JPH03119865A - Image data compression device - Google Patents

Abstract

PURPOSE:To perform the compression of image data with high compression efficiency by providing an encoder means to perform encode on data of difference outputted from an arithmetic means by using codes with different bit length and with bit length shorter as the absolute value of the data of difference is small. CONSTITUTION:A data compression part 53 is comprised of a latch circuit 91, a subtractor 92, a buffer memory 93, a CPU 94, a ROM 95, a RAM 96, and an input/output port circuit 97. The CPU 94 performs a data encoding processing to encode difference data CD stored in the buffer memory 93 transiently at every image data in one page to a compression code, respectively, and after that, outputs the compression code BD to an image memory 56 or a modem 57 via the input/output port circuit 97.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image data compression device used in facsimile machines and the like.

[Prior Art] Conventionally, when transmitting image data using a facsimile machine, the sending side compresses the image data so that the number of bits of the image data to be actually transmitted is smaller than the number of bits of the original image data. On the other hand, the receiving side decompresses the image data to restore the original image data, thereby reducing the transmission time.

For example, in the compression process in a facsimile machine, processing is performed to improve transmission efficiency when transmitting a general original document with characters, figures, etc. drawn thereon. That is, −
The white part of the manuscript to be sent for boat fishing is large compared to the entire manuscript without any text or figures, so the probability of consecutive white dots is very high, and the white image data of multiple dots is Compression processing is performed by converting the data into compressed data with a smaller number of dots.

Recently, there has been an increasing demand for using facsimile machines to transmit images including halftone images such as photographs, and facsimile machines having a halftone mode or a photo mode are being manufactured and sold.

When transmitting a halftone image using a facsimile machine, only white or black can be expressed in bits. Therefore, for example, a range of multiple N dots x N dots in a square shape is treated as one pixel, and the inside of that pixel is treated as one pixel. A halftone image is expressed by changing the black and white ratio.

[Problems to be Solved by the Invention] However, image data including halftone images has a lower probability of consecutive white or black colors compared to general character manuscripts, and when compression processing is performed using the conventional method, conversely There are problems in that the number of bits to be transmitted increases, the transmission efficiency deteriorates, and the transmission time increases.

For example, when one pixel is composed of four dots (2 dots x 2 dots) and expressed in halftones, as shown in Fig. 9,
The number of black dots within one pixel may be increased as the gradation of the image increases, and images with five types of density patterns may be used. In this case, the read image is converted pixel by pixel into one of the density patterns of gradations O to 4 shown in FIG. 9, and image data conversion processing is performed. A halftone image is expressed by a set of pixels having the above-mentioned density pattern, and when a part of the image is extracted, it becomes as shown in FIG. 10, for example.

The image data of the halftone image shown in FIG.
When compression processing is performed using the encoding method, a 1-dot white image is converted to 6-bit image data, and a 1-dot black image is converted to 3-bit image data.
For example, in processing the image of the first line, the number of bits of the image data after compression processing is 4.5 times that of the original image data, and the number of bits of the image data to be transmitted increases significantly.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned problems and to provide an image data compression device that can compress image data with higher compression efficiency than conventional methods even for image data including halftone images. be.

[Means for Solving the Problems] An image data compression device according to the present invention includes a storage means for temporarily storing image data indicating gradation that is input continuously;
Input image data and 1 output from the storage means
A calculation means for calculating the difference data from the image data before the pixel, and a calculation means for calculating the difference data output from the calculation means, the bit lengths of which are different from each other according to the difference data, and the absolute value of the difference data is different from each other according to the difference data. The present invention is characterized by comprising an encoding means for encoding using a code having a smaller bit length.

[Operation 1] With the above configuration, the storage means temporarily stores the image data indicating the gradation that is input continuously, and the calculation means stores the image data inputted and the storage means. Calculate the difference data from the image data of the previous pixel to be output. Next, the encoding means converts the difference data output from the calculation means into a code having different bit lengths depending on the difference data and having a shorter bit length as the absolute value of the difference data is smaller. Encode using .

Therefore, if the continuously inputted image data is, for example, image data including a halftone image, and the change in gradation of the image data is particularly small, the information of the image data is encoded into a code with a shorter bit length. Therefore, image data can be compressed with higher compression efficiency than conventional methods.

[Embodiment] Hereinafter, a facsimile machine that is an embodiment of the present invention will be described in the following order with reference to the accompanying drawings.

(a) Configuration of facsimile machine (b) Operation panel (c) Configuration of signal processing unit (d) Data encoding process 70- (e) Flow of data decompression process (f) Operation of facsimile machine (a) Facsimile machine FIG. 2 is a sectional view of a facsimile machine according to an embodiment of the present invention. This facsimile machine has a printer section l.
, a reading section 20 installed above it, and a signal processing section 100 built into the reading section 20. In the data compression section 53 of this signal processing section 200, input image data and l pixels are processed. The image data is compressed by calculating the difference data with the previous image data and encoding the data into the compression code shown in Table 2 based on the calculated difference data.

The printer unit 1 is a laser printer, and its operation will be briefly described below. First, the photoreceptor on the rotationally driven photoreceptor drum 2 is uniformly charged by the charger 3 .

Next, a laser beam is irradiated by the optical system 4 according to the image data to form an electrostatic latent image on the photoreceptor. Toner from the developing device 5 adheres to this electrostatic latent image. On the other hand, cut sheets are placed in the paper feed cassette 11, and the cut sheets are fed one by one toward the photosensitive drum 2 by rollers or the like. The toner adhering to the photosensitive drum 2 is transferred onto the cut paper by the transfer charger 6 and fixed by the fixing device 12, after which the cut paper is discharged onto the tray 13. On the other hand, the toner that has not adhered is collected by the cleaner 8, and this completes -th printing.

Next, in the reading section 20, reading of the transmitted original is performed in the same manner as in the past. That is, the originals placed on the original tray 21 are detected by the sensor 22, and are fed one by one to the position of the sensor 25 by the rollers 23 and 24, which have been lowered to the dotted line position. Next, the motor (not shown)
The rotation of the roller 24 and the reading by the contact type COD linear image sensor 26 are synchronized, the original illuminated by the lamp 30 is read by the COD linear image sensor 26, and the original image is converted into digital image data.

After the reading is completed, the document is discharged to the discharge section 28 by the discharge roller 27. In this embodiment, the rollers 24 and 27 are driven by the same drive system, and the original is ejected when the next original is read.

Further, a telephone handset 29 connected to a telephone line via the signal processing section 100 is provided on the side of the reading section 20, and when the facsimile machine is not in use, this handset 29 can be used to make a call. can. Further, an operation panel 15 for setting the operation of the facsimile machine and displaying the operating status is provided on the upper surface of the printer section 1 located next to the reading section 2o.

The image data received by the signal processing unit 100 from the destination facsimile device via the telephone line is processed by the signal processing unit 10.
After predetermined signal processing such as data compression processing, which will be described in detail later, is performed in the printer section 14, the image data read by the reading section 20 is decompressed in the signal processing section 100. After being subjected to predetermined signal processing such as geography, which will be described in detail later, the signal is transmitted to the other party's facsimile machine via the telephone line.

Note that the recording section 1 and the reading section 2o do not have to be integrated.

(b) Operation panel FIG. 3 shows the arrangement of keys, display panel, etc. on the operation panel 15. The operation panel 15 includes a numeric keypad 40, a liquid crystal display panel 41, a one-touch key 42, a send key 43, a memory key 45, a number print key 46, a registration key 47,
An automatic send key 48 is also configured.

The numeric keypad 40 displays the number of the telephone line to which the destination facsimile device is connected (hereinafter referred to as facsimile number).
The liquid crystal display panel 41 is used to display the operating status of the facsimile machine or to give instructions and warnings to the operator. The one-touch keys 42 are used to select the R of the facsimile number of the destination that corresponds to each key.
It is used for storing in AM72 (see Figure 1) or for one-touch transmission to a destination facsimile number.

The transmission key 43 is a key for starting transmission of the original image read by the reading section 20, and the memory key 45 is a key for starting transmission of the original image read by the reading section 20.
This key is used to temporarily store the image read by the reading unit 20 in the image memory 56 (see FIG. 1). Further, the number print key 46 is a key for printing the facsimile number of the other party corresponding to each key of the one-touch key 42, and the registration key 47 is used in a registration mode for registering the facsimile number of the other party and a non-registration mode. This is a key that only specifies one of the two. Furthermore, the automatic transmission key 48 is
This key is used to transmit the image data stored in the image memory 56 to the other party at a predetermined designated time.

(c) Configuration of signal processing unit FIG. 1 shows the signal processing unit 1 of the facsimile machine of this embodiment.
00 is a block diagram.

In FIG. 1, an original image 49 is irradiated by a lamp 30 whose on/off is controlled by a lamp controller 50, and the light reflected by the original image 49 is incident on the CCD linear image sensor 26.

The CCD linear image sensor 26 photoelectrically converts the received light and converts the electrical signal into an analog/digital signal (hereinafter referred to as A/D conversion) via an amplifier 51.

) output to the device 52. The A/D converter 52 converts the input electrical signal into 3-bit image data AD with 5 gradations per pixel according to the gradation level, as shown in Table 1, and then converts the input electrical signal into 3-bit image data AD with 5 gradation levels per pixel. Output to.

As shown in FIG. 4, the data compression unit 53 includes a latch circuit 91, a subtracter 92, a buffer memory 93, and a CPU.
94, ROM 95, RAM 96, and input/output port circuit 97. The latch circuit 91 temporarily latches the input 3-bit image data in synchronization with the clock signal SCK generated from the clock generation circuit 60 and outputs it to the subtracter 92. 4-bit difference data CD between the image data AD and the image data AD' of the previous pixel output from the latch circuit 91 is calculated and output to the buffer memory 93.

The CPU 94 stores the difference data CD temporarily stored in the buffer memory 93 for each page of image data.
After performing data encoding processing (see FIGS. 6(A) and (B)) to encode the compressed code according to Table 2, the compressed code BD is sent to the image memory 56 via the input/output port circuit 97.
Alternatively, it is output to a modulation/demodulation device (hereinafter referred to as a modem) 57. Here, the CPU 94 connects to the CPU 94 via the bus 98.
The CPU 94 is connected to a ROM 95 that stores the arithmetic control program 94 and compression codes corresponding to each difference data CD, a RAM 96 used as a work area for the CPU 94, and an input/output port circuit 97. Note that, for example, when processing an image having the density pattern shown in FIG. 1O, the data compression unit 53 performs data compression processing as shown in Table 3. Further, since there is no image data before the image data of the first dot of each line, the latch circuit 91 latches dummy data, which is image data "010" with an intermediate gradation level, as the image data of one pixel before.

The above processing of the image sensor 26, A/D converter 52, and data compression section 53 is performed in synchronization with the clock signal SCK generated from the clock generation circuit 60.

Further, the image memory 56 temporarily stores compressed image data for automatic transmission at a designated time, or for storing image data received via a telephone line when the printer section l is not operating or malfunctioning. This is a memory for temporarily storing image data and outputting it to the printer section 1 for printing after the printer section 1 recovers.

The modem 57 modulates the image data, which is a compressed code outputted from the data compression unit 53 or the image memory 56, using a predetermined modulation method, and then connects the modem 57 to a network control unit t (hereinafter referred to as NCU) that performs line connection control with a telephone line. ) 58 to the telephone line. Furthermore, the modem 57 demodulates the image signal received via the telephone line and the NCU 5g, converts it into image data that is a compressed code, and then outputs it to the image memory 56 or the data expansion section 81.

The CPU 70 is a control device that controls the operation of this facsimile machine, and is connected via a bus 75 to a ROM 71 that stores the system program of the CPU 70, etc., and a ROM 71 that stores the facsimile number of the other party and is used as a working area of the CPU 70. RAM72 and NCU
38, a modem 57, an input/output port circuit 74 that outputs a lamp control signal to the lamp control section 50 that controls on/off of the lamp 30, and a reading sensor section 76 to which the sensors 22 and 25 of the reading section 20 are connected. Connected. Also, CPU
70 is connected to the operation panel section 15 via the bus 75 and the input/output port circuit 73, and takes in information input using each key of the operation panel section 15, and displays the information on the liquid crystal panel 41 of the operation panel section 15. Displays the operation of the facsimile machine.

The data decompression unit 81 performs decompression processing on the image data input from the image memory 56 or the modem 57, which is a process opposite to the data compression processing of the data compression unit 53, and then
The image data after the decompression process is output to the printer section l. As shown in FIG. 5, this data decompression unit 81 includes two image buffer memories 120 and 121, a CPUll0 that performs data decompression processing, a ROM III that stores a processing program for the data decompression processing, and a workpiece of CPUll0. Consists of RAM 112 used as an area, C
PUll0 is connected to the buffer memory 12 via the bus 113.
0.121 Connected to ROMI l and RAM112. Image data input from the modem 57 or the image memory 56 to the data decompression unit 81 is transferred to the buffer memory 1.
After being temporarily stored in 20, the data is decompressed by CPUll0 (see FIG. 7). The image data after the data expansion process is temporarily stored in the buffer memory 121 and then output to the printer section 1.

Finally, the printer unit 1 performs printing on cut paper as described above based on the input image data.

(d) Flow of data encoding processing FIGS. 6(A) and 6(B) show flowcharts of data encoding processing in the data compression unit 53.

As shown in FIGS. 6(A) and (B), step 551
(Hereinafter, "step" will be omitted.), S53, S
55, S57, S59, S61, S63, and S65, the difference data CD stored in the buffer memory 93 and to be processed is 0100""0011""00"
10"0001""0000""1111"
"1110"1101" is determined and encoded into a compression code according to Table 2. That is,
If the difference data CD is “0100” (YES in S51)
) The compression code '10000001' should be output to the input/output boat circuit 97. S52) If the difference data CD is '001'
1" (YES in S53), the compression code is "10000"
1" to the input/output boat circuit 97 (554), and if the difference data CD is "0010" (YES in S55)
) compression code "to.

The bit is output to the input/output boat circuit 97 (S56). Also, if the difference data CD is "0001" (YES in S57), the compression code "11" is output to the input/output boat circuit 97, and if the 558L difference data CD is "oooo" (
YES in S59) Input/output compression code “0” to the input/output boat circuit 9
If the difference data CD is "111 bits" (YES in S61), the compressed code "101" is output to the input/output boat circuit 97 (S62). ” (YES in S63)
The compression code "+0001" is output to the input/output boat circuit 97 (564), and if the difference data CD is "1101" (YES in S65), the compression code "1000001" is output to the input/output boat circuit 97 (566), If the data CD is “1100” (No in S65), the compression code is “10”
0000001” to the input/output boat circuit 97 (
S67). Above S52, S54, S56, S58, S
60. After the compression code output processing in S62.564, S66, and S67, the process advances to S70.

In S70, it is determined whether or not the processing of data for one page has been completed. If not, the process returns to S51 and the above-mentioned process is repeated; on the other hand, if it has been completed, this data encoding process is completed. end.

(e) Flow of Data Decompression Processing FIG. 7 shows a flowchart of the main routine of data decompression processing in the data decompression section 81.

As shown in FIG. 7, first, in Sl, CPUll
0 is an image restoration process (FIG. 8 (A) to
See (B) and (C). ) is performed (S2). Then,
The image data after image restoration processing is stored in the buffer memory 121.
- After being temporarily stored, it is output to the printer section 1 (
S3).

FIG. 8 (A), (B), and (C) show image restoration processing (
A flowchart of the subroutine of S2) in FIG. 7 is shown. In this image restoration process, DN+M is the image data after decompression processing of the Mth pixel of the Nth line of the first page, and a is the bit number from the most significant bit in the compression code of one pixel (however, the most significant bit b is a parameter indicating the count value of 0'' data in the compression code.

First, after the parameter N is initialized to 0 (S100)
, parameters a and b are both initialized to O (Slo
t). Next, 1 is added to the parameter N, and the parameter M is initialized to O (S102). Furthermore, after 2 is set in the image data D, , M (S l 03),
If parameter a is 0 (YES in S I O4
), it is determined whether the input code data (hereinafter referred to as code data) is '0' (S105).

Here, if the code data is 11011, the image data I), , M-, of the previous pixel are converted to the image data D H,M
After adding 1 to the parameter M, the program proceeds to the check routine starting from 3180. If the code data is not 0'' (NO in S105), 1 is added to the parameter a, and then the process proceeds to 5180.

In 5104, if the parameter a is not "o", it is determined whether the code data is 0'' (S
IO9). Here, if the code data is 0'' (
After 1 is added to parameter b (S109: YES) (SIIO), it is determined whether parameter b, which indicates the count value of "O" data, is 8 or more. If the number of "θ" in one code is 8 or more, it is determined that there is an error in the transmitted data, and the process proceeds to the error routine starting from 5190. On the other hand, if the number of "0" is less than 8, 518
Proceed to check routine after 0.

On the other hand, in 5109, if the code data is not 0', after resetting the parameter a to 0, Sl 13,
5120, 5130, 5140°3150.5160.
5170, it is determined whether the parameter b is 0 to 6, it is determined which compression code in Table 2 the transmitted data is, and the image data DN corresponding to the compression code and indicating the gradation is determined. , M are stored in the buffer memory 121. Here, if the parameter b is O, then (S
113: YES) One pixel previous image data D, , -.

The image data obtained by subtracting 1 from the image data D is S.
l14), if parameter b is 1 (S12
0 = YES) Add 1 to the image data DNIM-1 of the previous pixel to create the image data D N, M5121)
, if parameter b is 2 (YES in S130)
The result obtained by subtracting 2 from the image data DN, M-1 of one pixel before is set as image data D, , M (S131). Also, if parameter b is 3 (YES in S140), 1
Add 2 to the image data I) and Il+i+-+ before the pixel as the image data D N+M S141),
If parameter b is 4 (YES in S150), 1
The image data DN and M obtained by subtracting 3 from the image data before the pixel D N r M-1 are the image data DN and M. If the 5t51L parameter b is 5 (YES at S160), the image data before the pixel Dll+M-1 is 3. The sum of these is set as image data D, , M (S l 61). Furthermore, if the parameter b is 6 (YES in S170), the image data DH,M is obtained by subtracting 4 from the image data D, . ), parameter b
If is 7 (No in 5170), the image data of the previous pixel D) IIM-1 plus 4 is the image data D
H9M (S 1 72). 5114.512 above
1.5131.5141,5151, S 161°51
After each restoration process of image data in 71 and 5172, parameter b should be reset to 0.S l l 5)
, after adding the parameter Milli (S116),
Proceed to check routine after 3180.

In 5180, it is determined whether or not the processing of the l line has been completed, and if the processing of the l line has not been completed, the process returns to step 5104.
Then, the process of restoring the compressed code described above is repeated. on the other hand,
If the processing of 1 line has been completed, it is determined whether the number of image data of 1 line is a predetermined number (C5181), and if they do not match, the process proceeds to the error routine after 5190; If they match, proceed to 5182.

At 5182, it is determined whether the restoration process for one page of image data has been completed, and if the process for one page has not been completed, the next line is processed.
Repeat the above process from step 1.

If the processing for one page has been completed, the process returns to the main routine shown in FIG.

When the number of image data in the l line does not match the predetermined number (No in 5181), or if the number of ``0'' data included in one compression code is 8 or more (S l
11), it is assumed that there is an error in the transmitted data, and the error routine from 5190 to Sl'95 is performed.

First, after performing the process of restoring the compressed code data of the next line of the data currently being processed into image data indicating gradation (5190), from 5191 to 5195, the image data of the next line and the previous line are The line currently being processed is calculated by calculating the average value of the image data of the line (this data indicates the average of the gradation of the image data of the previous line and the gradation of the image data of the next line). image data, and create pseudo image data. That is, first, after resetting the parameter M to 1 (S191), the average value data (DN-1, M + DN + l1M) /2
is calculated and stored in the buffer memory 121 as image data D N+M (3192), 1 is added to the parameter M (S193), and this process is repeated until 1942 minutes of processing is completed (S194).

After finishing processing for one line (YES at S l 94)
, after adding 1 to the parameter N and updating it (S195)
, 5104 and proceeds to process the next line.

As explained above, when the number of image data in one line does not match the predetermined number (No in S181), or when the number of "0" data included in one compression code is 8 or more, If there is (YES in S l l I), there is an error in the transmitted data, and 5190 to 5195
By performing the error routine up to this point, the original image data sent on the sending side can be restored in a pseudo manner.

(f) Operation of facsimile equipment Next, the transmission operation of the facsimile device will be explained.

First, when a document is placed on the document tray 11, the presence of the document is detected by the sensor 22 of the reading sensor section 76, and the information is sent to the CPU 70. In response to this, CPU7
0 displays a request for input of the recipient's facsimile number on the liquid crystal display panel 41 of the operation panel section 15, and then when the facsimile number is input using the numeric keypad 40, C is displayed.
The PU 70 reads out the facsimile device of the other party via the NCU 38 and the telephone line. When reading is performed,
The motor of the reading section 20 and the CCD linear image sensor 26
The image signal is read l lines at a time without synchronizing the image signals, and the image data of the l line is sent to the A/D converter 52 via the amplifier 51', and as shown in Table 1, 3 lines per pixel.
After being A/D converted into 5-bit image data AD, it is sent to the data compression section 53. In the data compression unit 53, as described above, the input 3-bit image data A
After the difference data CD between D and the image data AD' of one pixel before is calculated, the difference data CD is converted into compressed code image data as shown in Table 2 and sent to the modem 57. Next, the image signal is modulated into an image signal using a predetermined modulation method by the modem 57, and then transmitted to the destination facsimile machine via the NCU 38 and the t-communication line. When all the image data has been transmitted, the telephone line is disconnected according to a predetermined disconnection procedure, and the transmission is completed.

Next, the operation during reception will be described.

When a call is received from the other party's facsimile machine through the telephone line, the CPU 70 connects the line according to a predetermined communication procedure. When the line connection is completed, the image data sent from the other party's facsimile machine is transferred to the NCU3.
8 and the modem 57 to the data decompression unit 81,
After performing the data decompression processing, which is the reverse processing of the data compression section 53, the processed image data is sent to the printer section 1, and an image of the image data is printed on a cut sheet as described above. When all image data has been received and printed, the telephone line is disconnected according to the communication procedure. This completes the receiving operation.

As explained above, in this embodiment, the difference data CD between the image data AD indicating the gradation level and the image data AD' one pixel before is calculated, and the calculated difference data CD is converted into the compression code shown in Table 2. It is encoded using For example, when the image data shown in FIG. 1O is compressed using the data compression unit 53 of this embodiment, the number of bits of the encoded data is
20 as shown in the table. On the other hand, when the image data shown in FIG. 1O is encoded using, for example, a known MH encoding method, the number of bits of the encoded data is 290. Therefore, by using the data compression process in the data compression unit 53 of this embodiment, it is possible to compress image data with a compression efficiency of ten times or more. Further, the number of bits of the original image data shown in FIG. 10 is 48, and when compared with the original image data, in this example, 2 bits are used.
Image data can be compressed with compression efficiency that is more than double that.

Other Embodiments In the above embodiments, input image data AD and 1
Calculate the difference data CD with the image data AD' before the pixel, and divide the calculated difference data CD into the most significant bit as shown in Table 2, except when the difference data CD is "oooo". The least significant bit is "I", and between the most significant bit and the least significant bit there is a "0" bit with a different bit length depending on the absolute value of the difference data CD, and the value of the difference data CD The smaller the bit length, the shorter the bit length.However, the most significant bit and the least significant bit are set to "0",
The compression code may be configured to have a "1" bit between the most significant bit and the least significant bit.

In the above embodiment, 5 dots of 2×2 dots per pixel.
Although we have described the case of processing image data with a dither pattern of gradations, this is not the only option; processing image data with a data pattern having a high gradation of, for example, 4 x 4 dots or 8 x 8 dots per pixel. The data compression processing and data expansion processing of the present invention can also be applied to such cases.

In the above embodiments, a facsimile machine that performs data compression processing and data expansion processing of image data has been described, but the present invention is not limited to this, and can be applied to other image processing devices that perform image data processing. Further, although a fixed RAM memory is used as the image memory 56, the compressed image data may be stored in a removable memory card.

(Margin below) Table 1 Table 2 [Effect of the invention 1 As detailed above, according to the present invention, the data of the difference between the input image data and the image data one pixel before is calculated,
The difference data is encoded using a code that has different bit lengths depending on the difference data and has a shorter bit length as the absolute value of the difference data is smaller, so that it is input continuously. For example, if the image data is image data including a halftone image, and the change in gradation of the image data is particularly small, the information of the image data can be encoded into a code with a shorter bit length, so it is possible to encode the information of the image data into a code with a shorter bit length. It is possible to compress image data with higher compression efficiency compared to .

[Brief explanation of drawings]

FIG. 1 is a block diagram of a signal processing section of a facsimile device according to an embodiment of the present invention, FIG. 2 is a sectional view of the facsimile device of FIG. 1, and FIG. 3 is an operation panel of the facsimile device of FIG. 2. FIG. 4 is a block diagram of the data compression section of the facsimile machine of FIG. 1, FIG. 5 is a block diagram of the data decompression section of the facsimile machine of FIG. 1, and FIGS. ) is a flowchart of the data encoding process in the data compression section 53 of FIG. 4; 'M7 is a flowchart of the main routine of the data decompression process in the data decompression section of FIG. 5; FIG.
(B) and (C) are flowcharts of the image restoration processing subroutine shown in Fig. 7, Fig. 9 is a diagram showing an example of a density pattern with 5 gradations, and Fig. 10 is a diagram composed of the density pattern shown in Fig. 9. It is a figure showing an example of image data. 54...Data compression unit, 91...Latch circuit, 92...Subtractor, 93...Buffer memory, 94...CPU. 95...ROM. 96...RAM. 97...I/O port circuit.

Claims (1)

[Claims] (1) A storage means for temporarily storing continuously inputted image data indicating gradations;
a calculation means for calculating the difference data from the image data before the pixel; and a calculation means for calculating the difference data output from the calculation means, the bit lengths being different from each other according to the difference data and the absolute value of the difference data being different from each other according to the difference data. An image data compression device comprising: encoding means for encoding using a code having a smaller bit length.