JP3076070B2 - Encoded data transmission device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoded data transmitting apparatus for compressing binary data by predictive encoding and transmitting the compressed data.

[0002]

2. Description of the Related Art As one method of predictive coding for performing data compression, arithmetic codes (Arithmetic Co.
ding) is known.

[0003] In this arithmetic code, when the coded data is restored to the original data, a fixed operation process is performed between the coded data and various parameters obtained in the decoding process. I have.

[0004] For this reason, when transmitting the encoded data, if the receiving side once causes a reception error, the original data cannot be correctly restored thereafter.

Usually, a facsimile apparatus transmits a document image line by line in page units. For this reason, when a facsimile apparatus transmits image data after compressing the image data using arithmetic codes, if a transmission error occurs once, it becomes impossible to restore the image information of all subsequently received lines, and the received image is greatly reduced. Will be missing.

[0006]

As described above, conventionally,
When data is transmitted by compressing data using an arithmetic code, there is a problem that transmission data is lost when a transmission error occurs, which is not practical.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above-mentioned problems and to provide an encoded data transmitting apparatus which can reduce the loss of transmission data when a transmission error occurs and can be put to practical use.

[0008]

Normally, the encoding process starts in an initial state in which there is no past prediction result and no generated code, but the present invention sequentially transmits the codes generated by the encoding process. Each time the code is transmitted for a fixed number of bytes, the code is set to the initial state and a fixed identification code is transmitted.

[0009]

Even if a transmission error occurs, the receiving side can correctly restore the subsequent received data by initializing the decoding section when the above-mentioned identification code is received. As a result, loss of transmission data when a transmission error occurs is reduced, and the transmission data can be put to practical use.

[0010]

Embodiments of the present invention will be described below in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an image transmission system according to an embodiment of the present invention. In the figure, a data compression unit 1 includes a symbol probability evaluator 1a and an encoder 1b, and converts binary image data into a Q-COD.
It is encoded by a known method of an arithmetic code called ER. The transmitting unit 2 transmits the code information obtained by the coding to a communication line. The control unit 3 monitors and controls the operation of the data compression unit 1 and the transmission unit 2.

The receiving section 4 receives code information via the communication line. The data restoring unit 5 includes a decoder 5a and a probability evaluator 5b, and restores received code information to original binary image data. Control unit 6
Monitors and controls the operation of the receiving unit 4 and the data restoring unit 5.

Next, an image transmission operation of the image transmission system according to the present embodiment having the above configuration will be described.

When the transmitting side starts, as shown in FIG. 2, the control section 3 sets a variable r for counting transmission data to a constant value R (processing 101) and checks input of image data (step 101). Process 102).

At the time of image transmission, black and white binary image data is sequentially input line by line from a device (not shown). When image data is input (Y in process 102), the control unit 3 operates the data compression unit 1 and the transmission unit 2.

Thus, the probability evaluator 1a identifies the input image data as symbol information, and predicts the appearance probability of the symbol for each pixel based on the past pixel state. Then, it outputs the superior symbol information MPS indicating whether the superior symbol is black or white, and the appearance probability Qe of the inferior symbol.

The encoder 1b performs predetermined arithmetic processing based on the input image data, the above-mentioned superior symbol information MPS, and the appearance probability Qe of the inferior symbol, and sequentially generates an arithmetic code (the above processing 103). .

Next, the control unit 3 checks the number of bits of the generated code (step 104), and if the number is less than 1 byte (N in step 104), continues the above processing (step 10).
2). Then, when a code for one byte is generated (Y in process 104), the transmitting unit 2 transmits the generated code (process 105).

Next, the controller 3 decrements the variable r by 1 (step 106) and checks the value (step 10).
7). Here, if the variable r is not 0 (N in Step 107), the above processing is repeated (Step 102).

As a result, the variable r decreases each time the generated code is transmitted by one byte. In this way, when the variable r = 0 (Y in step 107), a preset identification code S is transmitted (step 108). This identification code is constant data that does not overlap with the code obtained by the above-described encoding, and is set to, for example, 2-byte data of all “1”.

Thereafter, the probability evaluator 1a and the encoder 1
b is initialized (process 109). As a result, initial values are shown for the values of the superior symbol information MPS and the inferior symbol appearance probability Qe output from the probability evaluator 1a.

Then, the above processing is repeated in the same manner from the beginning (to processing 101). As a result, as shown in FIG. 3, the arithmetic code of the image data is sequentially transmitted with the identification code S inserted for every fixed number of R bytes. Then, when there is no input of image data (N in process 102), the above-described transmission process ends.

On the other hand, as shown in FIG. 4, the receiving side sequentially receives the codes transmitted by the receiving unit 4 (process 201).
At this time, the control unit 6 monitors the code to be received (Process 2
02), if it is an arithmetic code (N in process 202),
The decoder 5a and the probability evaluator 5b are operated.

Thus, the decoder 5a restores the received arithmetic code to the original image data based on the superior symbol information MPS and the inferior symbol appearance probability Qe output from the probability evaluator 5b. In parallel with this, the probability evaluator 5b inputs the restored image data, and outputs the superior symbol information MPS and the appearance probability Qe of the inferior symbol (step 203).

The restored image data is output to a device (not shown) such as a recording device (step 204).

In this embodiment, it is assumed that the number of pixels of the received image is set to a fixed value in advance. At the time of outputting the image data, it is determined whether or not the image reception is completed by counting the number of output pixels (step 205). If image reception is continued (N in process 205), it is determined whether decoding of one received one-byte arithmetic code has been completed (process 206). If the processing of the one arithmetic code is not completed (N in processing 206), the process returns to the decoding processing (processing 203).

When the decoding of the one arithmetic code is completed (Y in step 206), the next code is received and the same processing is repeated (to step 201).

As can be seen from FIG. 3, the arithmetic code is a constant number R
Each time a byte is received, the identification code S is received. When the identification code S is received (step 20
(Y of 2), the decoder 5a and the probability evaluator 5b are initialized (process 207), and the same decoding process as described above is executed (process 203).

The arithmetic code following the identification code S is generated on the transmitting side by initializing the probability evaluator 1a and the encoder 1b. Therefore, the receiving side can also correctly decode the arithmetic code by the initialization.

When image data for the predetermined number of pixels is received (Y in step 205), the above-described reception processing is terminated.

As described above, in this embodiment, when image data is encoded by an arithmetic code and transmitted, the transmitting side
Each time a certain number of bytes of the arithmetic code are transmitted, the data compression unit 1
Is initialized and a certain identification code S is transmitted, and the receiving side initializes the data restoration unit 5 every time the identification code S is received.

By performing such initialization, when decoding the arithmetic code, the receiving side can restore the original image data without referring to various parameters obtained in the past decoding process. it can. As a result, even if a transmission error occurs, the image data can always be correctly restored from the time when the identification code S is received, so that the loss of transmission data is reduced.

Therefore, when transmitting an image with a facsimile apparatus, if initialization is performed every several lines, for example, even if a transmission error occurs, only a few lines of an image of one page are lost. No practical problem occurs.

Although the above embodiment has been described by taking as an example the case where image data is encoded by Q-CODER, which is a method of predictive coding, the present invention is similarly applied to other predictive coding methods. It is possible to do.

Although the description has been given of the case where image data is transmitted, for example, in a data transmission field where transmission data is partially lost, such as when an audio signal is converted into a digital signal and transmitted, there is no problem. In the above, the present invention can be applied.

[0036]

As described above, according to the present invention, every time the generated code is transmitted by a certain number of bytes, the code is returned to the initial state where neither the past prediction result nor the generated code is present, and a certain identification code is transmitted. Since the transmission is performed, the receiving side initializes the decoding unit when the identification code is received, so that the subsequent reception data can be reliably restored. Can be reduced, and can be put to practical use.

[Brief description of the drawings]

FIG. 1 is a block diagram of an image transmission system according to an embodiment of the present invention.

FIG. 2 is an operation flowchart showing data processing on the transmission side.

FIG. 3 is an explanatory diagram of transmission data.

FIG. 4 is an operation flowchart illustrating data processing on the receiving side.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Data compression part 1a, 5b Probability evaluator 1b encoder 2 Transmission part 3, 6 Control part 4 Receiving part 5 Data decompression part 5a Decoder

Continuation of the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N 1/41-1/419 G06T 9/00

Claims (1)

(57) [Claims] An input unit for sequentially inputting binary data;
Prediction evaluation means for predicting the appearance probability of each symbol based on the past symbol sequence of the binary data and the past prediction result, and sequentially generating a new code based on the prediction result and the code generated in the past In a coded data transmitting apparatus including a predictive coding unit for performing the above and a transmitting unit for transmitting the generated code, the prediction and the generation of the code are started in an initial state in which there is no past prediction result and the generated code. Processing starting means, and transmitting means for sequentially transmitting the generated code, and initialization means for setting the initial state in which neither the prediction result in the past nor the generated code is present each time the code is transmitted by a fixed number of bytes, A coded data transmitting device comprising: an identification code transmitting means for transmitting a fixed identification code each time the initial state is set.