JP2807293B2 - Image data encoding / decoding method and facsimile apparatus - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for encoding / decoding image data and a facsimile apparatus, and more particularly to an encoding / decoding method for superimposing / separating other data on image data. The present invention relates to a decoding method and a facsimile apparatus to which the method is applied.

[Prior art] Conventionally, in order to make recording and transmission of image data more efficient,
A technique of encoding and compressing image data is employed. For example, a modified read (MR) method and a modified MR (MMR) method that binarize an image into black and white and compress the image using a two-dimensional correlation of the binarized pixels are known. These methods do not perform encoding / compression using only one horizontal scanning line as in the modified Huffman (MH) method, which encodes the run lengths of binarized pixels as they are, but use adjacent scanning lines. Is used to classify the change in the boundary between black and white pixels in the coded scan line and the reference scan line immediately before the coded scan line into three types of modes: a pass mode, a vertical mode, and a horizontal mode. Encode. Since such an image data encoding / decoding method is used in a facsimile apparatus, a description will be given below in accordance with a facsimile method.

Regarding the arrangement of pixels, a specific pixel is defined as shown in FIG. A specific pixel whose color (white or black) has changed from the previous pixel on the same scanning line is called a changed pixel. That is, a0: a starting point start change pixel on the coded scan line. However, at the start of the coded scan line (left end), a0 is a pixel virtually placed immediately before (left) the first pixel. A1: The first changed pixel on the coded scan line and to the right of a0 a2: The first change pixel on the coded scan line and to the right of a1 b1: The first change pixel on the reference scan line and to the right of a0 and having the opposite color to a0 b2: On the reference scan line and to the right of b1 First change pixel However, if a1, a2, b1, and b2 except a0 cannot be set on the scan line, they are virtually set immediately after (right end) the last (right end) pixel of each scan line. Shall be

Next, three types of modes illustrated in FIG. 19 are defined as follows, and the codewords in Table 1 are applied.

This means that the pass mode b2 is located to the left of a1. At this time, immediately below b2 is encoded in the pass mode, and then a0 is moved directly below b2.
(FIG. 19 (A)) This refers to a case where the distance Δ between the vertical modes a1 and b1 is within ± 3 pixels.
At this time, the position of a1 is encoded in the vertical mode. next,
Move a0 to a1. (FIG. 19 (B)) That is, in the vertical mode, the pixel position shift in the scanning line direction between the changed pixels on the reference scanning line and the changed pixels on the coded scanning line having the same value (same color) (see FIG. The position of the changed pixel on the coded scanning line that satisfies the condition that the distance Δ) is within ± 3 pixels is coded with a code representing the shift.

It is neither horizontal mode nor horizontal mode. At this time, a0a1 and a1a2 are MH
(M (a0a1), M in the rightmost column in the second row of Table 1)
(A1a2)). Next, a0 is moved to the position of a2. (FIG. 19 (C)) In the MR method, the first scanning is performed at least every K scanning lines.
The scanning lines are encoded by the MH method, and the subsequent K-1 scanning lines are encoded according to the above-described mode while referring to the immediately preceding scanning line. This value is called a K parameter. In a facsimile machine, K = K at standard resolution (3.85 lines / mm).
2. K = 4 at high resolution (7.7 lines / mm).

Generally, document images have strong vertical correlation,
When the encoding is performed by the MR method, the ratio of the vertical mode, that is, the deviation of the pixel position in the scanning line direction between the changed pixels on the reference scanning line and the changed pixels on the encoded scanning line in the scanning line direction is ± 3 pixels. Is very high. For example, in CCITT test documents (high resolution), vertical mode averages 89%.

Further, the MMR method which is assumed to be used in a data network such as ISDN having a transmission error correction function is basically the same as the above-mentioned MR method, but differs in the following points.

(1) All scanning lines are two-dimensionally encoded (MR encoded). That is, a scan line encoded by the MH method is unnecessary to prevent error propagation, and the value of the K parameter is ∞.

(2) A virtual scanning line in which all pixels are white is set as a reference scanning line when encoding the first scanning line.

(3) In a G4 facsimile apparatus to which the MMR method is applied, all received data is once stored in a memory, and then a hard copy is taken. Therefore, a time for guaranteeing a scanning line synchronization signal (EOL) and a recording time on a receiving side. The fill signal is unnecessary.

(4) Two consecutive EOLs to indicate the end of the page
EOFB (end of facsimile block) is used.

(5) If it is necessary to set a certain block size, use EOFB
After that, insert the required number of pad bids “0”.

[Problems to be Solved by the Invention] However, since such an encoding method is a known method, even if the data is recorded in a compressed form or is being transmitted in a compressed form, the encoding method can be easily performed. There was a problem of being decrypted. So far, there has been proposed a method that employs a scramble encryption method by transposition of scanning lines and pixels to prevent leakage or eavesdropping on a line between facsimile terminals. Since it is restored to a form that can be recognized by anyone, there is no concealment effect for a third party at the receiving terminal, for example, an intermediary such as a document clerk. Even within the same company or group, documents are sent and received on a daily basis that should not be taken carelessly, but no consideration has been given to such issues. That is, in the conventional encoding / decoding method, it can be said that information security is hardly considered.

In addition, a method of encrypting data using a key or the like which is a multi-digit number has also been proposed, but since it is apparent at first glance that the data is encrypted, it is sufficiently considered that the data may be decrypted. It is not a realistic solution. Further, a method of combining the data to be concealed with other data (dummy data) and encrypting the data is also conceivable, but in this case, the data capacity becomes large, and the data to be concealed or the dummy data is not used. However, there is a compatibility problem that decoding cannot be performed by a normal method.

The present invention solves the above-described problems, and realizes improvement of data recording or transmission density by combining image data with other data within a frame of a normal MR system or MMR system, and concealment of data accompanying this. The purpose is to do.

Configuration of the Invention The configuration of the present invention that achieves the above object will be described below.

[Means for Solving the Problems] The following inventions of claims 1 and 2 are provided for combining other data with image data or separating the data. First, the encoding method of image data is based on image data obtained by binarizing each pixel of an original image, from an encoding scan line to be encoded and a reference scan line immediately before the encoding scan line. A change pixel at the head of an area where pixels of the same value are continuous on a scan line is extracted, and the change pixels on the reference scan line and the change pixels on the coded scan line have the same value. When a changed pixel on the coded scanning line that satisfies the condition that the displacement of the pixel position in the direction is within a predetermined number of pixels is detected, the position of the detected changed pixel on the coded scanning line is determined by the displacement. The two-dimensional encoding method of encoding with a code representing, or the one-dimensional encoding method of sequentially detecting the run length of the pixels on the scanning line and encoding the run length, and performing the image, Compress data In the image data encoding method, in the case of the two-dimensional encoding method, every time a changed pixel on the encoded scanning line that satisfies the condition is detected, another binary data to be combined with the image data is generated. Reading, according to the read binary data, when the binary data is the first value, the shift is within the predetermined number of pixels and is even, and the binary data is equal to the first value. By moving the detected changed pixel within one pixel on the coded scanning line so that the shift is within the predetermined number of pixels and odd when the second value is the opposite, Combining the read binary data with the displacement of the pixel position, and encoding the position of the changed pixel on the detected encoded scan line by a code representing the displacement after combining the binary data, The one-dimensional mark If by the method, in each time of detecting the run length,
The other binary data to be combined with the image data is read, and according to the read binary data, the run length becomes an even number when the binary data is a first value,
When the binary data is a second value opposite to the first value, the detected run length is corrected within one pixel so that the run length becomes an odd number. The read binary data is combined with a run length, and the run length after combining the binary data is encoded.

Also, whether the data encoded in this way is encoded by the two-dimensional encoding method according to claim 1,
Determine whether the one-dimensional encoding method has been encoded, according to the determination, if the two-dimensional encoding method, sequentially extract the code representing the shift from the image data, By decoding the extracted code and restoring the arrangement of pixels on each scanning line, an image is restored. In the case of the one-dimensional encoding method, image data is decoded and the run length of the pixel is reduced. In the method for decoding image data for extracting and restoring an image in accordance with the extracted run length, if it is determined that the encoding method is based on the two-dimensional encoding method, a neighboring pixel after restoring the arrangement of pixels is determined. For each of the two scanning lines, a changed pixel at the beginning of a region where pixels of the same value are continuous on the scanning line is extracted, and a changed pixel on one scanning line and a changed pixel on the other scanning line are extracted. Of the same value If the deviation of the pixel position in the scanning line direction between them is within the predetermined number of pixels and is an even number, it is determined that the other binary data as the first value is combined with the deviation, If the pixel position shift is within the predetermined number of pixels and is an odd number, it is determined that another binary data that is the second value is combined with the shift, and the other data is determined from the image data. To separate the binary data of
On the other hand, when it is determined that the run length is determined by the one-dimensional encoding method, if the extracted run length is an even number, it is determined that binary data is combined with the run length in addition to the first value. If the extracted run length is an odd number, it is determined that another binary data as the second value is combined with the run length, and the other binary data is determined from the image data. Is separated.

Further, the facsimile apparatus according to claims 3 and 4 is configured as follows as a suitable apparatus for directly executing the above-described encoding method and decoding method.

Reading means for reading an image written on a document by binarizing the image data; and a coding scan line to be coded and a reference immediately before the coding scan line based on the image data read by the reading means. For each of the scanning lines, a changed pixel at the head of an area where pixels of the same value are continuous on the scanning line is extracted, and a changed pixel on the reference scanning line and a changed pixel on the encoded scanning line have the same value. When a changed pixel on the coded scanning line that satisfies the condition that the pixel position shift in the scanning line direction between the coded scanning lines is within a predetermined number of pixels is detected, A two-dimensional encoding method for encoding the position with a code representing the shift, or run-length detection for sequentially detecting the run-length of pixels on a scanning line from image data read by binarization by the reading means A stage, encoding means for compressing the image data by a one-dimensional encoding method for encoding the run length detected by the run length detecting means, and a predetermined procedure for compressing the image data compressed by the encoding means. A facsimile apparatus comprising: a transmitting unit that outputs the binary data to the image data; a combined data setting unit that prepares another binary data to be combined with the image data; Every time a changing pixel on the coded scanning line that satisfies the condition is detected by the converting unit, the binary data prepared by the composite data setting unit is read out, and the binary data is read out according to the read binary data. When the data is a first value, the deviation is within the predetermined number of pixels and becomes even, and the binary data is a second value opposite to the first value. By moving the changed pixel detected by the encoding means within one pixel on the encoded scanning line so that the deviation is within the predetermined number of pixels and an odd number, the deviation of the pixel position is obtained. Synthesizing means for synthesizing the read binary data, wherein the encoding means includes a code representing the shift after the binary data is synthesized by the synthesizing means, on the detected encoded scan line. In the case of using the one-dimensional encoding method, every time a run length is detected by the run length detecting means, the binary data prepared by the composite data setting means is encoded. Reading, in accordance with the read binary data, when the binary data is the first value, the run length becomes an even number, and the binary data is the second value opposite to the first value. The read binary data is combined with the detected run length by correcting the detected run length within one pixel so that the run length becomes an odd number when the value is 2. Run length correction means, wherein the encoding means is configured to encode a run length after the binary data has been combined by the run length correction means.

On the other hand, the facsimile apparatus on the side receiving the information sent from the facsimile apparatus to be operated receives the image data transmitted from the facsimile apparatus according to claim 3 via a communication line, and Determine whether the received data is coded by a two-dimensional coding method or coded by a one-dimensional coding method. If the determination is a two-dimensional coding method, Restoring means for sequentially extracting codes representing the shifts from the data, decoding the extracted codes, and restoring the arrangement of pixels on each scanning line, and when the determination is a one-dimensional encoding method, A run-length extracting means for decoding the image data and extracting a run length of pixels; and reproducing an image in accordance with the arrangement of pixels restored by the restoring means, or An image reproducing unit that reproduces an image in accordance with the run length extracted by the length extracting unit; and a facsimile apparatus including: when the encoding method is determined to be a two-dimensional encoding method, the restoration unit For each of the two adjacent scanning lines in which the arrangement of the pixels has been restored, the changed pixel at the head of the area where pixels of the same value are continuous on the scanning line is extracted, and the changed pixel on one scanning line and the other With the changing pixel on the scanning line,
If the deviation of the pixel position in the scanning line direction between those having the same value is within the predetermined number of pixels and is an even number, the other binary data of the first value is combined with the deviation. If the shift of the pixel position is within the predetermined number of pixels and is an odd number, it is determined that another binary data that is the second value is synthesized with the shift, and the image Separating the other binary data from the data, on the other hand, if it is determined that the one-dimensional encoding method, if the run length extracted by the run length extraction means is an even number, the run length It is determined that the other binary data of the first value is combined, and if the extracted run length is an odd number, the other binary data of the second value is combined with the run length. Judge that
The image processing apparatus further includes a data separating unit that separates the other binary data from the image data, and a data output unit that outputs the binary data separated by the data separating unit.

[Operation] According to the image data encoding / decoding method of the present invention having the above configuration, the image data is handled as follows.

In the method for encoding image data according to claim 1 of the present invention, an encoding scan line to be encoded and an encoding scan line from the image data obtained by binarizing each pixel of the original image by the two-dimensional encoding method are used. For each of the reference scan lines immediately before the coded scan line, extract the changed pixel at the beginning of the area where pixels of the same value are continuous on the scan line, and change the changed pixel on the reference scan line and the coded scan line. When a changed pixel on the coded scanning line that satisfies the condition that the pixel position shift in the scanning line direction between the changed pixels having the same value is within a predetermined number of pixels, the detected coded scan is Image data is compressed by encoding the position of the changed pixel on the line with the code representing the shift.

That is, image data is compressed by a procedure such as the above-described vertical mode of the modified read (MR) method. Further, in the case of this two-dimensional encoding method, a changing pixel on an encoded scanning line satisfying the above condition is used. Each time is detected, other binary data to be combined with the image data is read out, and the binary data is combined with the image data by the following procedure.

That is, in accordance with the read binary data, when the binary data is the first value, the shift is within the predetermined number of pixels and is even, and the read binary data is the first value. By moving the detected changed pixel within one pixel on the coded scanning line so that the deviation is within the predetermined number of pixels and becomes an odd number when the second value is opposite to the second value, The read binary data is combined with the displacement of the pixel position. Then, by coding the position of the changed pixel on the detected coded scanning line by the code representing the shift after the combination of the binary data as described above,
When encoding and compressing the original image data, other binary data is synthesized.

On the other hand, a so-called modified Huffman, which sequentially detects run lengths of pixels on a scanning line from image data obtained by binarizing each pixel of an original image by a one-dimensional encoding method and encodes the detected run lengths, is called a modified Huffman. The image data is compressed by the (MH) method. In the case of this one-dimensional encoding method, every time a run length to be encoded is detected from the image data, another binary data to be synthesized with the image data is generated. The binary data is read out and combined with the image data by the following procedure.

That is, according to the read binary data, when the binary data is the first value, the run length becomes an even number, and when the read binary data is the second value opposite to the first value. In this case, the read binary data is combined with the detected run length by correcting the detected run length within one pixel so that the run length becomes an odd number. Then, by encoding the run length after synthesizing the binary data in this way,
When encoding and compressing the original image data, other binary data is synthesized.

For example, when the binary data to be combined is “1” as the second value, if the run length is an even number, the run length is shortened by the value 1 and its parallelism (ie, , Even or odd) to odd,
If the run length is odd, the run length is left unchanged. On the other hand, when the binary data to be synthesized is “0” as the first value, if the run length is even, the run length is left unchanged without modification, and if the run length is odd, the run length is lengthened by one. To make the oddity even.

Next, the combined binary data can be separated from the image data in which the binary data has been combined in the above procedure by the image data decoding method according to the second aspect.

That is, in the decoding method according to claim 2, claim 1
In the field encoded by the two-dimensional encoding method according to the, sequentially extract the code representing the shift from the image data,
An image is restored by decoding the extracted code and restoring the arrangement of pixels on each scanning line. In the case of this two-dimensional encoding method, an adjacent pixel after restoring the arrangement of pixels is restored. A change pixel is extracted for each of the two scan lines, and the change in the pixel position in the scan line direction between the change pixels on one scan line and the change pixels on the other scan line having the same value is the predetermined pixel. If it is within several minutes and it is an even number, it is determined that another binary data that is the first value is combined with the deviation, and the deviation of the pixel position is within the predetermined number of pixels and is an odd number. If there is, it is determined that another binary data, which is the second value, has been combined with the deviation, and the other binary data is separated from the image data.

On the other hand, in the case of using the one-dimensional encoding method, the image data encoded by the one-dimensional encoding method according to claim 1 is decoded, a run length of a pixel is extracted, and an image is formed in accordance with the extracted run length. Is restored,
In the case of this one-dimensional encoding method, every time a run length is extracted, other binary data is separated by the following procedure.

That is, if the extracted run length is an even number, it is determined that another binary data that is the first value is combined with the run length, and if the extracted run length is an odd number, the run length is determined to be the same. The other binary data, which is the second value, is determined to be synthesized, and the other binary data is sequentially separated from the image data.

For example, when binary data is combined with image data by the method of the above-described specific example, if the extracted run length is an even number, a value of “0 =
It is determined that the binary data having the “first value” is synthesized, and if the extracted run length is an odd number, the binary data having the value “1 = second value” is synthesized with the run length. That is, the other binary data can be sequentially separated from the image data.

The encoding method according to claim 1 is directly implemented.
In the facsimile apparatus described in the above, the reading means binarizes the image written on the document and reads it, and the encoding means converts the image data read in binary form by the reading means to a two-dimensional encoding method. In this case, a changed pixel is extracted for each of the encoded scan line to be encoded and the reference scan line immediately before the encoded scan line, and a changed pixel on the reference scan line and a changed pixel on the encoded scan line are extracted. When a change pixel on the coded scan line that satisfies the condition that the pixel position shift in the scan line direction between those having the same value is within a predetermined number of pixels is detected, the detected change on the coded scan line is detected. The image data is compressed by encoding the position of the pixel with a code representing the shift. Then, the transmitting means outputs the image data compressed by the encoding means to the communication line in a predetermined procedure.

In this case, in particular, in the facsimile apparatus according to the third aspect, the combined data setting means prepares another binary data to be combined with the image data, and the combining means performs the encoding scan satisfying the condition by the encoding means. Each time a changed pixel on the line is detected, the binary data prepared by the combined data setting means is read out. If the binary data is the first value according to the read binary data, the shift is performed. Is an even number within the predetermined number of pixels, and the binary data is
The changed pixel detected by the encoding means is not more than one pixel on the encoded scanning line so that the deviation is within the predetermined number of pixels and is odd when the value is a second value opposite to the value of And the read binary data is combined with the displacement of the pixel position. Then, the encoding unit encodes the position of the changed pixel on the detected encoded scanning line by using a code representing a shift after the binary data is combined by the combining unit.

In other words, in the facsimile apparatus according to the third aspect, when the synthesizing means and the encoding means function as described above, other binary data is synthesized when the image data is encoded and compressed.

In the case of the one-dimensional encoding method, the run length detecting means sequentially detects the run lengths of the pixels on the scanning line from the image data read by binarization by the reading means. Then, the encoding unit encodes the run length detected by the run length detection unit to compress the image data, and the transmission unit outputs the image data compressed by the encoding unit to the communication line in a predetermined procedure. .

Here, in particular, in the facsimile apparatus according to claim 3, the combined data setting means prepares another binary data to be combined with the image data, and in the case of the one-dimensional encoding method,
Every time the run length is detected by the run length detecting means, the run length correcting means reads the binary data prepared by the synthetic data setting means, and in accordance with the read binary data, reads the binary data. The run length is determined such that the run length is an even number when the value is 1, and the run length is an odd number when the binary data is a second value opposite to the first value. By correcting the length within one pixel, the read binary data is combined with the detected run length. Then, the encoding means encodes the run length after the binary data has been synthesized by the run length correction means.

In other words, in the facsimile apparatus according to the third aspect, when the run-length correcting means and the encoding means function as described above, other binary data is synthesized when encoding and compressing the image data. .

Further, in the facsimile apparatus according to claim 4 for directly executing the encoding method according to claim 2, the receiving means transmits image data transmitted from the facsimile apparatus according to claim 3 via a communication line. In the case of the two-dimensional encoding method, the restoring unit sequentially extracts codes representing the shift from the image data received by the receiving unit, decodes the extracted codes, and decodes the pixels on each scanning line. Is restored. Then, the image reproducing means reproduces the image in accordance with the arrangement of the pixels restored by the restoring means.

Here, in particular, in the facsimile apparatus according to the fourth aspect, in the case of the two-dimensional encoding method, the data separating unit includes:
A changing pixel is extracted for each of two adjacent scanning lines in which the arrangement of pixels is restored by the restoring unit, and scanning is performed between pixels having the same value between a changing pixel on one scanning line and a changing pixel on the other scanning line. If the displacement of the pixel position in the line direction is within the predetermined number of pixels and is even, it is determined that another binary data that is the first value is combined with the displacement, and the displacement of the pixel position is determined. Is within the predetermined number of pixels and if it is an odd number, it is determined that another binary data, which is the second value, is combined with the shift, and other binary data is received from the received image data. To separate. And the output means,
The binary data separated by the data separating means is output.

That is, in the facsimile apparatus according to the fourth aspect, in the case of the two-dimensional encoding method, when the data separating means functions as described above, when the image data is decoded and the image is reproduced, the image data is reproduced. From the other binary data.

In the case of the one-dimensional encoding method, the run length extracting means decodes the image data received by the receiving means and extracts the run length of the pixel. Then, the image reproducing means reproduces the image according to the run length extracted by the run length extracting means.

Here, in particular, in the facsimile apparatus according to the fourth aspect, in the case of the one-dimensional encoding method, the data separating unit includes:
If the run length extracted by the run length extracting means is an even number, it is determined that another binary data, which is the first value, is combined with the run length, and the run length extracted by the run length extracting means is determined. Is odd,
It is determined that another binary data, which is the second value, has been combined with the run length, and the other binary data is separated from the received image data. Then, the output means outputs the binary data separated by the data separation means.

That is, in the facsimile apparatus according to the fourth aspect, in the case of the two-dimensional encoding method, when the data separating means functions as described above, when the image data is decoded and the image is reproduced, the image data is reproduced. Are sequentially separated from the other binary data.

[Embodiment] In order to further clarify the configuration and operation of the present invention described above, a method for encoding and decoding image data according to the present invention and general characteristics of a G4 facsimile apparatus which directly implements the method will be described below. A preferred embodiment of a facsimile apparatus having an encoding system and an encoding control function will be described. FIG. 1 is a schematic configuration diagram of two facsimile apparatuses as an embodiment connected via a communication line. 2
The two facsimile apparatuses have both functions of G4 facsimile transmission and reception and have the same internal configuration. Therefore, here, the configuration of the facsimile apparatus 1 described as a transmission side will be mainly described.

In the facsimile apparatus 1, an image scanner 5 for inputting an image and a printer 7 for outputting an image are connected to a computer 3 having a terminal 2, and a board type G4 facsimile modem 10 is incorporated. Inside the computer 3, in addition to the well-known CPU 11, ROM 12, and RAM 13, a terminal controller 15 that controls input and output of data with the terminal 2, a scanner input port 17 that is connected to the image scanner 5 and inputs the data, and a printer 7. A printer control port 18 that is connected and outputs data is provided. The configuration of the facsimile machine 21 serving as the receiving side is the same as that of the facsimile machine 1, and the terminal 22, the computer 23, the image scanner 25, the printer 27, and the G4 facsimile modem 30
Are provided as well. Both facsimile machines 1, 21
It is connected to a public communication line 40 via built-in G4 facsimile modems 10,30.

Next, data combining / outputting processing and data receiving / separating processing in both facsimile apparatuses 1 and 21 will be sequentially described. FIG. 2 is a flowchart showing a processing routine executed by the facsimile machine 1 on the transmitting side, and FIGS. 3 to 5 are flowcharts showing details of the data synthesizing process in the processing routine. In the following description, a0, a1, a2, b1, and b2 are white or black pixels in the coded scanning line and the reference scanning line as described in the section of [Prior Art] shown in FIG. Indicate the pixel (changed pixel) at the position where the sequence is inverted.

The facsimile machine 1 on the transmitting side includes an image scanner 5
The original 51 to be transmitted is read, and the image data is expanded in the RAM 13 in advance. The image data of the document 51 is developed for each scanning line. The data to be combined with the data is input from the terminal 2 in the form of a character string, and is developed and stored in the RAM 13 in the computer 3.

When the transmission-side facsimile apparatus 1 is connected to the reception-side facsimile apparatus 21 via the communication line 40 by a signal sequence conforming to the protocol of the G4 device, the facsimile apparatus 1 thereafter generates facsimile encoded information of a predetermined block size. Shift to data synthesis and encoding processing. In the data synthesis and encoding processing, first, a virtual scanning line (one white reference scanning line) in which all pixels are white is set as a reference scanning line when encoding the first scanning line in the MMR method. (Step 100, hereinafter, the step is simply referred to as S), and the data of the coded scanning line to be coded next is read (S110).
Next, a process of setting a0 immediately before the first pixel of the encoded scan line is performed (S120). Next, as described in the section of [Prior Art], continuous white or black pixels on a coded scan line and a reference scan line where pixels change (referred to as change pixels)
Processing for detecting a1, b1, and b2 is performed (S130). See FIG. 18 for the definition of each pixel.

With respect to the pixels thus detected, it is determined whether the pixel b2 is on the left side of the pixel a1 (S140). If the pixel b2 is on the left side, the pixel corresponds to the pass mode.
The process of placing the pixel a0 immediately below the pixel b2 is performed for the next encoding (S160), and the process proceeds to the detection process of the paired pixels a1, b1, and b2 in S130. An example of the arrangement of the pixels corresponding to the pass mode and the movement of the pixel a0 are as shown in FIG. 19 (A).

On the other hand, if the pixel b2 is not to the left of the pixel a1 (S14
0), after detecting the changed pixel a2 in preparation for the processing described later (S170), it is determined whether or not the distance | Δ (a1b1) | between the pixel a1 and the pixel b1 is within the value 3 (S180). ).

If the distance | Δ (a1b1) | between the pixels a1 and b1 is within the value 3, it corresponds to the vertical mode as exemplified in FIG. 19B, and therefore, it is assumed that the data is synthesized in the vertical mode. Then, it is determined whether or not combination is possible, and if possible, the data Dm (i) is combined (S190). The determination of whether or not data combination is possible and details of the combination will be described later with reference to FIG. After combining the data in the vertical mode in this way, the data is encoded in the vertical mode (S200), and the pixel a0 is placed at the position of the pixel a1 in preparation for the next encoding (S210).

On the other hand, if the distance | Δ (a1b1) | between the pixels a1 and b1 is larger than 3, it corresponds to the horizontal mode as illustrated in FIG. 19 (C). First, the data synthesis in the first run length (hereinafter simply referred to as the first RL), that is, the first run length
Judgment as to whether synthesis is possible with RL and data Dm when possible
The combining processing of (i) is performed (S220). The determination of whether or not the first RL can be combined and the details of the combination will be described later with reference to FIG. After combining the data in the first RL in this way, the data is encoded in the horizontal mode (S230), and the next second
Pixel a0 in preparation for the encoding of the run length (2nd RL)
Is placed at the position of the pixel a1 (S240), and a process of detecting the changed pixel a2 is performed (S250).

After the detection of the changed pixel a2, data synthesis in the second RL is performed in the same order as in S220 to S240 described above (S260 described later with reference to FIG. 5 in detail), encoding in the horizontal mode (S270), and pixel
Each process of arranging a0 at the position of pixel a1 (S280) is performed.
Vertical mode (S190-S210), horizontal mode (S220-S280)
After the encoding is performed in each of the steps, it is determined whether or not the encoding of one line has been completed (S290). If the encoding has not been completed for one line, the process returns to S130 described above, and the setting after encoding is performed. The process is repeated from the process of newly detecting the pixels a1, b1, and b2 based on the pixel a0 thus obtained. On the other hand, if the encoding has been completed up to the end of the encoded scan line (S290), it is determined whether or not the end of the facsimile block has been reached (S300). Is performed (S310), and the process is repeated from reading of the next encoded scan line (S110).

If the end of the facsimile block is determined, the facsimile block end code (EOFB) is added to the end of the facsimile information encoded by the above-described processing (S320), and pad bits are added (S33).
0), the data synthesis / encoding processing routine ends.
The facsimile coded information of a predetermined block size thus coded is transmitted to the facsimile apparatus 21 on the receiving side by a predetermined signal sequence and is decoded.

The data combining and encoding processes have been briefly described above. Next, details of the data combining will be described. FIG. 3 shows details of the data synthesizing process in the vertical mode in S190 shown in FIG. In step 190, the pixel a of the encoded scan line
It is determined whether the distance Δ (a1b1) from 1 to the pixel b1 of the reference scanning line is greater than 0 (S400). In order to avoid the inconvenience of description, only Δ (a1b1) is simply indicated as Δ in the following description. There are certain restrictions between the pixels in order to perform data synthesis after the separation between the two pixels depending on whether or not the distance is greater than 0. This is due to the fact that data may not be able to be restored by combining 1-bit data depending on the arrangement of pixels. Therefore, in each case, it is determined whether or not the condition is satisfied. Details of when data cannot be restored will be described later.

The determination as to whether or not restoration can be made depends on whether or not the gap Δ is greater than 0. First, it is determined whether or not the gap Δ is greater than 0. Judge whether or not it is satisfied (S410-45
0).

Judgment of conditions: If the previous mode is the horizontal mode (S410), Δ
If (b0a1) ≧ 2 (S420) and Δ (b1a2) ≧ 1 (S430), it is determined that the condition is satisfied, and then data synthesis is performed.

If the previous mode is not the horizontal mode, that is, if the previous mode is the pass mode or the vertical mode (S4
10), Δ (b0a1) ≧ 1 (S440), and Δ (b1a2) ≧ 1
If it is (S450), it is determined that the condition is satisfied.

If it is determined that the condition is satisfied, it is determined whether or not the gap Δ is even (S460), and the character strings Dm to be combined are sequentially read as shown below, and
The pixel a1 is moved according to the bit (hereinafter referred to as bit Dm (i)).

When Δ is an even number, if the bit Dm (i) is a value of 1 (S4
70), a process of moving the pixel a1 to the right by one pixel is performed (S4).
80) If Δ is an odd number and bit Dm (i) is value 0 (S490), a process of moving pixel a1 right by one pixel is performed (S500). If the condition is not satisfied (S410-45
0), the processing for synthesizing the data is not performed.

As a result, as shown in FIGS. 6 (A1), (A2) and (A3), when Δ> 0 and the number is even,
As shown in (B2) and (B3), if Δ> 0 and the number is odd, the bit Dm (i) is inserted. If the value of the bit Dm (i) is 1, the gap Δ becomes odd and the bit Dm (i) has the value 0
In this case, the gap Δ is set to an even number.

As described above, when Δ> 0 in the vertical mode, the determination as to whether or not data synthesis is possible is made based on whether or not a condition is satisfied. This is set for the following reason. is there.

Conditions: (i) When Δ = 1 and RL (a1a2) = 1 (Fig. 7 (A1))
If the pixel a1 is moved to the right by the combination, RL (a1a2) disappears, and the combination cannot be performed (FIG. 7 (A2)). Therefore, Δ = 1
In this case, RL (a1a2) ≧ 2 must be satisfied. Next, Δ =
When RL (a1a2) = 2 in FIG. 2 (FIG. 7 (B1)), if pixel a1 is moved to the right by synthesis, RL (a1a
2) = 1 (FIG. 7 (B2)), and RL (a1a2) ≧ 2 is not satisfied. Therefore, when Δ = 2, RL (a1a2) ≧ 3 must be satisfied. Similarly, when Δ = 3, RL (a1a2) ≧ 4
Must. To summarize the above, the condition under which data synthesis is possible (hereinafter simply referred to as a constraint condition) is Δ (b1a2)
≧ 1.

(Ii) When Δ> 0 and Δ (b0a1) = 0 (FIG. 7 (C1))
When the pixel a1 is moved to the right by synthesis, Δ (a1 ′
If (b1 ')> 0, Δ (b1'a2') = 1 and a false bit is output (FIG. 7 (C2). Therefore, Δ (b0a1) ≧ 1 must be satisfied).

(Iii) When the previous mode is the horizontal mode and Δ (b0a1) = 1 (FIG. 7 (D1)), by moving the pixel a1 to the right by the combination, RL (a0′a1 ′) is obtained when decoding in the horizontal mode. If is an even number, a false bit is output with RL (a1'a2 ') = 2. (7th
(D2). Therefore, Δ (b0a1) ≧ 2 must be satisfied.

To summarize (i) to (iii), the constraint condition when Δ> 0, that is, the condition is Δ (b0a1) ≧ 1 and Δ (b1a2) ≧
One. However, when the previous mode is the horizontal mode, Δ (b0
a1) ≧ 2 and Δ (b1a2) ≧ 1.

Next, a description will be given of the determination of the constraint conditions and the data synthesis when Δ ≦ 0 is determined in S400 of FIG.

If it is determined that Δ ≦ 0, it is first determined whether or not a condition that is a constraint is satisfied (S510 to S530).

Judgment of conditions: If the previous mode is the horizontal mode (S510), Δ
If (b0a1) ≧ −Δ (a1b1) +3 (S520), it is determined that the condition is satisfied. If the previous mode is not the horizontal mode, Δ (b0a1) ≧ −Δ (a1b1) +2. In some cases (S530), it is determined that the conditions are satisfied. If the condition is not satisfied (S510 to S530), the process of synthesizing the data is not performed.

If it is determined that the condition is satisfied, it is determined whether the gap Δ is an even number (S540).
If Dm (i) is the value 1 (S550), a process of moving pixel a1 one pixel to the left is performed (S560). On the other hand, when the gap Δ is an odd number, if the bit Dm (i) is a value 0 (S57)
0), the pixel a1 is moved one pixel to the left (S580).

As a result, as shown in FIGS. 8 (A1), (A2) and (A3), when Δ ≦ 0 and the number is even,
As shown in (B2) and (B3), when Δ ≦ 0 and an odd number, if the bit Dm (i) is a value “1”, the gap Δ is an odd number and the bit Dm (i) is a value. In the case of “0”, the gap Δ is set to an even number.

Hereinafter, the constraint conditions used in the above-described processing and the processing described below are listed.

Condition: Δ (b0a1) ≧ −Δ (a1b1) +2 when Δ ≦ 0 in the vertical mode. However, when the previous mode is the horizontal mode, Δ (b0a1) ≧ −Δ (a1b1) +
3.

Condition: When RL (a0a1) is decoded in the horizontal mode When RL (a0a1) is an even number When encoding the first RL, RL (a1a2) ≠ 2 and Δ ≠ 4
And Δ (b0a1) ≠ 0 and Δ (a1b2) ≠ 0.

Condition: In horizontal mode, when RL (a0a1) is an even number When encoding the second RL, RL (a1a2) ≧ 2.

Condition: When RL (a0a1) is odd in horizontal mode When encoding the first RL, RL (a0a1) ≧ 1 and Δ (b0a
1) and Δ (a1b1) ≠ -4.

Condition: RL (a0a1) ≠ 1 when encoding the second RL when RL (a0a1) is an odd number in the horizontal mode.

The data composition in the vertical mode has been described above. Next, the data composition in the horizontal mode will be described. The details of the data synthesis process in the first RL of S220 shown in FIG.
Shown in the figure. In step 220, first, it is determined whether or not the run length (RL (a0a1)) of the pixels a0 to a1 of the encoded scan line is an even number in order to determine the constraint condition (FIG. 4, S60).
0) If the number is even, the condition (S610 to S640) is determined; if the number is odd, the condition (S670 to S690) is determined. In the conditions, RL (a1a2) ≧ 2 (S610), Δ (a1b1) ≠ 4 (S62
0), Δ (b0a1) ≠ 0 (S630), Δ (a1b2) ≠ 0 (S64
It is determined whether or not all of 0) are satisfied. If the condition is satisfied and the bit Dm (i) has the value 1, (S6
50), the pixel a1 is moved right by one pixel (S660).

On the other hand, if RL (a0a1) is an odd number, the following condition is determined. In the conditions, RL (a0a1) ≠ 1 (S670), Δ
It is determined whether all of (b0a1) ≠ 1 (680) and Δ (a1b1) （-4 (S690) are satisfied. If the condition is satisfied and the bit Dm (i) is 0 (S700), the pixel a1 is moved one pixel to the left (S710).

As a result, as shown in FIGS. 9 (A1), (A2) and (A3), when the run length is an even number, FIGS. 9 (B1), (B2),
As shown in (B3), when the run length is odd,
When Dm (i) has a value of 1, the run length is set to an odd number, and when bit Dm (i) has a value of 0, the run length is set to an even number.

After completion of the data synthesis processing for the first run length in the horizontal mode (S220), the processing of S230 to S250 is performed, and then the data synthesis processing for the second run length (S260) is performed. In S260, it is first determined whether or not RL (a0a1) is an even number (FIG. 5, S800). If the judgment is even, the judgment of the condition (S810) is performed next.
If it is an odd number, the condition (S840) is determined next. In the condition, it is determined whether or not RL (a1a2) ≧ 2 (S8).
Ten).

If the condition is satisfied and the bit Dm (i) is the value 1 (S820), the pixel a1 is moved right by one pixel (S830).
On the other hand, in the condition, it is determined whether or not RL (a0a1) ≠ 1 (S840). If the condition is satisfied and the bit Dm (i) is 0 (S850), the pixel a1 is moved one pixel to the left (S860).

As a result, when the bit Dm (i) is a value of 1, the run length is set to an odd number, and when the bit Dm (i) is a value of “0”, the run length is set to an even number. If the composition is performed in consideration of the constraint conditions (conditions to conditions) at the time of MMR encoding, the arrangement of pixels will not be changed at all and confusion as to whether or not data will be combined will not occur. Data series can be correctly extracted.

Next, a process of the facsimile machine 21 that receives the image data generated by executing the above-described data synthesis process and then output via the communication line 40 will be described. FIG. 10 is a flowchart showing a data separation processing routine executed by the facsimile machine 21 for reception.

When the receiving facsimile device 21 is connected to the transmitting facsimile device 1 via the communication line 40 by a predetermined signal sequence in accordance with the protocol of the G4 device, the received facsimile encoded information is received, and The image data is extracted, developed and stored in the RAM in the computer 23. In this state, when the data separation processing routine of FIG. 10 is started by the computer 23, first, processing of storing one line of white image data in the array A is performed (S1000), and then one line of image data to be decoded is processed. A process of reading, decoding, and storing in array B is performed (S111
0). Next, a process of setting a0 immediately before the first pixel of one line (hereinafter simply referred to as a decoded scanning line) stored in the array B is performed (S1120). Next, the changed pixels a1, b1, and b2 are detected (S1130), and it is determined whether or not the pixel b2 is on the left side of the pixel a1 (S1140). Since the data is coded, the process in the pass mode, that is, the process of placing the pixel a0 immediately below the pixel b2 is performed without performing data separation (S1150).

On the other hand, if the pixel b2 is not to the left of the pixel a1 (S114
0), after detecting the changed pixel a2 in preparation for the processing to be described later (S1160), it is determined whether or not the distance | Δ (a1b1) | is within a value 3 (S1170).

If the distance | Δ (a1b1) | is less than or equal to 3, it corresponds to the vertical mode, and data separation in the vertical mode is performed (S1180). For more information on this data separation, see
It will be described later with reference to FIG. After the data separation in the vertical mode in this manner, a process of placing the pair of pixels a0 on the pixels a1 is performed (S1190).

If the distance | Δ (a1b1) | between the pixels a1 and b1 is larger than the value 3, it corresponds to the horizontal mode.
(S1200). The details will be described later with reference to FIG. After the data separation in the first RL is performed, the pixel a0 is placed on the pixel a1 in preparation for the decoding of the second RL (S1210), and then the changed pixel a2 is detected on the decoded scanning line (S1220).

After detecting the changed pixel a2, data separation is performed in the second RL (S1230). The details will be described later with reference to FIG. After performing the data separation in the second RL in this manner, a process of placing the pixel a0 on the pixel a1 is performed (S124).
0).

Pass mode (S1150), vertical mode (S1180 to S119
0), after performing data separation and the like in each of the horizontal modes (S1200 to 1240), it is determined whether or not all the processing of one line has been completed (S1250). Returning to S1130, the process is repeated from the process of newly detecting the pixels a1, b1, and b2 based on the pixel a0 set after data separation or the like. On the other hand, if the processing is completed up to the end of the decoding scan line (S1250), the decoding scan line stored in the array B is changed to the array A in order to decode the next one line.
(S1260), and then determine whether a signal EOFB indicating the end of the facsimile block has been detected (S127).
0). If it is not the end of the facsimile block, the next decoding scan line is read and the process is repeated from the process of storing it in array B (S1110).

If it is determined that this is the end of the facsimile block, this routine is terminated, assuming that all the data separation processing has been completed. As described above, the data separated by this routine is temporarily stored in the RAM of the computer 23 and then recorded on the external storage medium.

The data separation processing has been briefly described above. Next, the details of the data separation will be described. S118 shown in FIG.
Details of 0 are shown in FIG. In step 1180, first, it is determined whether or not the gap Δ is greater than 0 (1300), thereby determining a constraint condition when the decoded data is encoded. If the gap Δ is larger than 0, it is determined that the constraint condition at the time of encoding depends on the condition, and the condition shown in S1310 is determined. Since the flowchart of the condition 1 to condition has already been described, the detailed description is omitted here.

If the condition is satisfied, that is, if the value of the bit Dm (i) is superimposed at the time of encoding, it is determined whether or not the gap Δ is an even number (S1320). If the gap Δ is an odd number, it is determined that the value “1” of the bit Dm (i) is superimposed, and a bit “1” is output as decoded data (S1330). On the other hand, if the gap Δ is even, the bit D
It is determined that the value 0 of m (i) is superimposed, and bit “0” is output as decoded data (S1340).

If the condition is not satisfied, it is determined that the value of bit Dm (i) has not been superimposed, and no decoded data is output.

When it is determined whether the restriction Δ at the time of encoding is performed according to the condition by determining whether the distance Δ is greater than the value 0 (S1300), the condition is determined next (S1350). If the condition is satisfied, the gap Δ
It is determined whether or not is an even number (S1360). If it is an even number, bit “0” is output as decoded data (S1370), and if it is an odd number, bit “1” is output (S1380).

As a result, in the vertical mode, when the condition or the condition is satisfied, it can be determined that the data is combined at the time of encoding, and the value of the combined bit Dm (i) is separated by the evenness of the gap Δ. And output it.

Next, details of data separation in the horizontal mode will be described with reference to FIGS.

In the data separation (S1200) in the first RL shown in FIG. 10, it is first determined whether or not RL (a0a1) is an even number (FIG. 12, S14).
00), if it is synthesized at the time of encoding, it is determined whether the value 0 is synthesized or the value 1 is synthesized. If it is an even number, it is determined that the value 0 may be synthesized, and it is determined whether the value 0 is actually synthesized by determining whether the condition is satisfied (S1410).

If the condition is satisfied, it is determined that the value 0 of the bit Dm (i) has actually been synthesized, and the bit “0” is output as decoded data (S1420).

On the other hand, when RL (a0a1) is an odd number, that is, when there is a possibility that the value 1 is synthesized, it is determined whether the condition is satisfied next (S1430). Assuming that the value 1 of the bit Dm (i) is actually synthesized, the bit “1” is output as decoded data (S144).
0).

Next, data separation (S123) in the second RL shown in FIG.
0) will be described in detail.

In the data separation in the second RL, first, it is determined whether or not RL (a0a1) is an even number (FIG. 13, S1500), and if it is synthesized at the time of encoding, the value 0 is synthesized or the value 1 is synthesized. To determine If it is an even number, there is a possibility that the value 0 has been combined, so it is determined whether or not the condition is satisfied next, thereby determining whether or not the combination has actually been made (S1510). If the condition is satisfied, bit “0” is output as decoded data (S1520).

On the other hand, if it is determined in S1500 that the number is an odd number, it is determined whether or not the condition is satisfied (S1530), and if so, a bit “1” is output (S1540).

Bit if condition or condition not satisfied
Assuming that the value of Dm (i) was not synthesized,
Move on to S1240.

As a result, the value of the bit Dm (i) combined with the first RL and the second RL in the horizontal mode is separated and output.

In the vertical mode and the horizontal mode, the separated bit data is decoded into characters for each predetermined number of bits by a character decoding routine (not shown), and is recorded as data Dm in the built-in RAM of the facsimile apparatus 21.

By the above processing, the receiving facsimile machine 21 that has received the data transmission prints the image on the document 51 by the printer 27. FIG. 14A is an explanatory diagram showing a part of a received image when only the image of the original 51 is transmitted at a high resolution of 1728 pixels × 2376 rows, and FIG. FIG. 9 is an explanatory diagram showing a part of a received image when data input from the MFP is combined and transmitted. As shown in the figure, the image is slightly rough due to the fact that the change point of the pixel is shifted by one bit due to the combination of the data, but the quality is not so much a problem. The G4 facsimile can select several types of resolution, and one dot in the horizontal direction in a facsimile image is about 0.12 mm to 0.06 mm, which is very small. As described above, since the facsimile image has a high resolution, even if one dot on the scanning line is increased or decreased, it hardly affects the human eyes. Also, as far as FIG. 14 (B) is concerned, there is no possibility that it is known that the image is combined with the data, and the confidentiality of the data is extremely high.

Using the system of the present embodiment, the results of experiments using CCITT text documents No. 1 to No. 8 as document images are shown in Table 2.
It is shown in Table 3. However, these images were input at a high resolution of 1728 pixels × 2376 rows. First, Table 2 shows MH, G
The data amount (in units of bytes) that can be combined with eight types of document images by each method of MR in 3 facsimile and MMR in G4 facsimile of this embodiment is shown. The larger the number of run-length images, such as documents NO.4 and NO.7, the larger the amount of data that can be combined. In the case of the MMR system according to the present embodiment, the amount of data that can be combined generally corresponds to 30% of the transmission code amount, and is the highest among the three systems. It is also shown that almost the same amount of data can be combined in all three systems. Next,
Table 3 shows a comparison of the transmission code amount (bit unit) between the case where the original document is encoded as it is in the three systems and the case where character data is multiplexed and encoded on the facsimile document. From this result, it is understood that the multiplexing of the character information hardly affects the transmission code amount. Also, the ratio of these code amounts is approximately MH: MR: MMR = 5: 3: 2 on average, and the compression ratio is almost constant.

Next, a process of extracting the data Dm extracted by the process shown in FIG. 10 and stored in the built-in RAM of the facsimile machine 21 for reception will be described. FIG. 15 is a flowchart showing a data restoration processing routine executed by the receiving facsimile machine 21. When this routine starts,
First, enter the password from the terminal 22 (S1600),
If the password matches the code string registered in advance, the user is further prompted to select a restoration method (S1620). If the printer is specified, the stored data Dm is sequentially read and output to the printer 27 (S1630). If the CRT is specified, the character string Dm is sequentially output to the CRT of the terminal 22. (S1640). Thereafter, it is determined whether or not the character string should be deleted (S1650). If an instruction such as deletion has been issued from the terminal 22, a process of replacing all the character strings Dm with the hexadecimal mode ＄ AA is performed (S1660). Even if the passwords do not match, the character string is similarly set to maintain the security of the information.
Performs Dm replacement processing. Since the character string Dm is not only deleted from the directory but also replaced with the code $ AA, there is no danger that data will be read out once the directory is deleted.

In this embodiment, the image scanner 5 corresponds to a reading unit, the processing of steps 100 to 160, 200, 230, and 270 in FIG. 2 corresponds to an encoding unit, and the facsimile modem 10 corresponds to a transmitting unit. The terminal 2 and the RAM 13 correspond to the synthesized data setting means, and the third data executed in step 190 of FIG.
The processing in the figure corresponds to the synthesizing means, and steps 170 and 180 correspond to the run length detecting means.

Steps 220 and 260 correspond to run-length correction means.

Further, the facsimile modem 30 corresponds to the receiving means,
The computer 23 and the printer 27 correspond to an image reproducing unit. The processing of step 1110 in FIG. 10 corresponds to the restoration means,
Steps 1200 and 1230 correspond to the run-length extracting means and the data separating means, the processing of step 1180 corresponds to the data separating means, and the RAM and the external storage medium of the computer 23 correspond to the data output means.

Encoding of data of the present embodiment configured as described above
The decoding method is extremely excellent in that data synthesis and separation can be performed extremely easily and is compatible with the highly versatile MMR method. Further, there is an excellent advantage that the amount of encoded data does not increase even by combining data. Further, according to the facsimile apparatuses 1 and 21 of the present embodiment, it is possible to easily combine and separate a large amount of data with the image data to be transmitted, and furthermore, it is possible to confirm that the data has been combined from the restored image. Since it is extremely difficult to visually recognize, there is an advantage that the confidentiality of information is extremely high. Moreover, the amount of information to be transmitted hardly increases while the data is being synthesized, and the data transmission efficiency is extremely high. Further, if the receiving-side facsimile apparatus has a conventional MMR receiving function despite the data being synthesized, there is an advantage that data can be exchanged. Therefore, although data can be synthesized and encrypted, data transmission is not limited to a special device.

Although the preferred embodiment of the present invention has been described above, it is also possible to perform image restoration to further reduce the roughness of the received image and take measures to improve the image quality.
For example, the following image restoration is performed.

In the data synthesizing method of this embodiment, the received image is slightly deteriorated due to the multiplexing of the second information source. This image quality degradation is in a practically acceptable range from the high resolution of the facsimile image, but if it is a high-performance G4 facsimile terminal having storage capacity and digital processing function, it is suitable for the properties of the composite image of this embodiment The quality of the output image can be improved by the smoothing process.

As is clear from the multiplexing method, the image quality of the composite image according to the present embodiment is deteriorated by one-dimensional horizontal noise as compared with the original image. In particular, visually, the influence of the unevenness of the black pixel is considered to be large. Therefore, the image quality is improved by removing the convex portion and filling the concave portion. Therefore, a minute area composed of 3 × 3 pixels centered on the target pixel is extracted, and the value of the target pixel is determined based on the state of eight surrounding pixels (hereinafter, referred to as reference pixels). That is,
If five consecutive pixels among the reference pixels are white as shown in FIG. 16 (A), the pixel of interest is converted to white to remove the convex portion, and conversely, five consecutive pixels as shown in FIG. 16 (B). If the pixel is black, the pixel of interest is converted to black to fill the recess. By the way, when such smoothing is performed, the reference pixels on the upper and lower two scanning lines sandwiching the target scanning line having the target pixel are:
This is an important factor in determining the pixel value of interest. If these pixel values do not match the original image due to synthesis, the target pixel value cannot be matched with the original image. Therefore, in order to make the smoothing process work effectively, information on these reference pixels is held. That is, in order to satisfy this condition, data is synthesized and transmitted only to the alternate scanning lines. On the receiving side, after separating the combined data, a smoothing process is performed on the combined scanning line. FIG. 17 shows the smoothed image thus restored. As shown in the figure, the uneven portion due to the multiplexing is deleted, and the difference from the original image shown in FIG. 14 (A) is hardly felt. As described above, the received image can be output as an image very close to the original image by the smoothing process.

Although the embodiment of the present invention has been described above, the present invention is not limited to such an embodiment. For example, a method for recording data on a magnetic recording device or as an apparatus, or encrypting and transmitting data It goes without saying that the present invention can be implemented in various modes as long as it does not depart from the gist of the present invention.

Effect of the Invention As described in detail above, encoding and encoding of image data of the present invention
According to the decoding method, the relationship between the pixel arrangement on the reference scanning line and the pixel arrangement on the encoding scanning line, and the encoding / decoding of data using run length is extremely easy. This provides an excellent effect that other data can be synthesized and separated. In addition, according to the method of the present invention, MR or the like which originally compresses image data efficiently can be used.
The MMR method has an advantage that other data can be synthesized without increasing the amount of information. That is, data can be efficiently synthesized.
As a result, the density of data recording / transmission can be further increased. In addition, it is not explicit whether the data is synthesized or not, and the advantage of excellent data confidentiality can be obtained. Further, the encoding / decoding method of the present invention can maintain higher compatibility with the encoding / decoding method of the conventional MR or MMR method.

A facsimile apparatus that directly implements the encoding method of the present invention can use the relationship between pixel shifts of image data and run length without increasing the amount of information of image data to be transmitted, and perform other data conversion. An excellent effect of being able to combine and transmit the data. Further, a facsimile apparatus that directly implements the decoding method of the present invention receives and decodes the data, and converts other data from the transmitted image data based on the relationship between pixel shifts and run length of the data. It has an excellent effect that it can be easily separated. Therefore, according to the facsimile apparatus of the present invention, it is possible to easily transmit and receive other data combined with image data with high transmission efficiency, and to realize high confidentiality of data.

In addition, the feature of this method that the amount of transmission code is not increased by combining information is that when applied to communications such as facsimile, a large amount of information can be transmitted in the same transmission time in the current line system, which is an advantage of cost reduction. Bring. Therefore, this means that data compression has been achieved without changing the existing coding system, which also leads to effective use of line resources. As information to be combined, not only a document but also other information (for example, an image or a program) can be used, so that the use form of the information medium in facsimile communication is expanded.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a facsimile apparatus for explaining an embodiment of the present invention, FIG. 2 is a flowchart showing a data synthesis encoding processing routine in the facsimile apparatus, and FIG. 3 shows details of data synthesis in a vertical mode. Flowchart, FIG. 4 is a flowchart showing details of data synthesis in the first RL, FIG. 5 is a flowchart showing details of data synthesis in the second RL, FIGS. 6 (A1), (A)
2), (A3), (B1), (B2), and (B3) are explanatory diagrams each showing a state of data synthesis, and FIG. 7 (A1), (A2), (B3).
1), (B2), (C1), (C2), (D1), and (D2) are explanatory diagrams illustrating cases where data cannot be combined,
Fig. 8 (A1), (A2), (A3), (B1), (B2), (B
3) is an explanatory diagram showing the state of data synthesis, and FIG. 9 (A
1), (A2), (A3), (B1), (B2), and (B3) are explanatory diagrams each showing a state of data synthesis, FIG. 10 is a flowchart showing a data separation processing routine, and FIG. Flow chart showing details of data separation in mode, twelfth
Fig. 13 is a flowchart showing details of data separation in the first RL, Fig. 13 is a flowchart showing details of data separation in the second RL, Figs. 14 (A) and (B) are explanatory diagrams showing examples of data synthesis, and Figs. FIG. 16 is a flowchart showing a processing routine for restoring the combined data in the receiving facsimile apparatus. FIGS. 16 (A) and (B) are explanatory diagrams of smoothing, and FIG. 17 is an explanatory diagram of an example of data smoothing. FIG. 18 is an explanatory diagram showing names of pixels on a scanning line to be handled, and FIG. 19 is an explanatory diagram showing a state of encoding of each mode in the MMR system. 1,21 ... Facsimile machine 3,23 ... Computer 5,25 ... Image scanner 7,27 ... Printer 10,30 ... Facsimile modem 40 ... Public communication line

Claims (4)

(57) [Claims] An image data obtained by binarizing each pixel of an original image, an encoding scan line to be encoded and a reference scan line immediately before the encoding scan line are the same on the scan line. Extract the change pixel at the beginning of the area where the value pixels are continuous,
On the coded scan line satisfying the condition that the pixel position shift in the scan line direction between the changed pixels on the reference scan line and the changed pixels on the coded scan line having the same value is within a predetermined number of pixels. Change pixel is detected,
The two-dimensional encoding method of encoding the position of the changed pixel on the detected encoded scan line with the code representing the deviation, or sequentially detecting the run length of the pixel on the scan line, and encoding the run length In the method of encoding image data for compressing the image data by selecting and performing one-dimensional encoding method, when the two-dimensional encoding method is used, the change on the encoded scan line satisfying the condition Each time a pixel is detected, another binary data to be combined with the image data is read out. According to the read out binary data, if the binary data is a first value, the deviation is equal to the predetermined value. Within the number of pixels and becomes even, and when the binary data is a second value opposite to the first value, the deviation is detected within the predetermined number of pixels and odd. Change pixel The read binary data is combined with the displacement of the pixel position by moving the pixel data within one pixel on the encoded scan line, and the code representing the displacement after the combination of the binary data is used. The position of the changed pixel on the detected coded scan line is coded. Reading, in accordance with the read binary data, when the binary data is a first value, the run length is an even number, and the binary data is a second value opposite to the first value. In some cases, the detected run length is corrected within one pixel so that the run length is odd, so that the read binary data is combined with the detected run length, and the binary data is read. A method for encoding image data, comprising encoding a run length after combining data. 2. It is determined whether the image is coded by the two-dimensional coding method according to claim 1 or is coded by the one-dimensional coding method.
In the case of the two-dimensional encoding method, a code representing the shift is sequentially extracted from the image data, and the extracted code is decoded to restore the arrangement of pixels on each scanning line, thereby restoring the image. In the case of the one-dimensional encoding method, in the image data decoding method of decoding image data, extracting a run length of a pixel, and restoring an image in accordance with the extracted run length, Is determined to be based on the two-dimensional encoding method, for each of the two adjacent scanning lines after the pixel arrangement has been restored, at the beginning of the area where pixels of the same value are continuous on the scanning lines. A certain change pixel is extracted, and a change pixel on one scan line and a change pixel on the other scan line are extracted.
If the deviation of the pixel position in the scanning line direction between those having the same value is within the predetermined number of pixels and is an even number, the other binary data of the first value is combined with the deviation. If the shift of the pixel position is within the predetermined number of pixels and is an odd number, it is determined that another binary data that is the second value is synthesized with the shift, and the image The other binary data is separated from the data. On the other hand, if it is determined that the data is obtained by the one-dimensional encoding method, if the extracted run length is an even number, the first value is added to the run length. It is determined that some other binary data is synthesized, and if the extracted run length is an odd number, it is determined that another binary data that is the second value is synthesized in the run length. The other binary data from the image data A method for decoding image data, comprising separating the image data. 3. A reading means for reading an image written on a document in a binarized form, an encoding scan line to be encoded and an encoding scan from the image data read in a binary form by the reading means. For each of the reference scan lines immediately before the line, the change pixel at the head of the area where pixels of the same value are continuous on the scan line is extracted, and the change pixel on the reference scan line and the change pixel on the encoded scan line are extracted. When a change pixel on the coded scanning line that satisfies the condition that the pixel position in the scanning line direction between pixels having the same value in the scanning line direction is within a predetermined number of pixels is detected, the detected coded scan is performed. The position of the change pixel on the line is
A two-dimensional encoding method for encoding with a code representing the shift, or run-length detecting means for sequentially detecting run-lengths of pixels on a scanning line from image data read by binarization by the reading means; Encoding means for compressing the image data by a one-dimensional encoding method for encoding the run length detected by the length detecting means; and outputting the image data compressed by the encoding means to a communication line in a predetermined procedure. A facsimile apparatus comprising: a combined data setting unit that prepares another binary data to be combined with the image data; and, if the two-dimensional encoding method is used, the encoding unit sets the condition. Every time a changed pixel on the encoded scanning line that satisfies is satisfied, reads out the binary data prepared by the synthetic data setting means. According to the read binary data, when the binary data is the first value, the shift is within the predetermined number of pixels and is even, and the binary data is opposite to the first value. In the case where the second value is, the changed pixel detected by the encoding means is moved within one pixel on the encoded scanning line so that the deviation is within the predetermined number of pixels and is odd. And a synthesizing unit that synthesizes the read binary data with the displacement of the pixel position. The encoding unit includes a code representing the shift after the binary data is synthesized by the synthesizing unit. By the above, the position of the changed pixel on the detected encoded scanning line is encoded. On the other hand, in the case of the one-dimensional encoding method, every time a run length is detected by the run length detecting means,
The binary data prepared by the composite data setting unit is read, and according to the read binary data, when the binary data is the first value, the run length becomes an even number, and the binary data is By correcting the detected run length within one pixel so that the run length becomes an odd number when the second value is opposite to the first value, the detected run length A run-length correcting unit that synthesizes the read binary data, wherein the encoding unit is configured to encode a run length after the binary data is synthesized by the run-length correcting unit. Facsimile machine. 4. A receiving means for receiving, via a communication line, image data transmitted from the facsimile apparatus according to claim 3, wherein the data received by said receiving means is encoded by a two-dimensional encoding method. Or a one-dimensional encoding method, and if the determination is a two-dimensional encoding method, sequentially extract the code representing the shift from the image data, Decoding means for decoding the encoded code and restoring the arrangement of pixels on each scanning line. On the other hand, when the judgment is a one-dimensional encoding method,
A run-length extracting unit that decodes the image data to extract a run length of a pixel; and reproduces an image in accordance with an arrangement of pixels restored by the restoring unit, or a run-length extracted by the run-length extracting unit. An image reproducing unit that reproduces an image according to the following.If the encoding method is determined to be a two-dimensional encoding method, an adjacent pixel whose pixel arrangement is restored by the restoration unit is determined. For each of the two scanning lines, extract the changed pixel at the beginning of the area where pixels of the same value are continuous on the scanning line, and determine the same between the changed pixel on one scanning line and the changed pixel on the other scanning line. If the deviation of the pixel position in the scanning line direction between the values is within the predetermined number of pixels and is even, the other two values that are the first value correspond to the deviation. If the shift of the pixel position is within the predetermined number of pixels and is an odd number, the other binary data as the second value is synthesized with the shift. And the other binary data is separated from the image data.On the other hand, if the one-dimensional encoding method is used, the run length extracted by the run length extracting means is an even number. If so, it is determined that the other binary data that is the first value is combined with the run length, and if the extracted run length is an odd number, the run length is represented by the second value. A data separating unit that determines that some other binary data is synthesized and separates the other binary data from the image data; and a data output that outputs the binary data separated by the data separating unit. With means Facsimile machine characterized.