KR100265133B1 - Moving image compression and reproduction system and method and moving image radio signal transmission and reception... - Google Patents

Abstract

An object of the present invention is to significantly reduce the amount of computation in a moving picture compression / reproduction device and the frame memory capacity of image data.

Motion detection / extraction circuit that detects motion in a moving picture signal for one frame, coding circuit for encoding and outputting a moving picture signal, complexing circuit for combining and feeding back a moving picture signal, and moving picture signal of the current frame combined And a frame memory for synthesizing a synthesized circuit and a frame memory for outputting the stored moving picture signal of one frame before the synthesized circuit and storing the synthesized moving picture signal of the current frame, wherein the compression device comprises at least an operation extraction circuit and a frame memory. A binarization circuit is provided at either the front end, the middle end, or the rear end of the image processing loop to convert a dark image signal into a binarized image signal to reduce the signal processing amount and the frame data storage capacity. Further, a reconstruction circuit for converting the binarized image signal into a dark image signal is provided at either the front end or the rear end of the reproduction image processing loop of the reproduction apparatus having the composite circuit, the synthesis circuit and the frame memory.

The signal processing amount of the image processing loop can be reduced from the dark image (about 8 bits) to the binarized image (1 bit), so that the processing speed and the storage capacity of the frame memory can be achieved.

Description

Video compression device and video playback device

1 is a block diagram showing a configuration of a moving picture compression apparatus according to a first embodiment of the present invention.

2 is a block diagram showing the configuration of a moving picture compression apparatus according to a second embodiment of the present invention.

3 is a block diagram showing the configuration of a moving picture compression apparatus according to a third embodiment of the present invention;

4 is a block diagram showing the configuration of a moving picture reproducing apparatus according to a fourth embodiment of the present invention.

5 is a block diagram showing the configuration of a moving picture reproducing apparatus according to a fifth embodiment of the present invention.

6 is a block diagram showing the configuration of a moving picture reproducing apparatus according to a sixth embodiment of the present invention;

Fig. 7 is a block diagram showing the construction of the seventh embodiment in which the moving picture compression apparatus according to the present invention is applied to the transmitting side of a wireless communication terminal.

Fig. 8 is a block diagram showing the construction of the eighth embodiment in which the moving picture reproducing apparatus according to the present invention is applied to a receiving side of a wireless communication terminal.

Fig. 9 is a block diagram showing the construction of the ninth embodiment in which the moving picture compression device and the moving picture reproducing device according to the present invention are applied to a portable television telephone as a two-way communication terminal.

Fig. 10 is a block diagram showing the configuration of the tenth embodiment in which the moving picture compression device and the moving picture reproducing device according to the present invention are applied to a portable television telephone as a two-way communication terminal.

FIG. 11 is a block diagram showing a moving picture compression apparatus incorporating first to third embodiments of the present invention. FIG.

12 is an explanatory diagram for explaining an operation region (movement region).

13 is a block diagram showing a moving picture reproducing apparatus incorporating fourth to sixth embodiments of the present invention.

14 is an explanatory diagram for explaining a range of a reference pixel.

15 is an explanatory diagram for explaining a positional relationship between a reference pixel and an operation region.

FIG. 16 is a block diagram showing a moving picture compression apparatus incorporating first to third embodiments of the present invention. FIG.

17 is a block diagram showing an eleventh embodiment of the present invention.

18 is an explanatory diagram for explaining the positional relationship of the processing order of binarization.

19 is an explanatory diagram for explaining a positional relationship between a reference pixel and an operation region.

20 is an explanatory diagram for explaining a positional relationship between a reference pixel and an operation region.

21 is a block diagram showing a twelfth embodiment of the present invention.

22 is a block diagram showing a thirteenth embodiment of the present invention.

23 is a block diagram showing a fourteenth embodiment of the present invention.

24 is an explanatory diagram for explaining a range of reference pixels in the case of performing vector quantization.

25 is a block diagram showing a fifteenth embodiment of the present invention.

FIG. 26 is an explanatory diagram showing a sixteenth embodiment in which the present invention is applied to a video communication system; FIG.

27 is a block diagram showing a seventeenth embodiment of the present invention.

28 is a block diagram showing an eighteenth embodiment of the present invention.

29 is a block diagram showing a nineteenth embodiment of the present invention.

30 is a block diagram showing a twentieth embodiment of the invention.

31 is a block diagram showing a twenty-first embodiment of the present invention.

32 is a block diagram showing the structure of a conventional moving picture compression apparatus.

33 is a block diagram showing the structure of a conventional moving picture reproducing apparatus.

* Explanation of symbols for main parts of the drawings

2: Motion detection and extraction means (circuit) 6: Coding means (circuit)

7, 12: decoding means (circuit) 8, 13: synthesis means (circuit)

9, 14: frame storage means (memory) 21, 31, 14: binarization means (circuit)

22, 26, 46: reconstruction means (circuit)

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture compression device and a moving picture reproducing device. In particular, in the transmission of moving pictures, halftones for converting a light and shade of an image into low grayscale image data of, for example, two or four values by a predetermined threshold value for each pixel ( A moving picture compression device which performs halftone processing on moving picture data, and a moving picture reproducing device capable of reproducing moving picture compressed data subjected to such halftone processing.

A conditional pixel replenishment system has already been proposed as a technique for compressing and transmitting signal data of a moving picture having a shade (see Hiroshi Harajima, "Image Information Compression" pp. 89-92, Ohm, 1991) ). This conditional pixel replenishment method uses a high correlation at rest in inter-frame prediction, and compares the prediction error between the frames with an appropriate threshold value and concentrates the pixels that need to be transferred to areas where the prediction value greatly deviates. The image data is separated into a pixel) area, and the other part is a pixel (analogous pixel) area in which the predicted value, that is, the value of the previous screen can be repeated. In the transfer of the image data, an address relating to a pixel is transmitted so that a significant / parable pixel can be distinguished, and encoding is performed for the significant pixel by quantizing its prediction error or separately transmitting a prediction error in a field. .

FIG. 32 shows a moving picture compression device to which the above conditional pixel replenishment method is applied, which includes an image input terminal 1 to which an image signal S1 of the current frame is input and an operation (movement) of the input image signal S1. A motion detection / extraction circuit 2 for detecting and extracting the extracted image signal S2, and outputting the encoded signal S3 by encoding the extracted image signal S2, and outputting the encoded signal S3. The decoding circuit 7 which decodes and generates the decoded signal S4, and the synthesis circuit 8 which combines the reference image signal S6 and the decoded signal S4 to generate the reproduced image S5 because the decoded decoded signal S4 has only an operation region component. And a frame memory 9 which accumulates the reproduced image S5 and outputs it as a reference image for the next frame, wherein the encoded signal S3 generated by the encoding circuit 6 supplies the output terminal 10. It is output through the outside.

The motion detection / extraction circuit 2 includes an image cutting part 3 for extracting an operation portion from the input image signal S1 of the current frame, and a reference output from the image signal S1 of this current frame and the frame memory 9. On the basis of the output of the difference circuit 4 and the difference circuit 4 which takes the difference with the image S6 and outputs the difference result S7, it is determined whether or not it is an operation part for each part of the screen by a predetermined determination method, and the determination signal S8 And an operation determination unit 5 for outputting the signal. As a determination method performed by this motion detection / extraction circuit 2, for example, a method of determining as an operation part when the absolute value error is larger than a predetermined predetermined value, and a method of determining that the operation value is not an operation part can be considered.

The encoding circuit 6 converts the motion-extracted image signal S2 into the encoded signal S3 by a Discrete Cosine Transform (DCT) coding method, a vector quantization coding method, or the like.

FIG. 33 shows a moving picture reproducing apparatus for reproducing a coded moving picture signal which is data-compressed and transmitted by the moving picture compression device shown in FIG. In FIG. 33, the encoded video signal S3 input through the input terminal 11 becomes the decoded signal S4 of the operation region by the decoding circuit 12 and is supplied to the combining circuit 13. The synthesizing circuit 13 synthesizes the reproduction image signal S6 of one frame before stored in the frame memory 14 and outputs the reproduction image S5 to the outside. The reproduced picture signal S5 of this current frame is stored in the frame memory 14 and used as the previous frame reproduced picture signal S6 at the time of reproducing the moving picture signal of the next frame.

Since the conventional moving picture compression device and the moving picture reproducing device having the above-described configuration and operation adopt the conditional pixel replenishment method, there is no need to transfer the still part of the screen, so that a large amount of data can be compressed. At this time, since the grayscale image normally uses 256 gray pixel values, an 8-bit data length per pixel is required. Therefore, when the image size is set to "352 x 244" pixels, the frame memories 9 and 14 are luminance signals and must hold 8 x 352 x 244 = 687104 (bits) of data, and a large memory is required. There was a problem with.

In addition, there has been a problem that the amount of calculation of the operation detection / extraction circuit 2, the encoding circuit 6, the decoding circuits 7 and 12, and the synthesis circuits 8 and 13 also increases as the amount of data increases.

On the other hand, as a technique for significantly reducing the amount of data, in the technical field of the still image transfer processing, a halftone process of halftones of images of multi-gradation is reduced in the processing of a shaded image for transmission, hard copying, etc. Several methods have been proposed.

Such binary image processing cuts out a figure portion from an image having a simple shade structure, assigns a value 1 to this figure portion, assigns a value 0 to the other portion, and converts it into a binary image, and then characterizes the shape of the figure portion. The sequence to interpret is used. In order to process the halftone image, a systematic dether method or an average error minimum method is used.

The systematic dither method regards the screen as a set of submatrices of " n × n " dots and determines a threshold only by the coordinate information in this submatrix. The arrangement of the order of the submatrix thresholds is a factor that greatly influences the image quality, and two methods have been proposed, a dot dispersion type that emphasizes resolution and a dot concentration type that emphasizes gradation. A two-step threshold determination method is known as the arrangement at a threshold for obtaining the required number of gradations and resolution performance.

In addition, the average error minimization method weights the error between the reading density and the display density of the reference pixel by the distance from the pixel of interest to binarize the error to the pixel of interest, and binarizes it to a predetermined threshold. This correction value is regarded as a concentration and is also called an error diffusion method. This method can satisfy both gradation expression and resolution, and is also strong in processing such as reduction.

However, such halftone processing is conventionally used in the field of still image processing, and is not used in the display of halftones in the field of moving image processing with a temporal component.

SUMMARY OF THE INVENTION An object of the present invention is to provide a moving picture compression device and a moving picture reproducing device which can reduce the amount of frame memory and at the same time significantly reduce the amount of computation during compression and playback by reducing the data amount of a moving picture signal by halftone processing a shaded image. It is to offer.

The moving picture compression apparatus according to the present invention for achieving the above object comprises: motion detection / extraction means for determining an error between past image data and current image data, extracting an operation region in the moving picture, and outputting motion extracted image data; And a frame storage means for encoding motion-extracted image data and sequentially updating coded image data to store image data for each frame and outputting the image data to the motion detection / extraction means. Halftone processing, which is provided at any one of three points inside the compressed image processing loop including the extraction means and the frame storage means, the front end and the rear end of the compressed image processing loop, to convert the input shaded image into a low gradation image. It is characterized by including a means.

Furthermore, the moving picture reproducing apparatus according to the present invention includes decoding means for decoding input coded image data, frame storage means for sequentially updating image data for each frame to be input, and storing image data for each frame, and decoding. A moving picture reproducing apparatus comprising a combining means for combining reproduced image data and image signals of past frames stored in a frame storing means to generate reproduced image data, comprising: at least a reproducing image processing comprising the combining means and the frame storing means It is provided in any one of the front end and the rear end of the loop, and it is characterized by including reconstruction means for converting the input low gradation image into a shaded image.

The moving picture compression apparatus of the present invention further includes decoding means for decoding a coded image data from the encoding means to generate a decoded image signal, wherein the halftone processing means detects and extracts the motion in the compressed image processing loop. Provided between the means and the encoding means, and between the decoding means and the frame storage means, a combining means for synthesizing the decoded image signal and the image signal of one frame stored in the frame storage means. Between the frame storage means and the motion detection and extraction means, a reconstruction means for converting a low gradation image into a shaded image is provided.

The moving picture compression apparatus further includes a decoding means for decoding a coded image data from the encoding means to generate a decoded image signal, wherein the halftone processing means is provided in front of the compressed image processing loop, and the compressed image The processing loop detects and extracts the operation of the input image signal halftone-processed at low gradation, encodes and outputs this operation data, decodes by the decoding means, and the image signal of one frame before stored in the frame storage means. It synthesizes and updates the frame image data stored in the frame storage means, and the motion detection / extraction means detects and extracts the motion data of the input image of the next frame using this frame image data.

In the moving picture compression apparatus of the present invention, the frame storage means stores input image data for one frame and outputs the same to the motion detection and extraction means, and constitutes the compressed image processing loop together with the motion detection and extraction means. Tone processing means are provided after the compressed image processing loop and between the encoding means.

The moving picture compression apparatus has an average error minimum in which the halftone processing means weights an error between a reading density and a display density of a reference pixel by the distance from the pixel of interest to binarize it by a certain threshold in addition to the pixel of interest. Means for converting a dark-gray image into a low gradation pixel using the method, and storing the low gradation image as a low gradation image of the past, and obtaining an error between the past low gradation image and the current image data to obtain the halftone. It further includes a subtractor which outputs to a processing means.

The moving picture compression apparatus further includes a storage means for storing a plurality of low gradation vectors consisting of a plurality of low gradation data, wherein the halftone processing means determines an output value in which the input image is halftone, in units of a plurality of pixels, It is determined as an output value that the error with the vector of the input image becomes smaller among the low gradation vectors.

The moving picture compression apparatus further includes a storage means for storing a plurality of low gradation vectors consisting of a plurality of low gradation data, wherein the halftone processing means determines the output value in which the input image is halftoned in a plurality of pixel units, An output value is determined by adding an error between the light and dark images and the output image in a predetermined reference range to an error between the vector of the input image and the low gray scale vector.

In the moving picture compression apparatus according to the present invention, since halftone processing means for converting a shaded image into a low gradation image is provided at any one of three points inside, before, and after the picture-axis image processing loop, the frame storage means stores the memory. Since the image data to be processed is halftone, the amount of data is greatly reduced, and there is no need to prepare a large frame memory. In addition, since the computation amount of data during the compression process is greatly reduced, it also contributes to speeding up the computation process.

Further, in the moving picture reproducing apparatus according to the present invention, since reconstruction means for converting a low gradation image into a shaded image is provided at either one of the front end and the rear end of the reproduction processing loop, the half tone processed low gradation coded image data After data transfer is performed, the low gradation image is reconstructed into a shaded image to reproduce a moving image.

Best Mode for Carrying Out the Invention A preferred embodiment of a moving picture compression and reproduction apparatus according to the present invention will be described in detail with reference to the accompanying drawings. In the following description, a binarization process is used as a typical example of halftone processing, and a binary image is used as a typical example of low gradation image.

1 to 6 are block diagrams showing schematic structures of the moving picture compression apparatus according to the first to third embodiments of the present invention and the moving picture reproducing apparatus according to the fourth to sixth embodiments, respectively. . In each drawing, the same reference numerals as those in FIG. 32 denote the same or equivalent components as those in the conventional moving picture compression apparatus. In addition, the path shown by the solid line among the signal paths in each figure is the signal path of the shaded image, and the one shown by the dashed-dotted line is the signal path of the binary image as a low gradation image.

FIG. 1 is a block diagram showing a moving picture compression apparatus according to the first embodiment of the present invention, wherein reference numerals "1" to "10" correspond to respective components in FIG. In Fig. 1, reference numeral 1 denotes an image input terminal, reference numeral 2 denotes an operation detection / extraction circuit including an image cutting unit 3, a differential circuit 4 and an operation determination unit 5. As in the moving picture compression apparatus of the conventional example of 32 degrees, the encoding circuit 6, the decoding circuit 7, the synthesizing circuit 8, and the frame memory 9 are also provided.

The configuration that characterizes the moving picture compression apparatus according to the first embodiment consists of an operation detection / extraction circuit 2, an encoding circuit 6, a decoding circuit 7, a synthesis circuit 8, and a frame memory 9. In the compressed image processing loop 20, a binarization circuit 21 serving as a halftone processing means for converting an input shaded image signal into a binarized image, and the binarization circuit 21 converts the binary image into a binarized image for signal processing. The reconstruction circuit 22 for inverting the converted image data back to a dark image signal is provided. This reconstruction circuit 22 performs reverse halftone processing on halftone processing. In the compressed image processing loop 20, looking at a specific point where the binarization circuit 21 and the reconstruction circuit 22 are provided, the binarization circuit 21 is a motion detection / extraction circuit 2 and an encoding circuit ( It is provided between 6) and the reconstruction circuit 22 is provided between the frame memory 9 and the operation detection / extraction circuit 2.

Therefore, the compressed image data S13 output from the output terminal 10 is also a binarized data signal. The binarization circuit 21 receives the motion extraction image signal S2 output from the motion detection / extraction circuit 2, and converts this signal S2 into a binarization image signal. As the above-mentioned still image, there is a systematic deserialization method and an average error minimization method ("Image Analysis Handbook" Makiou Takaki, supervised by Shimoda Yokyuu, Tokyo University Press, 1991), similar to the still image described above. By using these methods, the gradation of the pixel value is reduced to two values.

The motion-extracted image signal S12 binarized by the binarization circuit 21 is supplied to the encoding circuit 6 to perform encoding processing, and is output to the moving image reproducing apparatus via the output terminal 10, and at the same time, compressed image processing. The loop 20 is also used to process image data input next. That is, it is decoded by the decoding circuit 7 to become a binary decoded signal S14, and is synthesized by the combining circuit 8 with the binary image signal S16 before one frame stored in the frame memory 9. The frame memory 9 stores the synthesized binary image signal S15 of the current frame as an image signal of a new frame. The binarized image signal stored in the frame memory 9 is supplied to the reconstruction circuit 22, where it is converted into a shaded image signal from the binary image signal and output to the motion detection / detection circuit 2.

The reconstruction circuit 22 uses the technique described in the document "Dither Image Edge Detection and Smoothing Process" (Telecommunications Society Journal 1985/6, Vol. 168-D No. 6, pp. 1345-1355). The scaled image is converted into a shaded image and output to the motion detection / extraction circuit 2 as a reference image signal S6. The motion detection / extraction circuit 2 detects the difference between the input image signal S1 and the reference image signal S6 and outputs the difference signal S7 to the operation determination unit 5. The operation determination unit 5 compares the image data of the current frame with the image data of one frame before and cuts out an operation component to extract the operation, and the image signal from which this operation is extracted is supplied to the binarization circuit 21. .

According to this moving picture compression device, since the image signal inputted by the binarization circuit 21 is binarized, the data capacity stored in the frame memory 9 can be compressed to 1/8 of the conventional one. This is because the number of bits per pixel of the image held by the frame memory 9 can be reduced from the conventional 8 bits to 1 bit. For the same reason, the amount of data processed by the encoding circuit 6, the decoding circuit 7 and the combining circuit 8 can also be reduced to 1/8, and the amount of computation to be processed in the image compression processing loop is also reduced. can do. In this first embodiment, since motion detection and extraction is performed, halftone processing is performed on one frame as a reference even if halftone processing is not performed on the entire frame of a moving picture frame, and the subsequent frames are halftone processing of only motion components. This can reduce the amount of transmission.

Next, a moving picture compression apparatus according to a second embodiment of the present invention will be described in detail with reference to FIG. The compression apparatus according to the second embodiment is characterized in that the binarization circuit 31 is provided at the front end of the compressed image processing loop 30. Therefore, the image signal S1 input via the input terminal 1 is supplied to the binarization circuit 31, where it is converted into a binarized image from the shaded image and supplied to the compressed image processing loop 30. The configuration of the compressed image processing loop 30 is the same as that of the conventional processing loop in FIG. 32 except that the signal to be processed is a binarized signal. That is, the motion detection / extraction circuit 2 includes an image cutting section 3, a differential circuit 4, and an operation determining section 5, and further includes an encoding circuit 6, a decoding circuit 7, and a synthesis circuit ( 8) and frame memory 9 are also provided. Compared with the moving picture compression apparatus of the first embodiment shown in FIG. 1, since the signal processed in the compressed image processing loop 30 is a binary signal, it is not necessary to provide a reconstruction circuit to simplify the circuit configuration. Can be.

The operation of the compression device of the second embodiment will be described. The binarization circuit 31 converts the input image signal S1 from the shaded image to the binary image signal S12. As a method of binarization conversion, the method described in the literature mentioned above is used. The binarized image signal S12 is supplied to the motion detection / extraction circuit 2 and supplied to the coding circuit 6 as the binarized image signal S10 from which the motion is extracted. The encoding circuit 6 encodes the motion-extracted binary image signal S10, outputs the binary coded image signal S13 to the moving picture reproducing apparatus via the output terminal 10, and also supplies the signal S13 to the decoding circuit 7. The decoding circuit 7 decodes the encoded image signal and outputs the binary image decoded signal S14 to the combining circuit 8. The combining circuit 8 receives the reference image signal S16 one frame before, synthesizes this and the decoded signal S14, and stores the image signal S15 of the current frame in the frame memory 9.

The frame memory 9 outputs the binarized image signal one frame before to the motion detection / extraction circuit 2 as the reference image signal S16 and newly updates the binarized image signal S15 of the next frame supplied from the synthesis circuit 8. Remember In the motion detection / extraction circuit 2, the difference circuit 4 takes the difference between the input image binarization signal S12 and the binarization reference image signal S16 and outputs the difference binarization signal S17 to the operation determination unit 5. Based on the determination of the operation determining unit 5, an operation component in the input image is extracted and supplied to the coding circuit 6 as the operation extraction binary image signal S10.

According to the moving picture compression apparatus according to the second embodiment, since the binarization circuit 31 is provided in front of the compressed image processing loop 30 as described above, all the signal processing in the loop 30 is converted into a binary signal. It is possible to do this, and has a unique effect that it is not necessary to provide a reconstruction circuit in the loop 30.

Next, a moving picture compression apparatus according to a third embodiment of the present invention will be described with reference to FIG. The moving picture compression apparatus according to the third embodiment includes a frame memory 9 which stores a signal one frame before the input shaded image signal S1 supplied through the input terminal 1, and 1 stored in this frame memory 9. Means for inputting the shaded image signal before the frame as the reference image signal S6 to obtain a difference from the shaded image signal S1 of the current frame, and an operation detection / extraction circuit for outputting the motion extraction shaded image signal S2 based on the difference signal S7 (2) Is provided with a compressed image processing loop 40 constituted by the " The motion extraction shaded image signal S2 output from the loop 40 is supplied to the binarization circuit 41 and supplied to the encoding circuit 6 as the motion extraction shaded image binarization signal S12. The encoding circuit 6 encodes the motion extracting shaded image binary signal S12 and supplies it as a binary coded encoded image signal S13 to the image reproducing apparatus side through the output terminal 10.

Since the moving picture compression apparatus according to the third embodiment is configured to form a loop 40 for motion extraction on the shaded image side and to provide a binarization circuit 41 on the rear end side of the loop 40, the frame memory In (9), the one having the same storage capacity as the conventional one must be employed, but various components such as the decoding circuit, the synthesis circuit, and the reconstruction circuit on the compression side can be omitted. Therefore, there is an effect peculiar to this embodiment that the amount of calculation on the compression side can be reduced and a new simplification of the device can be achieved.

Next, a moving picture reproducing apparatus according to a fourth embodiment of the present invention will be described with reference to FIG. The reproducing apparatus according to the fourth embodiment corresponds to the moving picture compression apparatus according to the first and second embodiments, and the decoding circuit 12, synthesizing circuit 13 and frame memory 14 output a binary image signal. Except for the fact that it is processing, the configuration of the conventional playback apparatus shown in FIG.

In FIG. 4, a reconstruction circuit 26 for converting a binary image signal into a dark image signal is provided at the rear end of the reproduction image processing loop 25 formed by the frame memory 14 and the combining circuit 13. have.

In the above configuration, the binary coded compressed image signal S13 is supplied through the input terminal 11, and this signal S13 is decoded by the decoding circuit 12 and decoded by the decoding circuit 12 to the synthesis circuit 13 as the decoded binary image signal S14. Supplied. The synthesizing circuit 13 synthesizes the binarized frame image signal S16 and the decoded image signal S14 before one frame stored in the frame memory 14, and outputs the binary image signal S15. The binarized image signal S15 is supplied to the reconstruction circuit 26 to convert the binarized image signal into a dark image signal S5, and is output to the outside via the output terminal 15 as a reproduced image signal.

According to the moving picture reproducing apparatus according to the fourth embodiment, since the amount of image data stored in the frame memory in the reproduction image processing loop 25 is 1/8 bit, the large capacity of the memory, which is the biggest problem in the image information processing, is suppressed. There is a peculiar effect that can speed up processing and reduce the apparatus cost.

Next, a moving picture reproducing apparatus according to a fifth embodiment of the present invention will be described with reference to FIG. The device of the fifth embodiment shown in FIG. 5 has the same configuration as the device of the fourth embodiment shown in FIG. 4 except that a selection circuit is provided in front of the output terminal 15. In the fifth embodiment, when the reproduced image signal is displayed on a binary display or outputted in binary by a printer or the like, the binary image signal and the dark image signal can be switched by an external switching signal.

That is, the reconstruction circuit 36 is provided in the rear end of the reproduction image processing loop 35, and it is possible to convert the binarized image signal S15 into a shaded image signal S5 and output it. The configuration up to this point is the same as the playback apparatus of the fourth embodiment shown in FIG. A selection circuit 37 is provided at the rear end of the reconstruction circuit 36, and the selection switching signal S18 is supplied to the selection circuit 37 from the external input terminal 38, and the binary image is also obtained from the synthesis circuit 13. The signal S15 is supplied.

Therefore, when it is necessary to output the binarized image signal as mentioned above, the binarized image signal S15 can be output through the output terminal 15 by the external switching signal S18 supplied from the external input terminal 38. As shown in FIG. have.

Next, a moving picture reproducing apparatus according to a sixth embodiment of the present invention will be described with reference to FIG. The reproducing apparatus according to the sixth embodiment corresponds to the moving picture compression apparatus of the third embodiment shown in FIG. 3, and is reproduced image processing loop 45 formed by the synthesizing circuit 13 and the frame memory 14. The reconstruction circuit 46 is provided in front of the.

The operation of the playback apparatus according to the sixth embodiment will be described. The binary compressed image signal S13 supplied through the input terminal 11 is decoded by the decoding circuit 12 and supplied to the reconstruction circuit 46 as the binary decoded image signal S14. The reconstruction circuit 46 converts this binarized signal S14 into a shaded image signal S4 and supplies it to the synthesis circuit 13. The combining circuit 13 receives the image signal of one frame before as the reference image signal S6, synthesizes it into the reproduced image signal S5, and outputs it externally through the output terminal 15.

Although the specific effects cannot be obtained in the sixth embodiment, the compression apparatus of the third embodiment and the reproduction of the sixth embodiment can be performed in that signal processing can be performed by using a binarized image signal in the compression device axis. By combining the apparatuses, the present invention is effective.

Incidentally, the above-described first to sixth embodiments are applicable to both wired and wireless transmission methods, and show configurations between the input and output terminals on the transmitting side and configurations between the input and output terminals on the receiving side, respectively. 7 and 8 show the seventh and eighth embodiments in which the moving picture compression device and the moving picture reproducing device according to the present invention are applied to the wireless transmission method, respectively.

7 is a block diagram showing a moving picture compression apparatus according to the seventh embodiment of the present invention. The compression device of the seventh embodiment is a specific example applied to a moving picture radio transmission device, and wirelessly transmits a halftone processed low gradation image by using a radio frequency carrier and a signal.

The moving picture compression transmission device 50 shown in FIG. 7 performs halftone processing on the camera 51 which photographs a subject and receives it as a shaded image, and the shaded image signal S1 supplied from the camera 51. A halftone processing circuit 52 that outputs S12, a modulation circuit 53 that modulates a low gradation image signal S12, and outputs a moving image signal S20 loaded on a carrier signal, and an antenna 54 for transmitting the moving image signal S20; Equipped. The modulation circuit 53 generates a moving picture signal S20 having a predetermined transmission frequency, and the antenna 54 radiates radio waves S21 into the air.

8 is a block diagram showing a moving picture reproducing apparatus according to the eighth embodiment of the present invention. The reproduction apparatus of the eighth embodiment receives and reproduces the radio wave S21 transmitted through the compression apparatus of the seventh embodiment, for example, and performs half-tone processing on the demodulated low gradation image signal to reconstruct it into a shaded image.

The moving picture reception / reproduction device 55 shown in FIG. 8 receives the antenna 56 for receiving the radio wave S21 and the radio wave S21 received by the antenna 56 as the moving image signal S20 and demodulates it into a low gradation image signal S22. A demodulation circuit 57, a reconstruction circuit 58 for performing reverse halftone processing on the low gradation image signal S22 and reconstructing it as a shade image signal S5, and a display 59 for displaying an image based on the shade image signal S5 on a screen. ).

9 is a block diagram showing a moving picture compression and reproduction device according to the ninth embodiment in which the present invention is incorporated into a portable television telephone.

In FIG. 9, the portable television telephone 60 is separated into a compression side and a reproduction side through a transmission / reception circuit 65 having an antenna 54 or 56. As shown in FIG. The compression side uses the camera 51, the microphone 51, the image encoder 62 which encodes the image signal S1 photographed by the camera 51, and outputs the image code S24, and the acoustic-electricity by the microphone 61. A speech encoder 63 for encoding the converted speech signal S25 and outputting the speech code S26, and a multiplexer 64 for multiplexing the image codes and the speech codes output from the respective encoders to generate the multiplexed signal S27. . The transmission / reception circuit 65 converts the multiplexed signal S27 to a predetermined frequency and emits the transmission signal S20 as a radio wave through the antenna 54 or 56.

Transceiver circuit 65 receives radio waves from other portable television phones through antenna 54 or 56 and supplies them to separator 66 as multiplexed signal S28. The reproduction side receives the multiplex signal S28, and divides it into a separator 66 for separating the image code S29 and the voice code S30, an image decoder 67 for decoding the separated image code S29 to generate the image signal S5, and the separated voice code S30. Is provided with a speech decoder 68 that decodes the speech signal S31 to generate the speech signal S31, a display 59 that reproduces and displays the image signal S5, and a speaker 69 that converts the speech signal S31 into speech and outputs the speech.

In the above configuration, the image signal S1 received by the camera 51 and the audio signal S25 received by the microphone 61 are transmitted by the image encoder 62 and the audio encoder 63 at the time of transmission. It is encoded by voice code S26. The image encoder 62 performs image compression processing by detecting and extracting an operation during a moving image simultaneously with image encoding. This compressed image processing can drastically reduce image data to be transmitted by providing the halftone processing circuit or the binarization circuit according to the above-described embodiment in the front end or the loop of the compression processing loop. Similarly, the capacity of the frame memory in the image compression loop can also be reduced.

In addition, also on the receiving side, the image decoder 67 performs decompression of the compressed image data using a reproduction image processing loop at the same time as decoding of the image code. Therefore, by providing a reconstruction circuit for performing reverse halftone processing at the front end or the rear end of a reproduction image processing loop including a frame memory and a synthesis circuit, it is possible to process and reconstruct image data to be reproduced with a small amount of data. In this manner, bidirectional wireless communication using low gradation image data and audio data becomes possible.

The portable television telephone apparatus according to the tenth embodiment will be described with respect to the appearance of the bidirectional compression playback apparatus according to the ninth embodiment shown in FIG. As shown in FIG. 10, the portable television telephone apparatus 70 according to the tenth embodiment includes a camera 71, a display 72, a microphone 73, a speaker 74, An operation key 75 is provided, and an antenna 76 for transmitting and receiving is disposed above the main body. The other electronic circuits have the same configuration as that of the ninth embodiment shown in FIG. 9, for example, and are housed in a case of the main body.

All of the seventh to tenth embodiments described above show a configuration example in the case where the moving picture compression and reproduction apparatus according to the present invention is applied to a wireless information communication terminal. Although it is possible to apply to various communication terminals as described above, any modifications and changes are free as long as it includes not only the illustrated embodiment but also reconstruction as halftone processing and reverse processing of the image that is the subject of the present invention. For example, when the apparatus of the present invention is applied to a reception-only communication terminal, it is basically possible to include at least three components of a receiver such as an antenna, a decoder having reconstruction means, and a display output unit such as a display.

When the moving picture compression apparatuses according to the first to third embodiments described above are combined, it becomes as shown in FIG. Here, halftone means that a shaded image having a real value of 0 to 1 as a luminance value is represented using only two values of zero or one. In FIG. 11, the memory 9 has previously stored the shaded image one frame before. The shaded image S1 of the current frame is input to the image cutout part 3 and the subtractor 4. The shaded image S6 one frame before is provided from the memory 9 to the subtractor 4. From the subtractor 4, a difference S7 between the shaded image S1 of the current frame and the shaded image S6 one frame before is provided to the operation determining unit 5 for each part of the frame. The operation determination unit 5 sends the operation information S8 to the cutout unit 3 with the difference S7 larger than the predetermined value as the operation region and the other as the stop region.

For example, as shown in FIG. 12, the frame is divided into blocks, and each block determines whether it is a stationary block or an operation block. In Fig. 11, the cutout section 3 cuts out only the image S2 of the operation region from the shaded image S1 of the current frame by the operation information S8 and provides it to the binarizer 21 and the memory 9. The memory 9 is overwritten with the image S2 of the operation area, and prepares for motion detection of the next frame. In the binarizer 21, the shaded image S2 is halftoned, and the binary image S12 is provided to the encoder 6. The encoder 6 encodes the binary image S12 by entropy coding, vector quantization, or the like, and outputs the coded signal S13 to the outside.

The playback apparatus corresponding to this is shown in FIG. The memory 14 stores a reproduction binary image one frame before. The coded signal S13 is restored to the binary image S14 by the decoder 12 and sent to the memory 14. In the memory 14, the operation area is overwritten with the binary image S14 and output as the reproduced binary image S5 of the current frame.

By the way, there are various methods of halftoning, but one of them is the average error minimization method. In this technique, for example, as described on page 495 of the "Image Analysis Handbook" (Makiki Takaki, Supervised by Shimoda, Tokyo University Press, 1991), the value of each image is determined in order in the upper left of the image. The image value shown in Fig. 14 uses a predetermined image as a reference image. The value of the processed image is determined so that the error of the luminance value due to binarization in the reference image as well as the processed image is reduced, and the average error in the range of "processed image + reference image" is reduced.

Since the average error minimization method obtains a high quality binary image in which the resolution and gradation of an original picture are relatively maintained, there is a demand to use it for the binarizer 21 of FIG. In that case, since the image of the still area is excluded from the cutout part 3, the binarizer 21 cannot be used as a reference pixel. For example, in the case where the processed image has the upper end of the operation region as shown in FIG. 15, the error of the upper two rows cannot be referred to and binarized. When the binary image of the operation region thus created is combined with the binary image of the still region one frame before, the boundary between the still region and the operation region is easily perceived, and the quality of the reproduced image is deteriorated.

Further, as shown in FIG. 16, the binarizer 21 is disposed in front of the cutout 34 so that the still region can also be referred to the binarization of the operation region. However, since the binary image of the still area, which is also referred to in this case, is not transmitted eventually, the error there is different from the error in the same part of the reproduced image. As a result, distortion occurs at the boundary between the operation region and the still region of the reproduced image.

As described above, in the apparatus according to the first to tenth embodiments described above, there is a disadvantage that the boundary of the portion not to be transmitted is distorted to deteriorate the image quality.

In order to solve this problem, in the embodiments described below, motion detection / extraction means for determining an error between past image data and current image data, extracting an operation region in a moving image, and outputting motion extraction image data; Halftone processing means by the average error minimum method for changing the motion-extracted image data into a low gradation image, encoding means for encoding the low gradation image to output encoded image data, and image data for each input frame. Frame storage means for sequentially updating and storing image data of each frame and outputting the image data to the motion detection / extraction means, means for storing the low gradation image as a past low gradation image, and the past low gradation image And a subtractor for calculating an error of the current image data and outputting the error to the halftone processing means. And to include.

In the above-described embodiment, the low gradation image of one frame before is stored in advance, and the error between the low gradation image and the current shaded image is obtained, and the error is obtained from the reference pixel of the binarization of the operation region of the current frame. It can be used as. As a result, it is possible to reproduce a code that can be reproduced by an image having no distortion at the boundary between the untransmitted portion and the transmitted portion.

Hereinafter, a moving picture compression apparatus according to an eleventh embodiment of the present invention will be described with reference to the drawings.

17 is a block diagram showing an eleventh embodiment of the present invention. The operations of the memory 9, the cutout section 3, the subtractor 4, and the operation determination section 5 in FIG. 17 are the same as those in FIG. In another memory 81, the reproduction binary image one frame before is stored in advance. The shaded image S1 of the current frame is also input to the subtractor 82, and the reproduction binary image S40 one frame before read out from the memory 81 is also input to the subtractor 82. In the subtractor 82, the difference between the shaded image S1 of the current frame of each pixel and the reproduction binary image one frame before, that is, the error S41, is calculated and provided to the binarizer 21.

The image S2 of the operation region is input to the binarizer 21, and binarization is performed in order from the pixels on the left side of the operation region as shown in FIG. Here, binarization is performed by the average error minimum method with reference to the error S41. In the average error minimization method, as shown in FIG. 14, " error = light weight value-weighted average value S of binary values " in a predetermined reference pixel around a processing pixel to be binarized is first obtained. And when the luminance value f of a process pixel is larger than "1/2-S", it is determined as "binarization output of the corresponding pixel = g = 1", and when it is small, "g = 0". This is also equivalent to selecting g = 0 or 1 in which the distortion amount Z = | f + S-g | becomes smaller.

Although the processing pixel is naturally in the operating area, there are cases where the reference pixel is in the operating area and in the still area as shown in FIGS. 15, 19 and 20. FIG. Therefore, when the reference pixel is in the operation region, the error in the reference pixel = light gray value-the binary value of the current frame, and when the reference pixel is in the still region, the error S41 is used, that is, the error in the reference pixel. = Joke value-2 values before 1 frame. By doing this, distortion is not generated at the boundary between the operation area and the still area even when the current image is displayed as the current frame with only the operation area as a new value in the binary image one frame before playback.

Referring to FIG. 17 again, the binary image S12 is sent from the binarizer 21 to the memory 81 and the encoder 6. The memory 81 overwrites the operation area with the binary image S12, and prepares for processing of the next frame. In the encoder 6, a binary image is encoded by entropy encoding, vector quantization, or the like, and a code S13 is output to the outside. Reference numeral S13 is reproduced by the decoder of FIG. 13 described above.

Next, FIG. 21 shows a block diagram of a moving picture compression apparatus according to the twelfth embodiment utilizing motion compensation. 17 differs from the operation compensator (MC) 78 is inserted between the memory (9) and the subtractor (4). The motion compensator 78 also inputs the shaded image S1 of the current frame and performs matching by shifting the shaded image S6 one frame earlier and laterally, thereby detecting the motion vector of the entire frame, and subtracting the shaded image one frame previous one frame S42. (4) to send. The following is the same as FIG. 17, so description is omitted.

Next, FIG. 22 shows a moving picture compression apparatus according to the thirteenth embodiment in which the binarization method is switched by blocks. The image S2 of the operation area is input to the block type determiner 83 for each block. In the block type determiner 83, for example, when the variance of the block is smaller than the predetermined value, it is determined as a flat block, and the average value S43 of the luminance in the block is obtained and output to the outside. Then, one luminance value in the block is not used, and a binary image is created from the average value alone by a suitable method, and the binary image S42 of the flat block is overwritten in the memory 81. In the binarizer 21, only the image S44 of the uneven block is sent. In the halftone image, since the average luminance is perceived at a ratio of 0 and 1 in the block, when the average value can be known, 0 and 1 may be randomly arranged, for example, at a ratio that matches the average value. In a flat block, there is no need to represent any type in the arrangement of zeros and ones.

Next, a moving picture compression apparatus according to the fourteenth embodiment using the vector quantizer as the encoder 6 is shown in FIG. The binarizer 21 generates a binary image S45 from the operation area image S2 with reference to the error S41, and outputs it to the codebook generator 84. FIG. The codebook generator 84 groups the binary images S45 for each predetermined number of pixels and treats them as vectors. For example, the pixel value of the binary image S46 is

001101100100100011010110

011010000011110110010000

If the number of elements in the vector is set to four,

(0011) (0110) (0100) (1000) (1101) (0110)

(0110) (1000) (0011) (1101) (1001) (0000)

Treat as a vector Then, a plurality of vectors having a high appearance frequency are selected, and the indexes are associated with each other to form a codebook S46. In the example above,

0011 2

0110 three

1 0100

1000 2 pieces

2 1101

1 1001

0000 1

Therefore, when the index is 2 bits, the codebook S46 is

Index00 vector0110

Index 01 vector

Index10 Vector1000

Index11 Vector1101

Becomes The codebook S46 is sent to the memory 85 and simultaneously to the playback apparatus.

The image S2 and the error S41 of the operating area are also sent to the vector quantizer 86. The vector quantizer 86 also treats the operation region image S2 as a vector, and outputs the index of the vector providing the smallest vectorization distortion amount V as the symbol S47 of the vector. Here, the vectorized distortion amount V assumes that the vector of the image S2 in the operation region is [f1, f2, f3, f4] and the binary vector S48 read from the codebook is [g1a, g2a, g3a, g4a].

V = Z1 + Z2 + Z3 + Z4

= | F1 + M1-g1 | + | f2 + M2-g2 | + | f3 + M3-g3 | + | f4 + M4-g4 |

Obtain as M1 to M4 are weighted average values of errors obtained from the error S41, M1 is calculated in the range of the reference pixel 88 of f1 shown in FIG. 24, and M2 is calculated in the range of the reference pixel 89 of f2. In f1 included in the reference pixel 89 of f2, since the binary pixel value is not yet determined, the error therein uses f1-g1a. Similarly, M3 and M4 are obtained by shifting the range of the reference pixel. Thereby, image compression can be performed, keeping both the error of binarization and the error of vectorization small.

The codebook can also be stored in the memory 85 in advance. In this case, the binarizer 21 and the codebook generator 84 are unnecessary and there is no need to send the codebook to the playback apparatus.

The sign S47 output from the vector quantizer 86 is input to the vector dequantizer 87. The dequantizer 87 reads the binary vector S48 corresponding to the index from the codebook memory 85, reproduces the binary image S49, and overwrites the operation area portion of the memory 81.

25 shows a playback apparatus according to the fifteenth embodiment corresponding to the moving picture compression apparatus of FIG. Codebook S46 is stored in codebook memory 85, and code S2 is input to vector dequantizer 87. The binary image S49 reproduced by the vector inverse quantizer 87 is overwritten by the operation area portion of the memory 81, resulting in the reproduction binary image S40.

26 shows a sixteenth embodiment in which the present invention is used in a video communication system. This is a system 90 for transmitting a moving picture through wireless communication, and the moving picture binarization apparatus of the present invention is attached to the EWS 91. The image around the EWS 91 is input by the camera 92 as a shaded image S1, and the code S50 of the result of binarization and encoding is sent to the radio apparatus 93. As shown in FIG. In the radio 93, the signal is modulated to a predetermined transmission band, and the signal 94 is received by the network 100 and the signal 95 at the other radio 96 and demodulated by code S51. Reference numeral S51 is reproduced by the image reproducing apparatus in the personal computer 98, and displayed on the display 99 of the personal computer 98. In this system 90, since the code | symbol by the compression apparatus of this invention is used, a reproduced image has less distortion than before. In addition, if a compression device is installed in the personal computer 98 and a playback device is installed in the EWS 91, the image transmission from the personal computer 98 to the EWS 91 can also be performed. Communication is also possible.

27 is a block diagram showing a seventeenth embodiment of the present invention. In FIG. 27, the multi-gradation image signal S1 of the current frame input from the image input terminal 1 is sent to the image processing circuit 101. As shown in FIG. In the image processing circuit 101, halftone processing and encoding are performed on the image signal S1, and the coded signal S13 is sent to the code amount monitoring circuit 102.

Examples of the encoding method in the image processing circuit 101 include (1) a method of extracting and outputting only a portion of the image signal S1 which has an operation (movement) as compared to the previous frame, and (2) vector quantizing a halftone image. A method of outputting an index, (3) a method of outputting only the average value of the flat portion, and a combination of one or a plurality of these (described in detail in another embodiment) are used.

Since the code amount of the generated coded signal S13 fluctuates greatly in time due to the fineness of the picture or the intensity of the motion, the coded signal S13 cannot be directly output to the communication path or the like. Therefore, the code amount monitoring circuit 102 passes only signals within a predetermined allowable code amount of the coded signal S13, and this signal is output from the output terminal 10 as the transmission signal S55. In addition, the code amount monitoring circuit 102 sends a control signal S56 for controlling the amount of code generated to the image processing circuit 101 as necessary. In the image processing circuit 101, by the control signal S56, for example, in the case of (1): in the case of the threshold value (2) dividing the operation portion / stop portion: in the case of the number of bits (3) of the index: the flat portion / non-flat portion Encoding parameters, such as the threshold value which divides into, are switched, and the amount of generated codes is increased or decreased.

The code amount monitoring circuit 102 changes in configuration depending on the transmission condition to which the output terminal 10 is connected. FIG. 28 shows a configuration example of the eighteenth embodiment when the output terminal 10 is connected to a communication path in which a transmission code amount per unit time is determined. The coded signal S13 is once stored in the memory 103 once. Then, the temporarily stored signal is read out from the memory 103 by a predetermined amount in the old order and output as the transmission signal S55. The memory amount check circuit 104 always checks the memory amount of the memory 103, and generates a control signal S56 which reduces the generated code amount when the memory amount increases. On the contrary, when the memory amount approaches zero, a control signal S56 is generated which increases the generated code amount.

Next, although the transmission amount per unit time may fluctuate to some extent, when the allowable maximum value of the average transmission amount in a predetermined time is determined, it is comprised as shown in FIG. 29 as a 19th Example. The coded signal S13 is output as the transmission signal S55 via the switch 106 as it is, but the counter 105 always counts and counts the transmitted code amount within the latest predetermined time, and the count value exceeds the maximum value of the code amount. In this case, the switch 106 is disconnected to pause transmission of the transmission signal S55. Then, when the count value becomes small enough, reconnect the switch. When the coded signal S13 is too large and the switch must be disconnected frequently, the control signal S56 is generated so as to reduce the amount of the coded signal S13 and, conversely, to increase the code amount when the transmission amount is sufficient.

As described above, in the seventeenth to nineteenth embodiments, a transmission signal suitable for a transmission condition according to the code amount can be generated.

30 is a twentieth embodiment of controlling a frame rate, i.e., the number of transmission frames per unit time. In the case of moving pictures, if 30 frames are transmitted within 1 second, the receiving side can play smooth images.However, if the communication channel or recording medium does not have enough capacity, the frame rate must be reduced to 10 frames / second or 5 frames / second. There is.

From the input terminal 1A, the encoded image signal S57 having a high frame rate, for example, 30 frames / second is input and sent to the switch 108. In the input terminal 1B, the frame synchronizing signal S58 in which a pulse is formed at the beginning of every frame is input, and sent to the counter 109. In the counter 109, the number of pulses of the frame synchronizing signal S58 is counted. For example, when the number of pulses of the frame synchronization signal S58 is desired, the switch 108 is connected only when the count value is divided by three. Otherwise, the switch 108 is disconnected. A switching signal S59 is sent to the switch 108.

The image signal S1 passing through the switch 108 is input to the halftonization and encoding circuit 107 which is an image processing circuit, and is halftoneized and encoded and then output from the output terminal 10 as the encoded signal S13.

Thus, in the twentieth embodiment, the image can be transmitted at a predetermined frame rate.

FIG. 31 shows the twenty-first embodiment in which the function of switching the frame rate is added to the embodiment in which the amount of generated codes is controlled by switching the encoding parameter described above. The code amount monitoring circuit 102 sends a switch signal S59 to the switch 108 to disconnect the switch 108 when it is determined that the amount of generated codes is too large or that the allowable transmission amount cannot be suppressed only by switching the coding parameters. Suspends input. This suppresses unnecessary cases in which the amount of transmission is kept below the allowable value even when an image of very severe operation is suddenly inputted, and the half-toning or encoding of the image that cannot be transmitted is performed by the image processing circuit 107.

As described above, according to the present invention, the moving picture compression apparatus is provided with binarization means for converting a shaded image signal into a binary image signal at any point inside, before, and after the compression signal processing loop, and reproduced image processing of the moving picture reproducing apparatus. Since the reconstruction means for converting the binarized image signal into a dark image signal is provided at either the front end or the rear end of the loop, it is possible to significantly reduce the image data throughput on both the compression device side and the reproduction device side, and in particular, On the compression device side, binarization means are provided at the front or middle of the processing loop, and on the reproduction device side, reconstruction means are provided at the rear end of the processing loop, whereby the data storage capacity of the frame memory can be drastically reduced. It is also possible to solve the maximum problem in processing.

Moreover, although the technique of reducing the throughput of data by performing halftone processing on image data has already been implemented in the field of still image processing as described above, the present invention simply applies halftone processing to the field of moving image processing. Instead, using the correlation between frames specific to the field of moving picture processing, the operation part between the frames before and after is detected and extracted, and only this operation information is processed and transmitted, and the still part in the moving picture is not processed and transmitted. Therefore, the data transfer amount can be reduced by that amount.

Further, according to the invention, it is possible to reproduce an image without distortion at the boundary between the untransmitted portion and the transmitted portion.

Claims (8)

Motion detection / extraction means for determining an error between past image data and current image data, extracting an operation region in a moving image, and outputting motion extraction image data; and encoding for encoding motion extraction image data and outputting encoded image data. And a frame storage means for sequentially updating image data for each frame input sequentially, storing image data of one frame, and outputting the image data to the motion detection / extraction means, wherein at least the operation is performed. Halftones provided at any one of three points inside the compressed image processing loop including the detection / extraction means and the frame storage means, the front end and the rear end of the compressed image processing loop, and convert the input shaded image into a low gradation image. A moving picture compression apparatus, comprising processing means. 2. The apparatus according to claim 1, further comprising decoding means for decoding a coded image data from said encoding means to generate a decoded image signal, wherein said halftone processing means comprises: said motion detection / extraction means in said compressed image processing loop; Provided between the encoding means, a combining means for combining the decoded image signal and the image signal of one frame stored in the frame storage means is provided between the decoding means and the frame storage means. And a reconstruction means for converting a low gradation image into a shaded image between the means and the motion detection / extraction means. 2. The apparatus according to claim 1, further comprising decoding means for decoding a coded image data from said encoding means to generate a decoded image signal, wherein said halftone processing means is provided in front of said compressed image processing loop, The processing loop detects and extracts the operation of the input image signal halftone-processed at low gradation, encodes and outputs this operation data, and decodes it by the decoding means to store the image signal one frame before. And the frame image data stored in the frame storage means, wherein the motion detection / extraction means detects and extracts motion data of the input image of the next frame using the frame image data. Moving picture compression device. 2. The frame storing apparatus according to claim 1, wherein the frame storing means stores input image data for one frame and outputs the same to the motion detecting and extracting means, and forms the compressed image processing loop together with the motion detecting and extracting means. And a tone processing means is provided between the encoding means and the encoding means behind the compressed image processing loop. The average error of claim 1, wherein the halftone processing means weights an error between the read density of the reference pixel and the display density by the distance from the pixel of interest to binarize the error to a pixel of interest by a predetermined threshold value. Means for converting a dark image into a low gradation pixel using a minimum method, storing the low gradation image as a low gradation image of the past, and obtaining an error between the past low gradation image and the modern image data A moving picture compression apparatus, further comprising a subtractor output to the tone processing means. 2. The apparatus according to claim 1, further comprising a storage means for storing a plurality of low gradation vectors comprising a plurality of low gradation data, wherein the halftone processing means determines an output value in which the input image is halftone, in units of a plurality of pixels. And an output value for determining that an error with the vector of the input image is small among the low gradation vectors. 2. The apparatus according to claim 1, further comprising a storage means for storing a plurality of low gradation vectors comprising a plurality of low gradation data, wherein the halftone processing means determines the output value of the half-toned input image in a plurality of pixel units. And the output value is determined by adding an error of the light and dark images in a predetermined reference range to an error between the vector of the input image and the low gray scale vector. Decoding means for decoding input coded image data, frame storage means for sequentially updating image data for each frame input sequentially, and storing image data for each frame, and storing in decoded image data and frame storage means. A moving picture reproducing apparatus comprising a combining means for synthesizing a picture signal of a past frame to generate reproduced image data, wherein at least one of the front end and the rear end of a reproduced image processing loop including at least the above combining means and frame storage means And a reconstruction means for converting the input low gradation image into a shaded image.