontal pixel from the encoding target pixel in the vertical direction and the horizontal direction, respectively. In a configuration including a coding unit that codes a pixel to be coded by referring to pixels located at positions separated by an interval, the pixel interval having the maximum correlation in a pixel block is obtained in both the vertical and horizontal directions. . Two pixels located in the vertical direction and the horizontal direction at the same pixel interval as each other have a high probability of correlation even outside the pixel block. Therefore, when encoding the pixel to be encoded,
Pixels that are highly correlated will be referenced.

When the above configuration is specifically realized in the facsimile apparatus, the pixel referred to at the time of encoding correlates with the pixel to be encoded with a high probability, so that the encoding efficiency is increased and the compression rate is also improved. The transmission time of the image data is shortened.

Further, an original reading means, an image processing means for processing the image data read by the original reading means and giving the image data to the encoding means, and the image data encoded by the encoding means to the storage device. When the above-described configuration is realized in a copying machine and an image scanner including a transmitting unit for transmitting, the image data read by the document reading unit and processed by the image processing unit is coded with reference to highly correlated pixels. Therefore, it is efficiently encoded and compressed. Therefore,
The transmission time from the transmission means of the copying machine and the image scanner to the storage device is shortened, and the amount of data stored in the storage device is reduced.

[0027]

1 is a block diagram showing the configuration of a first embodiment in which the present invention is applied to a facsimile. The facsimile apparatus according to the present exemplary embodiment includes a reading unit 1 that reads a document image, an image processing unit 2 that processes image data output from the reading unit 1, a control unit 3 that controls the entire apparatus, and a ROM 4 that stores a control program and the like. RAM for temporarily storing image data, etc.
5, encoding / decoding device 6 for encoding and decoding image data, modem 7 for modulating and demodulating encoded image data, N for connecting to a telephone line and transmitting data
It has a CU (network control unit) 8, an operation unit 9 for a user to give an instruction to the control unit 3, and a recording unit 10 for outputting image data on a recording sheet.

The reading unit 1 is equipped with a light source and a CCD (charge coupled device). The image of the original is read by irradiating the original image with light from the light source and detecting the reflected light with the CCD.
The image is read by repeating scanning in the horizontal direction (main scanning) in the vertical direction at a predetermined pitch (sub scanning). CC
D outputs the detected light as an analog signal for each pixel.

The image processing unit 2 performs shading processing such as correction on the analog output from the CCD of the reading unit 1 due to variations in CCD sensitivity, and then compares the image data with a predetermined threshold value for each pixel. Binarize to. Further, it has two types of dither matrices, and when binarizing an image of an intermediate gradation, dither processing is performed using these matrices.

In addition to the device control program, the ROM 4 stores various tables including programs and model templates for performing various JBIG processes shown in FIG. The RAM 5 stores and holds the image data binarized by the image processing unit 2 until transmission, and also temporarily stores various calculation results performed by the control unit 3. Also, the received image data is stored and held. The encoding / decoding device 6 receives the image data for transmission,
Arithmetic coding is performed by the QM coder. For the received image data, the encoded image data is decoded in the reverse procedure to reproduce the image. Modem 7 and N
The CU 8 transmits and receives image data by a known standard method.

The control unit 3 is composed of a microcomputer and reads the program stored in the ROM 4 to control the entire facsimile apparatus. Moreover, the position of the adaptive template pixel referred to at the time of encoding is determined. The control unit 3 includes a timer circuit 11 for timer transmission that transmits image data at a time designated by the user.

FIG. 2 shows a partial external view of the operation section 9. The operation unit 9 is a numeric keypad 5 for inputting a numerical value such as a facsimile number.
1, a start key 52 for instructing the start of reading an original, a stop key 53 for instructing to stop the operation, a memory print key 54 for instructing the output of the received data held in the RAM 5 by the recording unit 10, a start of designation of the transmission time Together with various keys such as the timer key 55 to indicate the
It has a display panel 56 made of D (liquid crystal display device).

The destination facsimile number input from the ten keys 51 is displayed on the display panel 56 and can be confirmed. If there is an error in the entered number, the cursor key 57 is used to move the cursor on the display panel 56 to that digit, the numeric keypad 51 is used to re-enter a number, and the setting key 58 is used to confirm. When performing timer transmission, the timer key 55 is operated and then the transmission time is input from the ten keys 51. The input transmission time is displayed on the display panel 56, and can be changed by operating the cursor key 57 or the like, like the facsimile number. The display panel 56 also displays a simple message regarding the operation to assist the user in the input operation.

The operation section 9 includes a status display section 59 for displaying the status of the apparatus, a density designating section 60 for instructing the image density of the original, and an image quality designating section 6 for instructing the image quality of the original image.
1 is provided. The status display unit 59 has an LED (light emitting diode), and the received image data is held in the RAM 5 when there is an abnormality in the device or a user's erroneous operation, when the recording paper in the recording unit 10 runs out. When the automatic reception mode is set, the LEDs corresponding to error, paper supply, memory, and automatic reception are turned on to notify the user. Memory LED
When 62 is lit, by operating the memory print key 54, the image data stored in the RAM 5 is output from the recording unit 10 to the recording paper.

The density designating section 60 has a density designating key 63 and three LEDs 64, and the LED 64 is switched on each time the density designating key 63 is operated. The user specifies the density in three levels of dark, normal, and light according to the original image. When binarizing the output of the reading unit 1, the image processing unit 2 sets a threshold value corresponding to the stage specified here. As a result, the image data is appropriately binarized even if the image density varies from document to document.

The image quality designating section 61 has an image quality designating key 65 and five LEDs 66, and the LED 66 is switched on each time the image quality designating key 65 is operated. The user makes a selection according to the image quality of the document. That is, when the original image is a character image, normal, fine or super fine is designated. This determines the pixel density of the highest resolution layer. When the original image has halftone, halftone or super halftone is designated. The image processing unit 2 is provided with two types of dither matrices, and selects a matrix according to the designation here and performs dither processing, and binarizes the output of the reading unit 1.

The recording unit 10 employs a system in which an electrostatic latent image of an image is formed on the surface of a photosensitive drum, toner is attached to the electrostatic latent image to develop the image, and then the image is transferred to recording paper.

In the facsimile apparatus having the above structure,
All the image data read by the reading unit 1 and binarized by the image processing unit 2 are temporarily stored in the RAM 5. A part of the image data stored in the RAM 5 is the control unit 3.
The position of the adaptive template that is read out by the above and referred to when encoding is determined. The method of determining the adaptive template position will be described in detail later. In the progressive build-up, the image data stored in the RAM 5 is read by the control unit 3 to reduce the image, and the reduced image data is stored in the RAM 5.

At the time of transmission, information used for encoding the determined adaptive template position and selected model template is written in the header, modulated by the modem 7 and transmitted via the NCU 8. Then RAM5
The image data stored in is sequentially read and encoded by the encoding / decoding device 6 with reference to the model template and the adaptive template. The encoded image data is modulated by the modem 7, output from the NCU 8 to the telephone line, and transmitted to the receiving side.

On the other hand, the data received via the NCU 8 is first demodulated by the modem 7. A header is added to the beginning of the received data, and the control unit 3 reads out information at the time of encoding such as a model template and an adaptive template from the header. The encoding / decoding device 6 refers to these pieces of information and sequentially decodes the demodulated image data. As a result, the binary image data before being encoded by the transmission side device is reproduced. The binarized image data is immediately output to the recording paper by the recording unit 10 in the sequential buildup, once stored in the RAM 5 in the progressive buildup, and then by the operation of the memory print key 54 of the operation unit 9, or When the image with the highest resolution is reproduced, the recording unit 10 outputs the image on recording paper.

At the time of encoding, it is important to refer to a pixel having a strong correlation with the pixel to be encoded in order to improve the encoding efficiency. In the facsimile apparatus of this embodiment, the correlation between pixels is calculated by the following method, and the position of the adaptive template is determined based on the calculated correlation.

The m number of pixels lined up continuously in one line in the main scanning direction is set as a pixel line, and the one-dimensional autocorrelation coefficient ψ (k) is obtained by the equation (2).

[0043]

[Equation 2]

Here, X _i is 0 in the pixel line.
It is the value of the i-th pixel of the pixels from to m-1. In the numerator on the right side of Expression (2), the product of two pixels at positions separated by an interval k is calculated to obtain the sum. Only the combination in which the values of the two pixels are both 1 (H level) contributes to this sum. The denominator is the value 1 at that pixel line
Represents the total number of pixels having. Of the m autocorrelation coefficients ψ (0), ψ (1) ... ψ (m-1) thus obtained, the pixel interval k giving the maximum value is selected, and this is selected as the representative interval in this pixel line. And

The calculation of the autocorrelation coefficient and the selection of the representative intervals are performed for n consecutive pixel lines in the sub-scanning direction to obtain n representative intervals. Next, of these representative intervals, the one that appears most frequently is used as the reference interval and is used as the interval between the coded pixel and the adaptive template pixel. There are two pixels which are separated from the coded pixel by the reference interval in the front and rear in the scanning direction, but the pixel already coded in the rear in the scanning direction is referred to as an adaptive template pixel. In this way, the relative position of the adaptive template pixel with respect to the coded pixel is determined.

When the representative interval of the maximum appearance frequency is 0, that is, when the reference interval is 0, the adaptive template pixel is set at the default position. For example, in the case of the image data of the lowest resolution layer, the position indicated by A in FIGS. 7 and 8 is the adaptive template pixel.

In this embodiment, the number m of pixels forming a pixel line is 20, the number n of pixel lines is 10, and R
It is stored in OM4. Although these settings are arbitrary, the number of pixels forming the pixel line and the number of pixel lines are directly reflected in the calculation of the autocorrelation coefficient and influence the determination of the adaptive template position. The larger these are set, the more extensively the correlation of the image data is examined, and the appropriate adaptive template pixel position is determined. On the other hand, however, the time required to determine the adaptive template pixel increases. The number of pixels m and the number of pixel lines n may be input from the operation unit 9 and set.

Although the setting position of the pixel line is also arbitrary, when the pixel line is set in the peripheral portion of the pixel data for one page, the correlation coefficient and the representative interval obtained from this are the same as the image data. It may not be representative of the whole. For example, the upper part of the first page is usually used for stylizing the name of the destination, the date and time of transmission, etc., and has a lot of blank spaces as compared with other parts.
Therefore, there is a high possibility that the position of the adaptive template pixel determined based on the image data of this portion will not be suitable for use in encoding all the image data. In this embodiment, in consideration of such a point, the image data in the vicinity of the central portion of the first page among all the image data is used for the pixel line.

The transmission process of the facsimile apparatus of this embodiment is shown in the flowchart of FIG. First, step # 1
In 05, the document image is read by the reading unit 1, the image data is binarized for each pixel by the image processing unit 2, and stored in the RAM 5. This operation is repeated until the reading of all pages of the document is completed. Then # 11
At 0, the first page of the image data is read from the RAM 5 to the control unit 3.

In # 115, the control section 3 determines the first pixel line position according to the size of the read image data, that is, the size of the page, and in # 120, sets the variable n representing the pixel line number to the initial value. Set to 1. Step # 125
In, the autocorrelation coefficient ψ _n (k), (k = 0, 1, 2, ..., 19) of the n-th pixel line is calculated according to the equation (2). # 1
At 30, the k corresponding to the maximum value of these 20 autocorrelation coefficients is stored as kn. In # 135, 1 is added to n, and in # 140, it is determined whether n is 10 or less. When n is 10 or less, the process returns to # 125. By this iterative operation, k1 to k10 are obtained.

When it is determined in step # 140 that n exceeds 10, in step # 145, the one with the highest appearance frequency out of k1 to k10 is found, and this is set as the reference interval kmax. A position separated by the number of pixels kmax in the backward direction is determined as the position of the adaptive template pixel. When kmax is 0, the adaptive template pixel is set at the default position.

Then, in step # 155, a header including information necessary for decoding such as adaptive template pixel position is created and transmitted. In # 160, the image data is sequentially read from the RAM 5, and the encoding / decoding device 6 encodes the encoded pixel by the template operation. # 16
The image data encoded in 5 is transmitted.

According to the above method, the pixel interval having a large correlation is obtained for each pixel line, and the one having the highest appearance frequency in all pixel lines is selected as the reference interval, so that the image data not included in the pixel line is selected. Even in this case, there is a high probability that a strong correlation will occur at this reference interval. Therefore, when the adaptive template pixel is set at this reference interval, the coded pixel and the adaptive template pixel correlate with high probability, and the coding efficiency is improved. That is, the time required for encoding is shortened and the compression rate of image data is improved. Moreover, since the autocorrelation coefficient is obtained only for a predetermined number of pixels, the position of the adaptive template pixel is determined in a short time.

When various image data were encoded by the facsimile apparatus of the present embodiment, the compression ratio of 0.02 to 0.4 obtained by the above equation (1) was obtained. CCITT No. The compression ratio of one original was 0.02. Further, even in the image data that is subjected to the dither processing and the pixel values have a periodic correlation, the compression rate is improved to about 0.08. As described above, according to this method, binarized image data can be efficiently coded.

A facsimile apparatus according to the second embodiment of the present invention will be described. The facsimile apparatus according to the present embodiment has the same hardware configuration as that of the first embodiment shown in FIG. 1, and duplicate description will be omitted. In this embodiment, the position of the adaptive template pixel is determined as follows.

A pixel block is a set of m pixels arranged in the horizontal direction (main scanning direction) and n pixels arranged in the vertical direction (sub scanning direction), and a two-dimensional self-correlation is calculated for this pixel block according to the equation (3). Calculate the number.

[0057]

(Equation 3)

Here, X _i , _j represents the value of the pixel located i-th in the horizontal direction and j-th in the vertical direction in the pixel block. In the right-hand side numerator, the product of two pixels, which are separated by k1 in the horizontal direction and k2 in the vertical direction, is calculated, and the total sum is obtained. Only the combination in which the values of the two pixels are both 1 (H level) contributes to this sum. The denominator represents the total number of pixels having the value 1 in the pixel block.

From equation (3), m × n autocorrelation coefficients ψ
(0,0), ψ (1,0) ... ψ (m-1, n-1) are obtained. Pixel intervals k1 and k2 corresponding to the maximum value among these are defined as a horizontal reference interval and a vertical reference interval, respectively, and are defined as a horizontal relative position and a vertical relative position of the adaptive template pixel with respect to the encoded pixel. When the autocorrelation coefficient becomes maximum when both k1 and k2 are 0, that is, when ψ (0,0) is maximum, the adaptive template pixel is set to the default position.

In the present embodiment, the size of the pixel block is set to 20 for m and 10 for n, and is stored and held in the ROM 4. Although these settings are arbitrary, the size of the pixel block influences the determination of the adaptive template position. The larger this value is set, the more extensively the correlation of the image data is examined and the appropriate adaptive template pixel position is determined, but on the other hand, the time required to determine the adaptive template pixel increases. Further, the setting position of the pixel block is also arbitrary, but for the reason described in the first embodiment, the image data in the vicinity of the central portion of the first page among all the image data is used for the pixel block.
Note that m and n may be input from the operation unit 9 to set the size of the pixel block.

The adaptive template pixel whose relative position to the coded pixel is determined by the above method has a high probability of being correlated with the coded pixel over the entire image data. Therefore, the pixel coding efficiency is improved and the image data compression rate is also increased.

The processing of the facsimile apparatus of this embodiment at the time of transmission is shown in the flowchart of FIG. First, step # 2
In 05, the document image is read by the reading unit 1, the image data is binarized for each pixel by the image processing unit 2, and stored in the RAM 5. This operation is repeated until the reading of all pages of the document is completed. Then, # 21
At 0, the first page of the image data is read from the RAM 5 to the control unit 3.

In # 215, the control section 3 sets 20 × 1 according to the size of the read image data, that is, the size of the page.
The position of the pixel block of 0 pixels is determined. Next, in step # 220, the two-dimensional autocorrelation coefficients ψ (k1, k2), (k1 = 0,1,2, ..., 19, k of the pixel block according to the equation (3) are calculated.
Calculate 2 = 0, 1, 2, ... 9). # 225, these 200
Among these autocorrelation coefficients, k1 and k2 that give the maximum value are found, and they are set as the reference interval in the horizontal direction and the reference interval in the vertical direction. Based on these reference intervals, the relative position of the adaptive template pixel with respect to the coded pixel is determined.

In step # 230, a header including information necessary for decoding the adaptive template pixel position and the like is created and transmitted. In step # 235, the image data is sequentially read from the RAM 5, and the encoding / decoding device 6 encodes the encoded pixel by the template operation. The image data encoded in # 240 is transmitted.

The block diagram of FIG. 5 shows the configuration of the electronic copying machine according to the third embodiment of the present invention. The copying machine includes a scanner 31, an image processing unit 32, a printer 33, a control unit 34, an operation unit 35, an encoding / decoding unit 36, a transmission / reception unit 37, and a storage device 38. The original image is read by the scanner 31, and the image processing unit 3 operates on the read image data.
In step 2, shading, binarization and the like are performed, and the printer 33 outputs the recording sheet.

The control unit 34 controls the entire copying machine. The operation unit 35 includes various keys such as a numeric keypad,
By operating the keys of the operation unit 35, the user gives instructions to the control unit 34 such as recording paper size selection, number of copies, image enlargement / reduction, and image density setting. Further, the operation unit 35 has a display panel, and displays a message relating to the operation and an echo back display of the ten-key input to assist the operation of the user. The storage device 38 is provided outside the main body of the copying machine and is connected to the transmission / reception unit 37 via a data line 39. As a storage medium of the storage device 38, for example, R
AM, a magnetic disk or an optical disk is used.

The encoding / decoding unit 36 encodes the binarized image data according to the JBIG method. Transmitter / receiver 37
Outputs the image data encoded by the encoding / decoding unit 36 to the data line 39 and transmits it to the storage device 38. The storage device 38 stores and stores the received data. Further, the transmitting / receiving unit 37 receives the image data stored in the storage device 38 via the data line 39. The received data is decoded by the encoding / decoding unit 36.

The model template and the adaptive template are referred to when the image data is encoded by the encoding / decoding unit 36, and the position of the adaptive template pixel is determined by the method shown in the first embodiment or the second embodiment. Determined by That is, the one-dimensional autocorrelation coefficient is calculated for each of the plurality of pixel lines by the equation (2), and the pixel interval having the highest appearance frequency among the pixel intervals giving the maximum correlation coefficient is set as the reference interval. Alternatively, the two-dimensional autocorrelation coefficient is calculated for the pixel block by the equation (3), the pixel interval giving the maximum correlation coefficient is obtained in the vertical and horizontal directions, and these are referred to in the vertical and horizontal directions. Interval

The adaptive template pixel position is determined using the reference interval thus obtained as the interval between the coded pixel and the adaptive template pixel. The position of the adaptive template pixel is written in the header, stored in the storage device 38 together with the image data, and referred to by the encoding / decoding unit 36 at the time of decoding.

In the copying machine having the above-mentioned structure, the image data read by the scanner 31 and processed by the image processing unit 32 can be output to the recording paper by the printer 33, or can be encoded and stored in the storage device 38. You can also The user operates the keys of the operation unit 35 to select output to recording paper and storage in the storage device 38. Further, the image data stored in the storage device 38 is read out and output to the recording paper by the printer 33 at any time. This is also performed by the key operation of the operation unit 35.

The adaptive template pixels whose relative position to the coded pixel is determined by the above-described method correlates with the coded pixel with a high probability, so that the image data is coded efficiently and the compression rate is improved. As a result, the transmission time of the image data between the transmitter / receiver 37 and the storage device 38 is shortened, the amount of data to be stored is reduced, and the physical storage area occupied by the image data in the storage device 38 is reduced.

Although the storage device 38 is installed outside the main body of the copying machine in this embodiment, the storage device 38 may be installed inside the main body of the copying machine. Further, the configuration excluding the printer 33 from the copying machine functions as an image scanner, and the present invention is also applicable to an image scanner.

[0073]

As described above, according to the present invention, in the communication device that encodes image data according to the JBIG encoding method, the pixel interval exhibiting the maximum correlation is obtained for each of the predetermined number of pixel lines. Since the position of the adaptive template pixel is determined based on the one having the highest appearance frequency, the adaptive template and the pixel to be coded correlate with high probability. Therefore, the coding efficiency is improved, the time required for the coding is shortened, the data compression rate is increased, and the image data can be transmitted in a short time. This effect becomes more remarkable for image data in which a strong correlation appears in a constant cycle such as a dither image. Furthermore, since the correlation of a limited number of pixels is calculated instead of calculating the correlation of all pixels, the time required to determine the position of the adaptive template may be short.

Further, according to the communication device of the second aspect, the pixel interval giving the maximum correlation coefficient is calculated for each of the plurality of pixel lines, and the pixel interval having the highest appearance frequency among them is set as the reference interval. Pixels located at a reference interval away from are highly likely to be correlated with the encoding target pixel. Therefore, the coding efficiency of the binarized image data is increased, the coding time is shortened, the compression rate is increased, and the data transmission efficiency between the communication devices is also improved. The time required to determine the reference interval is also short.

According to the third aspect of the invention, since the two-dimensional autocorrelation coefficient is calculated within the pixel block, the correlation between pixels can be investigated in a wide range. Among these, the pixel interval showing the maximum correlation is obtained in both the vertical direction and the horizontal direction, and is set as the relative position of the pixel to be encoded and the pixel to be referred to, so that there is a high probability that both pixels are correlated. Therefore, 2
The time required for encoding the binarized data is shortened, the compression rate of the image data is increased, and the data transmission efficiency between the communication devices is also improved.

In the facsimile apparatus according to the fourth aspect, when the transmission side apparatus performs encoding, the pixel which is highly likely to be correlated with the encoding target pixel is referred to, so that the encoding efficiency and the compression rate are increased. Along with this, the time for transmitting the image data from the transmission side device to the reception side device is shortened, and the transmission efficiency is improved.

Further, in the copying machine according to the fifth aspect and the image scanner according to the sixth aspect, since the encoding is performed with reference to the pixel having a high probability of being correlated with the pixel to be encoded, the encoding efficiency is improved. The data compression rate is also increased, and the image data transmission time to the storage device is shortened. Further, the storage area of the storage device required to store the image data is reduced.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of a facsimile apparatus according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an appearance of an operation unit of the facsimile apparatus according to the first embodiment.

FIG. 3 is a flowchart showing a process at the time of transmission of the facsimile apparatus of the first embodiment.

FIG. 4 is a flowchart showing a process at the time of transmission of the facsimile apparatus according to the second embodiment of the present invention.

FIG. 5 is a block diagram showing the configuration of an electronic copying machine according to a third embodiment of the present invention.

FIG. 6 is a block diagram showing an outline of a JBIG base system.

FIG. 7 is a diagram showing default positions of a two-line model template and an adaptive template of an image of a lowest resolution layer.

FIG. 8 is a diagram showing default positions of a 3-line model template and an adaptive template of an image of a lowest resolution layer.

[Explanation of symbols]

 1 reading unit 2 image processing unit 3 control unit 4 ROM 5 RAM 6 encoding / decoding device 7 modem 8 NCU 9 operation unit 10 recording unit

Claims (6)

[Claims] 1. A communication device for encoding image data according to the JBIG encoding method, wherein a plurality of pixels lined up in a row are defined as pixel lines, and an autocorrelation coefficient of pixel values is calculated for each of a predetermined number of pixel lines. Means for performing an algorithm for obtaining a pixel interval that gives a maximum correlation coefficient, and a code for deciding the position of the adaptive template pixel based on the pixel interval having the maximum appearance frequency among the pixel intervals and encoding the image data. And a communication unit. 2. A communication device for encoding and transmitting image data composed of binarized pixels, wherein an autocorrelation coefficient of pixel values is provided for each of a plurality of pixel lines in which the pixels are arranged in a row. To obtain a pixel interval that gives a maximum correlation coefficient, and a pixel located at a position distant from the encoding target pixel by the reference interval with a reference interval being the one that appears most frequently among the pixel intervals. A communication device, comprising: an encoding unit that refers to and encodes a pixel to be encoded. 3. A communication device for encoding and transmitting image data composed of binarized pixels, wherein a two-dimensional autocorrelation coefficient of pixel values is provided for a pixel block in which pixels are continuously arranged in a vertical direction and a horizontal direction. And a means for calculating the vertical pixel interval and the horizontal pixel interval that give the maximum correlation coefficient, and the vertical pixel interval and the horizontal pixel from the encoding target pixel in the vertical direction and the horizontal direction, respectively. A communication device, comprising: an encoding unit that encodes a pixel to be encoded by referring to pixels located at a distance. 4. The communication device according to claim 1, 2 or 3 is a facsimile device. 5. The communication device according to claim 1, 2 or 3, wherein a document reading unit, an image data read by the document reading unit, and an image which gives the image data to the encoding unit. A copying machine comprising a processing means and a transmission means for transmitting the image data encoded by the encoding means to a storage device. 6. The communication device according to claim 1, 2 or 3, wherein an image is read from a document reading unit, an image data read by the document reading unit is processed, and the image data is given to the encoding unit. It is an image scanner comprising a processing means and a transmission means for transmitting the image data encoded by the encoding means to a storage device.