JPH09149263A - Prediction coder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a high compression rate by allowing a predict section to have plural prediction object picture elements and to predict them in the lump in the unit of plural picture elements to reduce the signal quantity for prediction signals and a code quantity. SOLUTION: A prediction device 2 predicts a pattern of plural prediction object picture elements from a pattern of input image data effecting on plural reference picture elements to provide an output of a prediction signal (a). A comparator 3 compares the received prediction signal (a) with image data received from an image input terminal 1 to discriminate whether or not the both are coincident. When the both are coincident (prediction is hit), a level '0' is outputted as a prediction error signal (b). When the both are not coincident, the comparator 3 provides an output of information denoting which picture element prediction is not hit among the prediction object picture elements as the prediction error signal (b). This information is made up of plural bits whose most significant bit is '1'. A coding section 5 generates a run length code depending on the run length of the prediction error signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a predictive coding apparatus, and more particularly to a predictive coding apparatus for compressing image data of a facsimile apparatus.

[0002]

2. Description of the Related Art Conventionally, a predictive coding apparatus that compresses image data of a facsimile apparatus has a reference pixel and a prediction target pixel as shown in the example of FIG. , The prediction target pixel is predicted to be a white pixel or a black pixel. In this conventional prediction encoding apparatus, a prediction value is output from a prediction unit for each reference pixel.

[0003]

However, the conventional prediction coding apparatus has a problem that the compression ratio is limited because the prediction is performed by the prediction unit in units of one pixel.

[0004] The present invention has been made in view of the above points,
A prediction encoding device that has a plurality of pixels to be predicted in a prediction unit and reduces the signal amount of a prediction signal by collectively performing prediction in a plurality of pixel units, thereby reducing a code amount and obtaining a high compression rate. The purpose is to provide.

[0005]

To achieve the above object, the present invention simultaneously predicts a plurality of prediction target pixels from a plurality of reference pixels of input image data, and predicts the prediction result and the actual input image data. And a coding unit that receives the prediction error signal as an input signal and that converts the prediction error signal into a code assigned according to the run length occurrence probability. To do.

Here, the prediction unit according to the present invention comprises: a predictor that outputs a prediction signal obtained by predicting a pattern of a plurality of pixels to be predicted from a pattern that appears in a plurality of reference pixels of input image data; A comparator that compares the prediction signal with the input image data and outputs a prediction error signal indicating whether or not prediction has failed in a plurality of prediction target pixels.

In the present invention, a plurality of prediction target pixels are simultaneously predicted in the prediction unit, and a prediction error signal indicating the prediction result is output. In particular, when the predictions match, a 1-bit prediction error signal can be output.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a block diagram of an embodiment of the present invention. As shown in FIG. 1, the predictive coding apparatus according to this embodiment includes a predicting unit 4 including a predictor 2 and a comparator 3, and a coding unit 5.

The predictor 2 performs prediction from image data input from the image input terminal 1 to generate and output prediction data a. The comparator 3 is an image data input terminal 1
The image data and the prediction data a received as input signals are received as input signals, and a prediction error signal b which is a difference signal between the two.
Is output. The encoding unit 5 generates a run length code according to the run length of the prediction error signal, and outputs the generated run length code to the code output terminal 6.

Next, the prediction method in the predictor 2 will be described in more detail. FIG. 2 shows the arrangement of reference pixels and the arrangement of prediction target pixels inside the predictor 2 of FIG. In FIG. 2, a pattern of four prediction target pixels indicated by squares with X is predicted from patterns of input image data appearing at 20 reference pixels indicated by white squares.

As shown in FIG. 3, four pixels to be predicted adjacent to each other in two rows and two columns are represented by a combination of white pixels on a white background and black pixels indicated by diagonal lines. Could take one of the following: The predictor 2 predicts which of these 14 states these reference pixels take, and outputs it as a prediction signal a.

Next, a comparison method in the comparator 3 will be described. The comparator 3 compares the input prediction signal a with the image data input from the image input terminal 1 and determines whether the two match. If they match (prediction is successful), “0” is output as the prediction error signal b. In this case, four pixels to be predicted have been replaced by one bit of the prediction error signal “0”, and the signal amount of the prediction error signal decreases, thereby increasing the compression rate of the encoding unit 5. Can be expected.

When the prediction signal a and the image data input from the image input terminal 1 do not match, the comparator 3 predicts information indicating which of the four prediction target pixels is out of prediction. Output as an error signal b. This information is composed of 5 bits starting with "1", and each bit has the meaning shown in FIG.

That is, in FIG. 4A, the first 1
The value “1” of the bit S serves as an identification flag indicating that prediction has been missed, and the following four bits A to D indicate a value corresponding to the prediction result of each of the four prediction target pixels. Show. Here, as shown in FIG. 4B, the second bit A indicates the prediction result of the upper left pixel of the four pixels to be predicted, and the third bit B indicates the lower left pixel of the four pixels to be predicted. The fourth bit C indicates the prediction result of the lower right pixel of the four pixels to be predicted, and the fifth bit D indicates the prediction result of the upper right pixel of the four pixels to be predicted. Shows the prediction results. The prediction result takes a value of “0” when the prediction of the pixel is correct, and takes a value of “1” when the prediction is out of place.

As described above, in this embodiment, four pixels to be predicted are predicted at once, and a 1-bit prediction error signal is output if the predictions match, and 5 bits if the predictions do not match. Although a bit prediction error signal is output, the above-mentioned prediction usually has a high probability of coincidence depending on the transmission path, so that the prediction error signal is higher than in the conventional case where a prediction error signal is output for each pixel. The signal amount of the signal can be reduced, and a high compression ratio can be obtained.

The present invention is not limited to the above embodiment, and the present invention can be applied to a plurality of prediction target pixels other than four.

[0017]

As described above, according to the present invention,
The prediction unit predicts a plurality of prediction target pixels simultaneously,
A prediction error signal indicating the prediction result is output, and the prediction can be performed in units of a plurality of prediction target pixels. In particular, when the predictions match, the prediction result of the plurality of prediction target pixels is a 1-bit prediction error signal. Is replaced by
The signal amount of the prediction error signal can be reduced, and a high compression ratio can be obtained, as compared with the related art in which prediction is performed in units of one pixel.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of the present invention.

FIG. 2 is a diagram illustrating an arrangement of reference pixels and an arrangement of prediction target pixels inside a predictor according to an embodiment of the present invention.

FIG. 3 is a diagram showing a possible pattern of a prediction target pixel.

FIG. 4 is an explanatory diagram of a method of generating a prediction error signal in the comparator of FIG. 1;

FIG. 5 is a diagram illustrating an example of a reference pixel and a prediction target pixel in a conventional device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Image data input terminal 2 Predictor 3 Comparator 4 Prediction part 5 Encoding part 6 Code output terminal a Prediction signal b Prediction error signal

Claims (3)

[Claims] 1. Prediction in which a plurality of prediction target pixels are simultaneously predicted from a plurality of reference pixels of input image data, and a prediction error signal that collectively indicates an error between the prediction result and actual input image data is output. And a coding unit that receives the prediction error signal as an input signal and converts the prediction error signal into a code assigned according to a run length occurrence probability. 2. The prediction unit outputs a prediction signal that predicts a pattern of a plurality of prediction target pixels from a pattern that appears in a plurality of reference pixels of the input image data, and an output prediction of the predictor. The prediction code according to claim 1, further comprising a comparator that compares the signal with the input image data and outputs a prediction error signal indicating whether or not the prediction in the plurality of prediction target pixels has failed. Device. 3. The prediction unit outputs 1 bit of a predetermined value as a prediction error signal when the prediction target pixel is n pixels and all predictions in the prediction target pixels of the n pixels are correct. , When the prediction is deviated by one pixel or more, the bit indicates a value different from the predetermined value and n indicates whether or not the prediction target pixel corresponds to each pixel in a one-to-one correspondence. 3. A prediction error signal of a total of (n + 1) bits consisting of bits is output.
The predictive coding device described.