JPH01108861A - Picture coder - Google Patents

Abstract

PURPOSE:To improve a compression rate with respect to a picture including a pseudo halftone picture by applying run length coding while a segmented part is buried by a white level and other parts use a data of the original, applying predictive coding while the segmented part uses the data of the original and other parts are buried by a white level and then applying run length coding. CONSTITUTION:A picture signal stored in a frame memory 1 is inputted to white bury means 2, 4, a rectangular area including a pseudo halftone picture of an original is set and the picture signal is stored all as a white picture signal and the picture signal other than the rectangular area is stored as it is. The picture signal stored in the frame memory 1 is inputted to segmentation means 3, 5, the picture signal in the rectangular area is stored as it is and the picture signal other than the rectangular area is stored as the white picture signal and given to a prediction coder 6, where prediction coding is applied. A select means 7 selects the output of the prediction coder 6 and the output of the white level bury means 2, 4, gives the result to a run length coder 8, where the signal is coded. Thus, excellent compression rate is attained to the picture including the pseudo halftone picture.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is applied to an image encoding device for image communication. In particular, the present invention relates to an encoding device for an image including a pseudo intermediate image.

〔overview〕

The present invention uses an image encoding device to cut out a rectangular area containing a pseudo-halftone image from a document containing a pseudo-halftone image, fill in the cut-out portion with white, and run data other than the cut-out portion using original data. By performing length encoding, using the extracted part as the original data, filling the other parts with white, performing predictive encoding, and then performing run-length encoding, an image containing a pseudo halftone image can be created. In contrast, the compression ratio is improved.

[Conventional technology]

FIG. 3 is a block diagram of a conventional image encoding device.

Conventionally, image encoding devices have been equipped with predictive encoders and predictive encoders that convert pseudo halftone images such as dithered images or halftone images in which many short runs of white and black are generated into long runs in order to efficiently perform run length encoding. , and a run-length encoder for run-length encoding. Its operation will be explained with reference to FIG.

In FIG. 3, 1 is a frame memory that stores images in page units, 6 is a predictive encoder, and 8 is a run-length encoder. Further, sl is an image signal, Sl□ is an image signal output from the frame memory 1, and 513 is an image signal S1
□ sampling clock, Sol is the image request signal from the predictive encoder 6 to the frame memory 1, S22 is the predicted image signal that is the output of the predictive encoder 6, and S23 is the predicted image signal S2□ that is the output of the predictive encoder 6. A sampling clock S31 is a predicted image request signal from the run-length encoder 8 to the predictive encoder 6, and S32 is a run-length code signal output from the run-length encoder 8.

After the image is stored in the frame memory 1, the run-length encoder 8 converts the predicted image request signal S31 into “
"on" to request a predicted image from the predictive encoder 6. (3rd
The frame memory status monitoring unit is omitted in the figure. )
The predictive encoder 6 requests an image from the frame memory 1 by turning on the image request signal S21. Frame memory 1 outputs image signal 312 in synchronization with sampling clock S13. The predictive encoder 6 receives the image signal S
12 into a predicted image, and the sampling clock 323
The predicted image signal S2□ is output in synchronization with the .

The run-length encoder 8 performs run-length encoding on the predicted image signal 322 and outputs a run-length encoded signal 332.

The above series of operations continues until the frame memory 1 runs out of data. Here, as the run-length encoder 8, a well-known encoding circuit such as modified Huffman (MH) encoding or modified read (MR) encoding is used. Further, as the predictive encoder 6, for example, the time series signal X of the image signal binarized by the dither matrix (period T) and this time series signal X are each 1.2, .
There is a predictive encoder that has logic conversion means for converting a signal delayed by (T-1) samples into a time series signal Y representing whether the sum of a total of T signals is an even number or an odd number.

In addition to the logic conversion means, the predictive encoder 6 shown in FIG. 3 has a through mode in which no conversion is performed on the image signal. In the through mode, the image signal 312 and the predicted image signal S22 are the same.

Although not shown in FIG. 3, it is possible to select whether the predictive encoder 6 performs logic conversion or enters the through mode.

Refer to Japanese Unexamined Patent Publication No. 57-140069 [Problems to be Solved by the Invention] However, in such conventional image encoding devices, predictive encoding is performed on a page-by-page basis of a document, and run-length encoding is performed, or The configuration is such that either predictive coding or run-length coding is selected. In addition, predictive encoding processing has the effect of lengthening the run length for pseudo-halftone images such as dither images and halftone images, but it has the effect of increasing the run length for binary images that are not pseudo-halftone images. Since this has the effect of shortening the length, when a single page of a manuscript contains a mixture of pseudo-halftone images and other binary images such as characters, it is better to run-length encode the predicted image instead of using predictive encoding. This method has the drawback that the compression effect may be worse than when images are run-length encoded.

SUMMARY OF THE INVENTION The present invention is intended to solve the above-mentioned drawbacks, and it is an object of the present invention to provide an image encoding device with a good compression rate for images including pseudo-halftone images.

[Means for solving problems]

The present invention provides an image encoding device including a frame memory, a predictive encoder, and a run-length encoder, which inputs an image signal stored in the frame memory and converts a rectangular area containing a pseudo halftone image of a document into an image encoding device. a whitening means that inputs the image signals stored in the frame memory and stores the image signals as all white image signals, and stores the image signals other than this rectangular area as they are; is stored as is, all image signals other than the rectangular area are stored as white image signals, and the output of the whitening means and the output of the predictive encoder are selected. The above run range gives 7 to the encoder.
The present invention is characterized by comprising a selection means.

[Effect]

The whitening means inputs the image signal stored in the frame memory, sets a rectangular area containing the pseudo halftone image of the original, converts all the image signals into white image signals, and stores them. Store it as is. The image signal stored in the frame memory is input to the extraction means, the image signal of the rectangular area is stored as is, all the image signals other than the rectangular area are stored as white image signals, and then sent to the predictive encoder for prediction. encode. The selection means selects the output of the predictive encoder and the output of the blanking means, and supplies the selected outputs to the run-length encoder for encoding. The above operations can improve the compression rate for images including pseudo-halftone images.

The signal encoded by this circuit can restore the original image by overlapping the whitened image and the cut out image.

A control code indicating that the encoded signal has undergone the processing of the present invention is added, and superimposition is performed during decoding based on this control code.

The rectangular area may be set by either a method of predetermining the area in which the pseudo-halftone image appears, or a method of detecting an input image and setting the area adaptively each time. When decoding, there is no need to know the size or position of this area, and it can be restored simply by overlapping.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings. 1st
The figure is a block diagram of an image encoding device according to an embodiment of the present invention. In FIG. 1, the image encoding device includes a frame memory 1 that inputs and stores an image signal S1 for one page of a document including a pseudo-halftone image, and an image signal that predictively encodes the image signal and includes many short runs. A predictive encoder 6 converts the image signal into a long-run image signal, and a run-length encoder 8 that encodes the image signal using Modified Huffman or Modified Read.

Here, the feature of the present invention is that the frame memory 1
A whitening means inputs an image signal stored in the image signal, sets a rectangular area containing a pseudo halftone image of the original, converts all of the image signals into white image signals, and stores the image signals outside this rectangular area as they are. and a cutting means that inputs the image signal stored in the frame memory, stores the image signal of the rectangular area as it is, converts all the image signals other than the rectangular area into white image signals, and stores the converted image signals to the predictive encoder; The present invention includes a selection means for selecting the output of the whitening means and the output of the predictive encoder and applying the selected output to the run-length encoder.

The whitening means is a whitening circuit that inputs the image signal S12 from the frame memory 1, converts all the image signals in a rectangular area including the pseudo halftone image into white image signals, and outputs the image signals other than this rectangular area as they are. 2, and a white image memory 4 which receives a white image signal 32.4 from the white image circuit 2 and stores one page of a document.

The clipping means includes a clipping circuit 3 which inputs the image signal S12 from the frame memory 1, outputs the image signal of the rectangular area as it is, and outputs all the image signals outside the rectangular area as white image signals; It includes a cut-out image memory 5 which inputs the cut-out image signal S34 from the output circuit 3, stores one page of the original, and supplies it to the predictive encoder 6.

The selector means selects the white plum image signal S'47 from the white plum image memory 4 and the predictive encoded signal S6 from the predictive encoder 6.
It includes a selector 7 that inputs and selects the image signal 5ell to supply the image signal 5ell to the run-length encoder 8, and a part of the control section 10 that supplies the selector control signal 5l (H) to the selector 7.

The image encoding device further includes a selector 9 that selects and outputs the encoded signal SSS from the run-length encoder 8 and the original control code addition signal 5109 from the control unit 10, and a control unit 10 that controls the selector 9. , and a fixed storage circuit 11 that stores the output encoded signal Ss+ of the selector 9.

The operation of the image encoding device having such a configuration will be explained. FIG. 2 is a diagram showing image conversion by the image encoding device of the present invention. In FIG. 2, A indicates an image stored in the frame memory 1, B indicates an image stored in the white plum image memory 4, and C indicates an image stored in the cut-out image memory 5. In the image A of the frame memory 10, a rectangular area including the pseudo halftone image in the image A is filled with white by the whitening circuit 2, and is stored as an image B in the whitening image memory 4. Here, the image is the same as image A except for the rectangular area. Also,
The image A in the frame memory 1 is converted into the image A by the cutting circuit 3.
The area other than the rectangular area including the pseudo halftone image is filled in white and stored as image C in the cutout image memory 5. Here, the rectangular area is the same as the image A. The image B is encoded by a run-length encoder 8 through a selector 7 and stored in a fixed storage circuit 11 through a selector 9. The pseudo halftone image portion of the image C is converted into a long run by the predictive encoder 6. The white part of one line is not changed. The output of the predictive encoder 6 is passed through a selector 7, encoded by a run-length encoder 8, and then sent to a selector 9.
The data is stored in the fixed storage circuit 11 through the process. Here image A
The control unit 10 adds a document control code indicating that the data has been encoded as two images, image B and image C, to the front of the encoded data.

In this embodiment, the predictive encoder does not affect the all-white data for one line, and the run-length encoding effect of the all-white image signal is extremely high, so that the compression rate can be increased.

(Effects of the Invention) As explained above, the present invention has an excellent effect of increasing the compression rate of encoding.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an image encoding apparatus according to an embodiment of the present invention. FIG. 2 is a diagram showing image conversion by the image encoding device of the present invention. FIG. 3 is a block diagram of a conventional image encoding device. 1...Frame memory, 2...White Ume circuit, 3...
- Extraction circuit, 4... Shiroume image memory, 5... Extraction image memory, 6... Predictive encoder, 7, O... Selector, O: Run length encoder, 10... control unit, 11
...Fixed memory circuit, Sl, Sl□, Sfu8... Image signal, S13, S23... Sampling clock, S
21... Image request signal, S22... Predicted image signal,
S24...white plum image signal, S! l... Predicted image request signal, SS2... Run length code signal, S34,
S58...Cut out image signal, S6? ...Prediction encoded signal, 5sss Ss+...Encoded signal, 5cot
...Selector control signal, 5hos...Original control code addition signal. Patent applicant: NEC Corporation - Representative: Patent attorney Naotaka Ide 1. , 11.
-3.・

Claims (1)

[Claims] (1) An image encoding device that includes a frame memory that stores an image signal of a document including a pseudo halftone image, a predictive encoder that predictively encodes the image signal, and a run-length encoder that encodes the image signal. In this step, the image signal stored in the frame memory is input, a rectangular area containing the pseudo halftone image of the original is set, and all of the image signals are converted to white image signals and stored, and image signals other than this rectangular area are left unchanged. a whitening means for inputting the image signal stored in the frame memory, storing the image signal of the rectangular area as it is, converting all image signals other than the rectangular area to white image signals, and storing the image signal as a white image signal; 1. An image encoding device comprising: a clipping means for applying to the run-length encoder; and a selection means for selecting the output of the whitening means and the output of the predictive encoder and applying the selected output to the run-length encoder.