MXPA03006756A - Moving picture information compressing method and its system. - Google Patents

Abstract

A moving picture information compressing method and its system for compressing, at high compression rate and at high speed, data such that image information can be prediction-coded and for improving the quality of image. In a first working example, the image in one frame is divided into blocks before the inter-frame compression, and each block is approximated to (replaced with) a single plane defined by three elements of the pixels in the block. In a second working example, the pixel of an original image is compared with that of its decompressed compressed-image, information about the difference of each pixel element is outputted, and the smaller block size is applied if a pixel element which leads to a difference larger than a parameter (threshold) P is present, so as to carry out intra-frame compression. In a third working example, I blocks produced by spatially dividing an image are dispersed along the time axis between frames in such a way that no I block is inserted at a block position in a frame updated when the difference between the frames larger than the parameter (threshold) P occurs in a designated period.

Description

METHOD OF COMPRESSION OF IMAGE INFORMATION IN MOTION AND SYSTEM TECHNICAL FIELD The present invention relates to a method and system for compressing image information in motion, which can compress, with a high percentage of compression and at high speed, data or image information that can be subjected to prediction coding, and which can improve the quality of the image.

Prior Art Conventionally, it is a general procedure to convert an image signal to another type of a signal, and assign suitable codes to that converted signal based on the statistical characteristics of the converted signal, and transmit the resulting encoded signal. In this case, the so-called prediction coding, which can compress the information with a high compression ratio, is done for a redundant image inside a frame or an image that includes a regular pattern or a flat pattern, in such a way that : because there could be a high correlation between the adjacent pixels (pixels), it is possible to predict, to some degree, the next value of the pixel to be encoded from a pixel value that has already been encoded; only the components that could not be predicted are extracted and coded. In the case of a moving image in a videophone or the like, because the adjacent frames are frequently similar to one another, temporal changes are limited; In accordance with the above, this temporary redundancy can be removed by means of the inter-frame prediction coding, which makes the prediction between the frames. At this time, the so-called block-based coding can generally be performed in such a way that: a block code is used in which a keyword is assigned to a single symbol; each frame is divided into a plurality of pixel blocks; using the feature that the difference in luminance within each block is smaller, the information is compressed. Huffman coding is known as a method to generate a high efficiency code, which is one of the entropy coding that data compression can achieve by assigning a high efficiency code for the converted signal. One representative of it is arithmetic coding, which generates one keyword after another through an arithmetic calculation by dividing a line of probabilistic numbers into segments according to the probability of occurrence of each sequence of symbols, and determine to accept the tenth binary indicating a location in a segment as a code for the sequence of symbols. A conventional three-step block coding system for efficiently encoding an image signal comprises the steps of sampling, transforming, and quantizing. In order to retain the two-dimensional resolution and the high-frequency components for a given image signal, it is generally required to perform sampling at a frequency of twice the highest frequency component. With MPEG, it is preferable that the coding efficiency be as high as possible so that images can be compressed with a large amount of information. In accordance with the above, there is conventional forward prediction coding (P-frame based processing), which uses as a prediction signal only the past image signal that has already been encoded, and the two-way prediction encoding (processing). based on frame B), which uses as a prediction signal a future image signal, as well as the past image signal. The conventional interframe prediction coding performs the transmission of the difference signal between an input image signal and the corresponding prediction image signal, and the decoding side performs the addition of the difference signal transmitted to the signal of Prediction image that was already decoded so as to reconstruct the original image. In this way, it is impossible for the decoding side to reconstruct on the basis of inter-frames if the prediction image signal is not provided. In accordance with the above, not using the past and future image signals as prediction signals, but using an intra-frame coded I frame (ie, a reference frame that allows the reconstruction of an image from only that frame), this I frame is inserted inside a sequence of frames at a fixed interval, which enables the reconstruction of an image partially through the sequence of the same and that solves possible data errors. However, because the conventional image signal compression technique employs the complicated block-based coding method, it is difficult to compress with a high compression ratio and at high speed, the image data such as the audio that can be subjected to prediction coding. When the difference information is generally compressed through a moving image compression procedure, that is, when the successive values Al and? 2 are expected to be similar to each other and when the Al value is known before the occurrence of the value? 2, assuming that the probability that the occurrence of the difference between? 2 and? 1 is equal to or close to zero is high, the compression is performed using conventional Huffman coding or arithmetic coding; in accordance with the above, if Al and A2 can each take one of the values 0, ..., n, the difference between? 2 and Al can have one of the values 2n + 1, so that the keywords of Huffman 2n + 1 are necessary. Because there are actually n possible values for? 2, but not all n codes are used locally, redundant keywords can be generated. In addition, there is a problem that when the difference between the frames is large, the quality of the image deteriorates intensely and it is impossible to provide a high image quality. Also, if a larger block size is used, the compression percentage can be improved; however, the detail of the original image may be lost, which causes the quality of the image to deteriorate. A phenomenon occurs where the fine lines can be lost completely when the original image is configured by those fine lines, differing their intensity from those of a fixed color background. In addition, because the frame I that is to be inserted periodically into a frame sequence is subjected to intra-frame coding, the coding efficiency is worse compared to that of inter frame coding, which encodes the difference between the tables r so that the amount of information that is generated is increased; therefore, in the case of a high-speed communication line that is not available, the insertion frequency of frames I is limited. Also, because the amount of data within a frame I is between two and ten times more than the data within a differential box, that technique is against a fixed proportion that is required to enable communication. This is, conventionally, because the I frames are inserted periodically within a sequence of frames, the processing time is quite long so that the visual display of the image that was reconstructed is delayed for a long time. In addition, because the amount of data itself is large, the probability of an error impossible to restore within a table I is very high. Also, in the case where that error occurs that makes reconstruction impossible ( or decoding) of a frame I, the reconstruction processing is stopped until the subsequent frame I is reached if a dedicated element is not provided to solve the problem. For example, in the case where the data error occurs due to a cause, its initially small adverse influence can be amplified over all the tables; in the worst case, reconstruction processing stops. Additionally, with the conventional technique to insert tables I at a fixed interval, it is necessary to search, when the reconstruction processing starts in the frame in a certain temporary position, the nearest frame I by means of some element, to reconstruct the corresponding image from of it, and visually display the reconstructed image after the frame is reached in the temporary objective position; however, this search for the I-frame takes a long time. If a certain dedicated element is provided to solve the problems mentioned above, the corresponding load to be imposed in the decoding process naturally increases. In addition, because a heavy load will be imposed on the process to reconstruct the I frames, an additional function that can process the I frames for the process is required. The present invention is provided by considering the above problems, and its first objective is to provide a method and system for compressing image information in motion, which can compress with a high proportion of compression and at high speed, data or image information that can be compressed. subject to prediction coding, and which can improve the quality of the image.

The second objective of the present invention is to provide a method and system for compressing image information in motion, preserving the detail of the original image and without deterioration of the quality of the image even if the compression ratio is improved by enlarging the image. block size. In addition, the third objective is to provide a method and system for compressing image information in motion, which can visually display the reconstructed image in an arbitrary temporal position in a simple manner by means of preventing an initially adverse influence due to an error occurrence. of data during a reconstruction process prevails over all the tables and, in turn, that the reconstruction process stops, without it taking much time to look for the closest square I by some element when the reconstruction process begins in the frame in an arbitrary temporal position and then the corresponding image is reconstructed from it.

DESCRIPTION OF THE INVENTION In accordance with a method of compressing moving image information of a first embodiment of the present invention, which compares adjacent pixels spatially within a frame with each other or compares adjacent pixels in a manner between the tables, outputs the information of the resulting difference for the pixels, stores in a bitmap the information regarding whether the information of the difference that came out is greater than a given parameter (threshold), and compresses the information stored in the bitmap that is greater than the parameter (threshold), thereby reducing the redundant information; by dividing an image within a block frame and by approximating (substituting) each block divided by a plane that is represented by the three components for the pixels within each block, the problems mentioned above are solved. On the other hand, a system for compressing moving image information, according to the first embodiment of the present invention, comprises: information of the bitmap that re-encodes the element to compare the adjacent pixels spatially within a frame one with the other or that compare the adjacent pixels temporarily between the frames that output the resulting difference information for the pixels, and which stores in a bitmap the information with respect to whether the output difference information it is greater or not than a given parameter (threshold), and the information compression element to compress the information that is stored by the element to re-encode the bitmap information that is greater than the parameter (threshold), reducing redundant information by the same; the system further comprises a block approach element for dividing an image into a block frame, before an inter-frame compression is performed, and which approximates (replaces) each block divided by a plane that is represented by the three components for the pixels within each block, solving with the same the problems mentioned above. According to a moving image compression method of the second embodiment of the present invention, intra-frame compression is performed by compressing the entire image in a block unit of pixels of nxm (n and ra are integers, respectively ) using a method of information compression, comparing the pixels between the original image and the expanded image after compressed output to the information of the difference resulting from each pixel, and if there is a pixel that has caused a greater difference to occur than a given parameter (threshold), repeatedly using a smaller block size for a portion or performing a surrounding area that includes this pixel until a designated minimum block size is reached, solving with the same the problems that were mentioned previously. Additionally, in accordance with a moving image compression system of the second embodiment of the present invention, the block approach element performs an intra-frame compression by compressing the entire image into a pixel block unit of nxm (n and m are integers, respectively) using an intra-frame compression method, which compares the pixels between the original image and the expanded image after tablets, which outputs the information of the resulting difference of each pixel, and if it exists a pixel that has caused a difference greater than a given parameter (threshold) to occur, repeatedly using a smaller block size for a portion or a surrounding area that includes this pixel until a designated minimum block size is reached, also solving with the same the problems mentioned above. In accordance with a moving image compression method of the third embodiment of the present invention, an intra-frame coded frame I is used (i.e., a reference frame exclusively from which an image can be reconstructed); Table I is divided spatially into blocks I; and no block I is inserted into any block within the table that has been updated due to the difference between the frames that are larger than a given parameter (threshold) within a specific period of time when the I blocks are scattered between each picture along the time axis, also solving by the same the problems mentioned above. Additionally, a moving image compression system of the third embodiment of the present invention comprises the insertion element of the block I, which uses an intra-frame coded frame I (i.e., a frame of reference exclusively from the which one can reconstruct an image), which spatially divides the I frame into blocks I, and which scatters the I blocks between each frame along the time axis; the insertion element of the block I does not insert any block I into a block within the frame that has been updated because the difference between the frames is greater than a given parameter (threshold) within a specific period of time, also solving by the same as the problems mentioned above. In accordance with the method of compressing moving image information and the system thereof, because the block conversion procedure is omitted, compression can be performed with a high compression rate and at a high speed of the data or the information of the image that can be submitted to the prediction coding, improving by the same the quality of the image. With the conventional technique, in particular, when the difference between the frames is large, the quality of the image deteriorates drastically; however, according to the present invention, the deterioration of that image quality can be reduced. More specifically, in accordance with the first embodiment of the present invention, it is possible to provide the linear change in the quality of the image without drastically deteriorating the image quality due to a threshold for a block. In accordance with the above, the adjustment of the bit rate of the communication without the deterioration of the quality of the image can easily be made and, in addition, the improvement can be carried out by approximately -20 percent to -50 percent of the compression ratio, maintaining the quality of the image as it was. In addition, the Huffman adaptation coding and the arithmetic matching coding collectively perform the prediction coding procedure which includes the generation of conventional difference information and the Huffman coding and / or the generation of difference information and the arithmetic coding, generating efficient key words and efficiently compressing the data, so that the information of the image can be submitted to the prediction coding. In addition, the reduced (compressed) data according to the first embodiment of the present invention is used to define a plane and, when expanded, represents a plane with a gradació. According to a method and system of compressing moving image information of the second embodiment of the invention, even in cases where the compression ratio is improved using a larger block size, the detail of the original image, and in compliance, the deterioration of image quality can be reduced. Even in cases of an original image that is configured by intensely deferential intensities of fine lines with a fixed colored background, it is possible to avoid fine lines being completely lost. In accordance with a method of compressing moving image information and the system thereof of the third embodiment of the present invention, when spatially divides the I frame into blocks, and when the divided I blocks are scattered between each table along the time axis, because no block I is inserted into any block within the table that has been updated because the difference between the tables is larger than the parameter (threshold), it is possible for the image reconstruction processing to begin reconstruction due to a previously determined number of previous frames from which an image can be reconstructed completely, and to visually display a reconstructed image after which the table in the target temporary position has been reached; in this way, without a time-consuming search for a frame I, a reconstructed image in an arbitrary position can be visually displayed in a simple manner. In addition, because the amount of data distributed in a communication server and / or the data transmission path is temporarily uniformed during the transmission of the moving image, a higher transmission performance can be obtained for the transmission of the content that one with the conventional technique. In addition, because on the receiving and reconstruction side the change in the amount received per unit time is small, a necessary amount of memory can be reduced to buffer, the expected reconstruction loads are regulated, and even a system with low capacity can be reconstructed in a stable manner. In addition, because the possible influence of data errors on reconstruction is small. It is possible to continue reconstructing with the data error unattended, therefore it is not necessary for the lateral system to send the data again, and reducing the load on the side of the transmission. In addition, it is also possible to provide in a simple manner the ability to distribute multiple distributions, etc., for the transmission of the moving image.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram showing the outline of a structure for compressing image information in motion. Figure 2 is a block diagram detailing the structure for compressing the moving image information of Figure 1. Figure 3 illustrates an example of a specific structure for coding. Figure 4 is an exemplary block diagram showing an example of a specific structure for decoding. Figure 5 is an explanatory figure showing a plane that is represented by three pieces of data: intensity Z of a pixel within a block, the gradient of the block in the X direction, and the gradient of the same block in the Y direction, which are used to approximate the corresponding divided image block. Figure 6 is a plan view of an image that explains the operation of using a smaller block size for a portion or a surrounding region of a pixel, which causes a difference greater than a given parameter (threshold) to occur. Figures 7A and 7B show an image explaining an inter-frame compression procedure, wherein Figure 7? is a planar view of the square t and Figure 7B is a planar view of the square t + 1. Figure 8 is a plan view showing a plurality of blocks I forming a square I. Figure 9 is an explanatory figure showing a state where blocks I are inserted between the frames. Figure 10 is a flow chart showing the step of performing an intra-frame compression. Figure 11 is a flow chart showing the step of performing an inter-frame compression; and Figure 12 is a flow chart showing the step of performing the insertion of block I.

DESCRIPTION OF THE REFERENCE NUMBERS P ... parameter (threshold) 1 ... analog-to-digital converter 2 ... buffer 3 ... encoder / compression unit 4 ... bitmap information recording element 5 ... information compression element 6 ... entropy coding element 7 ... block insertion element I 8 ... block generation element I 12 ... comparison element 13 ... data of current frame 14 ... previous frame data BEST MODE FOR CARRYING OUT THE INVENTION To begin with, a method for compressing image information according to a first embodiment of the present invention is described. The present invention includes a method of compressing moving images to compare adjacent pixels spatially within a square one with gold or to compare adjacent pixels temporarily between the frames that output the information of the resulting difference for the pixels. pixels, store in a bitmap the information regarding whether the information of the difference that came out is greater or not than a given parameter (threshold), and compress the information that is stored in that bitmap that is larger than the parameter (threshold), reducing redundant information by the same; where an image within a frame is divided into blocks, and each divided block is approximated (replaced) with a single plane that is represented by three components for the pixels within the block, before inter-frame compression processing begins .

In addition, the information that is stored in the bitmap no larger than the parameter (threshold) P is processed (deleted) as a pixel without changes. Additionally, in accordance with a block approach method to configure a single plane that is represented by three components for the pixels, the average and the least squares method are used. In addition, in accordance with the intra-frame compression procedure, the plane is represented by three pieces of data: the intensity of a pixel within a block, the intensity gradient inside the block in the X direction, and the intensity gradient. within the block in the Y direction. The information that is stored in the bitmap is compressed by at least one binary image coding method that is selected from the group consisting of length execution coding, modified READ coding (MR, MMR), modified Huffman (MH) coding, and JBIG coding. The information larger than a parameter (threshold) P is compressed using the adaptive Huffman coding, which uses as many Huffman tables as the amount of information expected. Redundant information between frames is further reduced using entropy coding.

Entropy coding is done either through the Huffman coding procedure, which encodes using a table that is selected from many Huffman tables as an expected amount of information, or the adaptive arithmetic coding procedure, which encode using a table that is selected from many arithmetic tables as an expected amount of information. This decoding procedure is performed based on the information of the difference between the pixels. The difference information is the difference output through the comparison of the pixel t and the pixel t - 1 between the frames. In addition, the information output of the difference is used through the comparison of a block of pixels of n x m (where n and m are integers of 2 or more) and the corresponding block thereof between the frames. Additionally, the output of the difference information is used through the comparison of the pixel t and the pixel 7 - 1 between the frames, where a block is configured by the pixels of nxm (n and m are integers of 2 or more) inside of a painting. Also, with the pixels of n x m between the squares, n denotes 2K (K is a complete number), and m denotes 2? ' (? 'is a complete number). Additionally, intra-frame compression can be performed while changing the size of the blocks that are divided within the same frame, before the inter-frame compression procedure is started. Next, a system for compressing moving image information according to a first embodiment of the present invention is described. The present invention includes a system for compressing moving image information comprising: bitmap information recording element 4 for comparing adjacent pixels spatially within a frame or for comparing adjacent pixels temporarily between frames. frames, outputting the information of the difference resulting from the pixels, and storing in a bitmap the information with respect to whether the information of the difference that came out is greater than a given parameter (threshold), and compression element of information 5 for compressing the difference information that is stored by the re-coding element of the bitmap information 5 which is larger than the (threshold) parameter P, thereby reducing the redundant information; the system further comprises a block approach element for dividing an image into a block frame, and approximating (replacing) each block divided by a single plane which is represented by the three components for the pixels within the block. Additionally, the information compression element 5 processes (deletes) the information that is stored by the information register element 4 of the bitmap no greater than the parameter (threshold) P as an image without changes. In addition, the block approach element approximates a single plane that is represented by the three components for the pixels, using the average and the minimum square method. In addition, with the block approach element, the plane is represented by three pieces of data: the intensity of a pixel within a block, the gradient of the block intensities in the X direction, and the gradient of block intensities in the Y direction. In addition, the information that is stored by the information register element 4 of the bitmap is compressed by at least one binary image coding method that is selected from the group consisting of length execution coding. , modified READ coding (MR, MMR), modified Huffman (MH) coding, and JBIG coding. Additionally, the information compression element 5 for compressing information that is larger than the (threshold) parameter P, performs an adaptive Huffman encoding that uses as many Huffman tables as a quantity of prediction information. The entropy encoding element 6 is also provided, which reduces the redundant information between the frames; this entropy-encoding element 6 performs either adaptive Huffman coding, which it encodes using a table that is selected from as many Huffman tables as an expected quantity of, information, or adaptive arithmetic coding, which it encodes using a table that is selected from as many arithmetic tables as an expected amount of information. Additionally, the information of the difference that is stored in the information register element 4 of the bitmap is an output of the difference through the comparison of the pixel t and the pixel t-1 between the frames, wherein a block is set by the pixels of n X m (n and m are integers of 2 or more) within a frame. In the following, an embodiment of a method and system for compressing moving image information is described, in accordance with the first aspect of the present invention. Figure 1 is a block diagram showing the outline of a structure for compressing image information in motion. An analog signal output composed from a device such as a video camera, a disc player, or a video cartridge player in accordance with the NTSC standard, is converted to a digital signal to represent a single line of a video frame through an analog-to-digital converter 1, and output digitally to the buffer 2, where it is stored in turn. Note that, although it is described that the output of the analog signal from the NTSC device is converted to a digital signal by the analog-to-digital converter 1 and the resulting digital signal is output and stored in the buffer 2, The present invention is not limited to this structure. In other words, in accordance with the present invention, any video signal that includes the output of common video signals from any of the different types of devices can be efficiently compressed. As shown in Figure 1, the information storage circuit 4 of the bitmap, which sequentially compares the pixel and the pixel t-1 between the frames, and stores in a bitmap that is based on a only bit, the information is provided as to whether or not the resulting difference is greater than the parameter (threshold) P. The comparison of this pixel t and the pixel t-1 is made based on a pixel component (i.e. or dye). This is to temporarily compare a pixel (pixel t) in the current frame and the corresponding pixel (pixel t - 1) in the previous frame, where t denotes time. In accordance with the foregoing, the difference between the pixels tyt -1 that is stored by the information register element 4 of the bitmap is compressed, which is greater than the parameter (threshold) P, but others are determined for be of pixels without changes and then they are processed (deleted). That information (difference) greater than the parameter (threshold) P is compressed by the information compression element 5 using the adaptive Huffman coding, which uses as many Huffman tables as the expected amount of information. The entropy encoding element 6 is provided, which performs the comparison of adjacent pixels spatially or temporally, outputs the resulting difference information, and then performs the adaptive arithmetic coding using an arithmetic table, which is selects from as many arithmetic tables as, for example, an expected amount of information that is based on prediction information so that redundant information can be reduced between tables. After the encoding is performed by the compression encoder 3, a data block is then transmitted within each frame to a memory of the information register element 4 of the bitmap, as shown in Figure 2. Then the data of the current frame 13 and the data of the previous frame 14 are stored, which are erased by the time of a single frame. After this, the comparator 12 determines the data of the current frame 13 and the data of the previous frame 12 in terms of the redundancy between the frames and calculates the difference thereof. That is, each coded block is compared to the corresponding block in the previous frame. Each block is marked with a single bit that identifies whether each block is changed or not from the corresponding previous block. Through this procedure, a bitmap of the frame is generated with a single bit per block. In the present, the bitmap for each frame is distinguished from another bitmap by making the comparison between the frames. This modality uses a basic technique, the technique of intra-frame compression in which the size of the blocks is not changed. As shown in Figure 5, an image within a frame is previously divided into blocks, and each previously divided block is approximated (replaced) with a single plane that is represented by three pieces of data: the intensity Z of a pixel in each block, the gradient of the intensities within each block in the X direction, and the gradient of the intensities in each block in the Y direction. More specifically, with the intra-frame compression procedure, a image in a plurality of blocks, and each block is replaced with a single plane approaching each block. This plane can be represented by three components for the pixels in each block, such as the intensity z, the gradient x of the intensities in the X direction, and the gradient and the intensities in the Y direction. Alternatively, the plane it can be represented by the intensity z of a pixel inside a block, the gradient of the intensities of the pixels between the blocks in the X direction, and the gradient of the intensities of the pixels between the blocks in the Y direction. For example, the average and the minimum square method can be used for the approach. The resulting diminished data (tablets) represent a plane; by expanding that data, you get the plane with the gradation. When a block is configured by s pixels, an expected compression ratio within a single frame is 3 / s; the compression ratio increases while s increases, however, the quality of the image deteriorates: Note that the size and shape of a block is of pixels of n x m, where n and m are any integer. Also, with the pixels of n x m between the squares, n can denote 2K (K is a complete number), and m can denote 2? ' (K 'is a complete number). Next, the basic technique that is used for the compression of inter-frames in accordance with this modality is described. More specifically, in accordance with the first method for inter-frame compression, the block in table t-1 that is placed in the same location as that in table t, is compressed by intra-frames, and z ( t + 1), x (t + 1), ey (t + 1) are obtained in terms of three components: denoting z the intensity of a pixel, x denoting the gradient of the intensities in a block in the x direction, ey denoting the gradient of the intensities in the block in the direction y. The sum-squared-average error is calculated between a group of z (t), x (t), ey (t) and a group of z (t + 1) rx (t + 1), ey (t + 1 ), and then it is compared to the threshold P. As a result, if it exceeds the threshold P, the determination is made such as "THERE IS DIFFERENCE". Alternatively, each of a group of z (t), x (t), ey (t) and a group of z (t + 1), x (t + 1), ey (t + 1) is compared with a group of thresholds Pz, Px, and Py; if the resulting difference exceeds the threshold P, the ATHERE IS DIFFERENCE 'determination is made. If the ATHERE IS DIFFERENCE 'determination is made, the portion in the bitmap corresponding to the block in the box is marked. In the previous case, a single-bit map is used while three-bit maps are used later. The bitmap comprises a configuration that includes O and / or 1 (ie, binary data), and is compressed using, for example, length execution encoding. In addition, the pieces of the difference data Az (t) = z (t + 1) - z (t), Ax (t) = x (t + 1) - x (t), and Ay (are compressed by entropy. t) = y (t + 1) - y (t). It is noted that, according to the first method, because the expansion is not performed, the load that is imposed in the calculation is slight, however, calculation errors can accumulate. In accordance with the second method for inter-frame compression, the data that is compressed is expanded using a basic technique for the inter-frame compression mentioned above, and the pieces of the pixel data that make up a pixel are reconstructed. block. It is calculated and compared with the threshold P a sum-squared-error error between the respective parts of the data of the pixel that is located each in the same place within the same block in the following table t + 1 and the reconstructed pieces corresponding to the pixel data. As a result, if it exceeds the threshold P, the determination of "THERE IS DIFFERENCE" is made, if the determination VTHERE IS DIFFERENCE 'is made, the portion in the bitmap corresponding to the block in the table is marked. of bits comprises a configuration that includes 0 and / or 1 8 that is, binary data), and is compressed using, for example, length execution coding In addition, the pieces of difference data Az (t) are compressed by entropy = z (t + 1) - z (t), Ax (t) = x (t + 1) - x (t), and Ay (t) = y (t + 1) - y (t). that, according to the second method, because the expansion is carried out, the load imposed in the calculation is heavy, however, the calculation errors do not accumulate In accordance with the third method for the compression of inter-tables , an error of sum-squared-error between the respective parts of the data of the pixel within a block in the current t-box and the pieces corr is calculated and compared with the threshold P of the pixel data which is each located in the corresponding identical place within the corresponding identical block in following table t + 1. As a resultif it exceeds the threshold P, the XTHERE IS DIFFERENCE 'determination is made. If the determination "THERE IS DIFFERENCE" is made, the difference is calculated from the corresponding pixel data which is located in the corresponding identical place within the corresponding identical block in the following table t + 1 and is compressed by interest The portion in the bitmap corresponding to the block in the frame is marked.This bitmap comprises a configuration including 0 and / or 1 (ie, binary data), and is compressed using, for example, the length execution coding.The difference data are compressed by entropy.It is noted that, according to the third method, because the compression is made after the determination of the difference is made, the amount of the calculation is the minimum and the calculation errors can not be accumulated.The entropy encoding element 6, as shown in Figure 1, compresses the bitmap information that is based on a single bit that is stored m using the information re-encoding element 4 of the bitmap, using a binary image coding such as length execution coding, modified READ coding (MR, MR), modified Huffman coding (MH), or JBIG coding. More specifically, in the case of a binary document image that is manipulated by a facsimile machine or the like, there is a high probability that white pixels or black pixels successively appear in one or more continuous areas; in accordance with the execution length encoding method, a one-dimensional segment is used that includes only white or black pixels, which are called execution, is used as a unit to encode, and using the number of identical continuous pixels that they are included within each execution as the length of each execution, the coding is performed. For example, in digital facsimiles using the public telephony network, modified Huffman codes are generally used for the length execution model that is being established separately for the black and white pixels. Modified Huffman (MH) coding is used as a one-dimensional coding method in the facsimile transmission of monochromatic pixel information that includes 1728 pixels per scan line, which is obtained by scanning at a density of pixel of, for example, 8 pixels / mm, where the MH codes represent the lengths of execution that are of the length of each segment that includes only continuous white pixels (white execution) or only continuous black pixels (black execution), and Variable length codes are assigned to the respective runs using the statistical tendency that white or black runs having particular lengths occur more frequently than other lengths, which is the theory for reducing the amount of data. Modified READ coding (MR or MMR) is used as a standard method for two-dimensional coding, as well as one-dimensional coding, where the MMR coding is the one that adjusts infinitely for both standard and high resolution resolution for MR coding. The basic structure of the method of compressing moving image information and the system thereof, in accordance with the present invention, is to compare the adjacent pixels spatially or temporally and the output of the information of the resulting difference, so that redundant information can be reduced between tables. More specifically, the pixel t and pixel t-1 are compared sequentially between the frames, and the information with respect to whether or not the resulting difference is larger than the (threshold) parameter P is stored as a piece of a single bit of the bitmap information. The entropy encoding element 6 predicts the codes that may occur within each frame and between the frames, and outputs a small error from the predicted value that the average code length per pixel is never less than, or equal to, to the average information content (ie, entropy) when the code assignment is made and the resulting code sequence is transited. The algorithm of the adaptive Huffman coding is described below. Adaptive Huffman coding is performed, in order to efficiently generate the keywords by collectively performing a series of prediction coding procedures that includes the generation of difference information and the Huffman coding thereof. In accordance with conventional Huffman coding, the key words are usually generated using a Huffman table, and the Huffman table that was generated is updated whenever each individual word is coded, or the Huffman coding procedure is performed in a manner dynamic. In contrast, in accordance with the adaptive Huffman coding, which uses as many Huffman tables (code table) as the predicted amount of information, one of those many tables is selected by a table selector in accordance with the predicted information; with that, the coding of the selected table is performed. In accordance with the above, the data such as the audio information, which can be subjected to prediction coding, it is compressed effectively. ? The algorithm of adaptive arithmetic coding is described below. Adaptive arithmetic coding is carried out, in order to efficiently generate the key words by means of collectively performing a series of prediction coding procedures, which includes generating the difference information and the arithmetic coding thereof. In accordance with conventional arithmetic coding, keywords are usually generated using a single occurrence probability table, and the occurrence probability table that is generated is updated whenever each individual word is encoded, or the arithmetic coding procedure is performs dynamically. In contrast, in accordance with adaptive arithmetic coding, using as many arithmetic tables (code table) as the predicted amount of information, one of those many tables is selected by a table selector in accordance with the predicted information.; with that, the coding of the selected table is performed. According to the above, the data such as the image information, which can be subjected to the prediction coding, is effectively compressed. In Figure 3 a specific structure of a prediction coding circuit is shown, in which in order to encode the input image data, which is converted from analog to digital, it is conveniently delayed and coupled to the selector of table. The data of the input image is also transmitted without any delay to the coding unit, which then encodes it. The resulting pieces of the coded data are compared and then their difference is calculated. The table selector selects a code table for the input image data in accordance with the prediction information, which transmits it to the coding unit, which in turn, compresses the data of the input image so that you can get the adjusted keywords. Figure 4 shows a specific structure of a prediction decoding circuit, in which a keyword is transmitted to the decoder, and at the same time the keyword that was transmitted directly is temporarily sent to the table selector, which selects then a decoding table in accordance with the prediction information that sends it back to the decoder, which, in turn, calculates the difference from the pixel value that was previously decoded so that the adjusted keywords can be obtained . A method for compressing the moving image information according to a second embodiment of the present invention is further described. The present invention is a method for compressing moving image information by comparing the adjacent pixels spatially within a frame or comparing the adjacent pixels temporarily between the frames, outputting the difference information resulting from the frames. pixel values, store them in a bitmap information regarding whether the information of the difference that came out is greater or not than a given parameter (threshold) P, and compress the difference information that is greater than the parameter (threshold ) P in accordance with the information that is stored in the bitmap, thereby reducing the redundant information, where the intra-frame compression is performed while the block size is changed divided within the same frame, before inter-frame compression is performed. In accordance with the intra-frame compression procedure, the pixels within each block are compared, outputting the difference information of the pixels while the size of the divided block is changed, and if that information of the difference is greater than the parameter (threshold) P, a smaller block size is used for the portion that includes this difference information. Also, if the information of the difference between the pixels is larger than the (threshold) parameter P, a smaller block size is repeatedly used. An image within a frame is divided into blocks, and each block approaches (replaces) a single plane that is represented by at least three components for the pixels within each block. Furthermore, in accordance with the intra-frame compression procedure, the plane is represented by three pieces of data: the intensity of a pixel within a block, the gradient of the intensities within the block in the X direction, and the gradient of the intensities within the block in the Y direction. In accordance with the intra-frame compression procedure, the whole image is compressed in a block unit of nxm pixels (n and m are integers, respectively), using a compression method of intra-frames, the pixels are compared between the original image and the image that was expanded after they were compressed and each information is output of the resulting difference for the pixels, and if there is a pixel that has caused a difference greater than the parameter (threshold) P to occur, the operation of using a smaller block size or a surrounding area that includes this pixel is performed repeatedly until a size is reached of designated minimum block. In addition, if there is no change in block size during the intra-frame compression procedure, the inter-frame compression procedure is started. If you change the size of the block to a larger one, no additional calculation is made for the difference in the data in that block and it is output as it is. If the block size is changed to a smaller one, the difference is calculated from the previous expanded data within each portion, and compressed into that smaller block size. Next, a system for compressing moving image information according to a second embodiment of the present invention is described. In accordance with the intra-frame compression procedure, the whole image is compressed in a block unit of nxm pixels (n and m are integers, respectively) using an intra-frame compression method, the pixels are compared between the image original the compressed image after it compressed, outputting each resulting difference information for the pixels, and if there is a pixel that causes a difference larger than the parameter (threshold) P to occur, the operation of using a block size even smaller for a portion or a surrounding area that includes this pixel is performed repeatedly, until a designated minimum block size is reached. In addition, with the block approach element, if there is no change in block size during the intra-frame compression procedure, the inter-frame compression procedure is initiated. Also, with the block approach element, if you change the size of the block to a larger one, no additional calculation is made for the data difference in that block and it is output as is. Furthermore, with the block approach element, if the size of the block was changed to a smaller one, the difference is calculated from the previous expanded data within each portion, and compressed into that smaller block size. In the following, a method and system for compressing moving image information according to the second embodiment of the present invention are described, while referring to Figures 6, 7 and 10. As described above, if used a larger block size, the compression ratio can be improved; however, the detail of the original image may be lost, which deteriorates the quality of the image. A phenomenon occurs where the fine lines are completely lost when the original image is configured by those fine lines, the intensity of which differs from that of a fixed colored background. The following method is used to solve these problems. To simplify the explanation, the case (an example) of 16 x 16 pixels of a white image is described. More specifically, as shown in the Figures 6 and 10, the entire image is compressed (expanded) into a 16 x 16 pixel block unit using an intra-frame compression method as described above (STEP 1 in Figure 10). The pixels between the original image and the image that was expanded after being compressed are compared by outputting the information of the difference resulting from each pixel, which is then compared to the parameter (threshold) P (STEP 2 in Figure 10) ). As a result of this comparison, if there is a pixel with the difference exceeding the (threshold) parameter Pl, a portion or a block of 8 x 8 pixels including this pixel is compressed (expanded) (see STEP 3 in the Figure) 10 and the largest circle in Figure 6). In addition, the surrounding area of that portion is compressed into an 8 x 8 pixel block unit. After this, the pixels are compared between the original image and the image that was expanded after compressing it, outputting the information from the resulting difference of each pixel, which is then compared to the parameter (threshold) P2 (STEP 4 in Figure 10). As a result of this comparison, if a pixel exists with the difference exceeding the (threshold) parameter P2, a portion or block of 4 x 4 pixels including this pixel is compressed (expanded) (see STEP 5 in the Figure) 10 and the intermediate circle in Figure 6). In addition, the surrounding area of that portion is compressed into a block unit of 4 x 4 pixels. The pixels are compared between the original image and the image that was expanded after compressing it, outputting the difference information resulting from each pixel, which is then compared to the parameter (threshold) P3 (STEP 6 in Figure 10). As a result of this comparison, if a pixel with difference exceeding the parameter (threshold) P3 exists, a portion or block of 2 x 2 pixels including this pixel is compressed (expanded) (see STEP 7 in Figure 10) and the smaller circle in Figure 6). In addition, the surrounding area of that portion is compressed into a 2 × 2 pixel block unit. The procedure continues to the inter-frame compression step (STEP 8 in Figure 10). In this way, it is possible to compress the original image while maintaining its detail while maintaining a high compression ratio. Then, as a result of the intra-frame compression as mentioned above, an inter-frame compression method is described in the case where a compressed image of the frame t in Figure 7 (?) And an image is obtained. compressed from table t + 1 in Figure 7 (B), while reference is made to Figure 11. Block sizes are compared (STEP 9); because there is no change in the size of the block size between 1 in Figure 7 (A) and 1 'in Figure 7 (B), and between 2 in Figure 7 (A) and 2' in Figure 7 (B), the difference and the compressed inter-frame are calculated using one of the methods described in the inter-frame compression procedure mentioned above (STEP 10). After this, it is determined whether the difference in each block size is equal or not to the parameter (threshold) P or greater (in STEP 15). If it is determined that the difference in each block size is equal to the parameter (threshold) P or greater, the effect of THERE IS DIFFERENCE is recorded in a bitmap that outputs the difference (in STEP 16). Otherwise, if it is determined that the difference in each block size is smaller than the parameter (threshold) P, the bitmap is stored and updated with the effect of THERE IS NO DIFFERENCE, to the difference (see STEP 17). Incidentally, in the case where there is change in block size, particularly where there is a change towards a coarser resolution (STEP 11) such as the change between 4 and 4 'in Figure 7, 4 is used. 'as a key block (or a keyframe) that can extend itself independently of the previous table. In this case, no difference is calculated. In other words, the difference in the data in the r-block is not calculated and output is given as is (STEP 12). In the case where there is a change in the size of the block, in particular where there is a change towards a finer resolution (STEP 13) such as the change between 3 in Figure 7A and 3 'in Figure 7B, it is calculated the difference from the extended data within block 3 for each portion and is compressed in terms of the block size unit (in STEP 14). Next, a method for compressing the moving image information according to the third embodiment of the present invention is described. In accordance with the present invention, intra-frame encoded I frames (ie base frames only from which the corresponding image can be reconstructed) are used.; each frame I is previously divided spatially into a plurality of blocks I, which are then scattered between each frame along the time axis. When the respective I blocks, which are spatially divided, are scattered between each frame along the time axis, no block I is inserted into any block within the frame that is updated at a time when a state occurs in that the difference between the tables is greater than the parameter (threshold) P. Additionally, a compression method is provided where: an image within a table is previously divided into blocks, all the divided blocks are approximated (replaced) each with a single plane that is represented by three pieces of data: the intensity of a pixel in each block, the gradient of each block in the V direction, and the gradient of each block in the Y direction; and using an intra-frame encoded I frame (ie, a reference frame only from which an image can be reconstructed), frame I is inserted into a sequence of frames, where frame I is previously divided spatially in blocks; when the divided I blocks are scattered between each frame along the time axis, no block I is inserted into any block within the frame that has been updated because the difference between the frames is greater than the parameter (threshold) P within a specific period of time. A method for compressing moving image information is provided, wherein the adjacent pixels are compared spatially within a frame or the pixels are compared between adjacent frames temporarily to output the information of the resulting difference between the pixels; the information regarding whether the information of the difference that came out is greater or not that a given parameter (threshold) is stored in a bitmap; and the information of the difference that is stored in the bitmap that is greater than the parameter (threshold) P is compressed, thereby reducing the redundant information; wherein: the use of an intra-frame coded frame I (ie, a reference frame only from which an image can be reconstructed), the frame I is previously spatially divided into blocks; and when the divided I blocks are scattered between each frame along the time axis, no block I is inserted into any block within the frame that has been updated because the difference between the frames is greater than the parameter (threshold) P within a specific period of time. In addition, a method for compressing moving image information according to the third embodiment of the present invention is described. In accordance with the present invention, intra-frame encoded I frames (i.e., base frames only from which the corresponding image can be reconstructed) are used; each frame I is previously divided spatially into a plurality of blocks I, which are then scattered between each frame along the time axis. The block insertion element I 7 does not insert any block I into a block within the frame that has been updated, because the difference between the frames is greater than the parameter (threshold) P within a specific period of time. Additionally, a compression system is provided which comprises a block approach element for dividing an image into a block frame and which approximates (replaces) each of all the blocks divided by a single plane which is represented by three pieces of data: the intensity of a pixel in each block, the gradient of each block in the X direction, and the gradient of each block in the Y direction; the system further comprises: block generating element I 8 for spatially dividing an intra-frame encoded frame I (ie, a reference frame only from which an image can be reconstructed) in block I; and block insertion element I 7 to insert a block I into a portion, except for a block within the frame that has been updated because the difference between the frames is greater than the parameter (threshold) P within a period of specific time, when scattered I blocks divided between each frame along the time axis. In the following, a method and system for compressing moving image information according to the third embodiment of the present invention are described, while reference is made to Figures 8, 9 and 12. The present invention is a coding method , which corresponds to the reconstruction (decoding) of partial sequence image and / or the data error that occurs during the reconstruction of the image. It is noted that there is a premise that a compression algorithm is used without using any movement prediction and correction technique for more than three complete frames to be compressed. First, as shown in Figure 12, a coded table of intra-frame prediction or a frame I in a single or multiple blocks I (in STEP 1) is spatially divided, and these I blocks that were divided are scatter along the time axis (generation of blocks I in STEP 2). It is noted that the size of the block, the shape of the divided block, etc., due to that generation of blocks I, can be changed optionally, and in addition they can be selected in a random manner. More specifically, as shown in Figure 8, a pixel array of 8 x 8 is divided spatially, each having 2 x 2 pixels, and these are inserted into a frame sequence at period intervals. fixed. As a result, when the portion where the output of the inter-frame difference occurs (ie, the portion with large information content, where there is a movement) and a block I (whose information content is greater) overlaps that one of the other tables), a useless block I must be inserted, drastically increasing the information content, which can cause errors that are impossible to recover inside the inserted box I to happen. In order to avoid this, as shown in Figure 12, in cases where there is no problem with the processing speed on the coding side, whether or not the inter-frame difference state is greater than the parameter (threshold) P within a designated period of time and it is determined that the update (difference output) is done in accordance (in STEP 3), and no I block is inserted into any block that has been updated (or that you leave a difference) (in STEP 4). On the other hand, a block I is inserted into the block that was not updated (or gives a difference) (in STEP 5). With reference to Figure 9, a specific coding method is described. It is noted here that a frame I of pixels 8 x 8 is spatially divided into a block of pixels of 1 x 2 by the block generating element I 8 which configures thirty-two I blocks in total, as an example. It is also noted here that a moving image is provided as an example comprising an image frame with a block of pixels of 8 x 8 and with the maximum block of pixels of 16 x 16. In Figure 9, as a matter of convenience, the (n + H) a or (n + 32) avo is omitted. First, I blocks are inserted which each have horizontally 1 x 2 pixels, which are marked in black in the Figure. An object (dark gray regions that represent a difference output block that moves against the background), which is initially placed in the upper left corner of the image and which requires the maximum of 2 pixels to be updated. x 2 (that is, to output the difference), moves to the lower left. Up to the frame (n + 3) °, a block I is usually inserted (in STEP 5 of Figure 12). By contrast, because the block corresponding to the object appeared in the upper right in the table (n + 3) ° is updated (that is, it outputs the difference), the block I that is to be inserted into the table (n + 4) ° is not actually inserted (see the shaded portion and STEP 4 in Figure 12). It is noted that the light gray portion denotes an output block of the difference that is returning to a portion of the original background, because the object has been moved. In this case, the processing in which no block I is inserted (see STEP 4 in Figure 12) appears in the table (n + 7) ° and in the table (n + 8) °. More specifically, as a result of the movement of the object in the frame (n + 5) °, if a difference output block (the light gray portion) that should be back to the bottom portion exists as a portion that will be updated, within the frame (n + 7) °, only the single block on the right side of the blocks I that each have 1 x 2 pixels is not inserted in that portion. As a result of the movement of the object in the frames (n + 4) ° and (n + 5) °, if a block of difference output (the light gray portion) of 1 x 2 pixels placed horizontally it should be back in the bottom portion exists as a portion to be updated, inside the box (n + 8) °, a block I of pixels of 1 x 2 is not inserted in that portion. In this case, the reference time (the near past) during which no I block should be inserted, is represented by the number of frames needed to insert blocks I in each position of the block (8 x 8 / 2 = 32 frames). In other words, no block I is inserted in any block that is updated (it outputs the difference) within a sequence of thirty-two frames due to the movement of an object, and so on. In order to initiate the reconstruction of a desired frame in an arbitrary temporal position, the decoding should initiate a predetermined number of frames in advance to allow the complete reconstruction of a single image.

INDUSTRIAL APPLICABILITY As described above, with a method of compressing moving image information and the system thereof, in accordance with the first embodiment of the present invention, an intra-frame image is previously divided into blocks, each divided block it approaches with a single plane that is represented by three pieces of data: the intensity of a pixel within each block, the gradient of each of those blocks in the X direction, and the gradient of each of those blocks in the direction And, performing the same with efficient intra-table compression. In accordance with a method of compressing moving image information and the system thereof, intra-frame compression is performed by compressing the entire image into a block unit of nxm pixels (n and m are integers, respectively) , the pixels are bought between the original image and the image that was expanded after being compressed, leaving the information of the difference resulting from each pixel, and if there is a pixel that caused a difference greater than the parameter to occur (threshold) P, a larger block size is used repeatedly for a portion or a surrounding area that includes this pixel, until the designated minimum block size is reached, thereby maintaining the detail of the original image and avoiding the deterioration of the quality of the image.

With a method of compressing moving image information and the system thereof, according to the third embodiment of the present invention, a frame I is previously spatially divided into blocks I, and when the divided blocks I are scattered between each frame along the time axis, no block I is inserted into any block within the table that has been updated because the difference between the frames is greater than a given parameter (threshold) within a specific period of time; therefore, reconstruction of an image can be performed by initiating the reconstruction of a previously determined number of frames earlier so that a single image can be completely constructed, and visually displaying the reconstructed image after it is reached the objective frame that was placed temporarily, easily visualizing by means of the same a reconstructed image in an optionally temporary position without the search for frame I taking much time. In addition, because the amount of data distributed in a distribution server and / or in the data communication path during the distribution of a moving image is uniformly temporal, a distribution performance higher than that is obtained. with the technique of conventional content distribution. On the reception / reconstruction side, because the change in the amount received per unit time is small, a necessary amount of intermediate zone memory can be reduced, and also because the load imposed on the unit is regulated. processing and reconstruction, even a system with poor performance can play the stable reconstruction. In addition, because the influence of data errors on reconstruction processing is small, reconstruction processing can continue with those data errors that were neglected; therefore, it is not necessary for the system on the distribution side to send the data again, and a lighter load is imposed on the distribution side. In addition, it is also possible to provide the distribution capacity of multiple distributions, etc., for the transmission of the moving image. As described above, the present invention is an optimal means for efficiently compressing moving image information, and can be widely used in the fields of transmitting, receiving, and reconstructing a variety of moving image information.

Claims (42)

1. CLAIMS 1. A method for compressing moving image information comparing the adjacent pixels spatially within a frame or comparing the pixels between adjacent frames temporarily to output the difference information between the pixels; it stores in a bitmap the information regarding whether the information of the difference that came out is greater or not than a given parameter (threshold), and compresses the information of the difference that is stored in the bitmap that is larger than the parameter (threshold) P, reducing the redundant information by the same; the method comprising dividing an image into a block frame and approximating (substituting) each block as a single plane that is represented by at least three components for the pixels within each block, before the compression procedure is initiated. -picture. 2. The method for compressing moving image information according to claim 1, wherein the information that is not larger than the parameter (threshold) that is stored in a bitmap is processed (deleted) as a pixel without changes . 3. The method for compressing moving image information in accordance with either claim 1 or 2, wherein the approximation of each block as a single plane that is represented by at least three components for the pixels, uses an average or the minimum square method. 4. The method for compressing moving image information according to one of claims 1 to 3, wherein intra-frame compression is performed with the plane that is represented by three pieces of data: intensity of a pixel within a block, the intensity gradient within the block in the X direction, and the intensity gradient within the block in the Y direction. The method for compressing motion picture information according to one of claims 1 to 4, wherein the information that is stored in the bitmap is compressed using at least one binary image coding method that is selected from the group consisting of length execution encoding, modified READ coding (MR, MR), coding of Modified Huffman (MH), and JBIG coding. 6. The method for compressing moving image information according to one of claims 1 to 5, wherein the information greater than the parameter (threshold) P is compressed using the adaptive Huffman coding, which uses as many Huffman tables as the expected number of pieces of information. The method for compressing moving image information according to one of claims 1 to 6, further comprising reducing the redundant information between the frames, using entropy coding. The method for compressing moving image information according to claim 7, wherein the entropy coding is performed either through the adaptive Huffman coding, which it encodes using a table that is selected from so many Huffman tables as an expected number of pieces of information, or adaptive arithmetic coding, which codes using a table that is selected from as many arithmetic tables as an expected number of pieces of information. 9. The method for compressing moving image information according to one of claims 1 to 8, wherein the information of the difference between the pixels is used. The method for compressing moving image information according to one of claims 1 to 9, wherein the difference information is the output of the difference through the comparison of the pixel t and the pixel t -1 between the picture . The method for compressing moving image information according to one of claims 1 to 10, where the difference information output is used through the comparison of the pixel t and the pixel t -1 between the frames, where the pixels of nxm between the frames form a single block (n and m are integers of 2 or plus) . The method for compressing moving image information according to one of claims 1 to 11, wherein the difference information output is used through the comparison of the pixel t and the pixel t -1 between the frames , where the pixels of nxm within a frame configure a single block (n and m are integers of 2 or more). The method - for compressing moving image information according to one of claims 1 to 12, wherein with the pixels of n x m between the frames, n denotes 2K (K is a complete number), and m denotes 2? ' (K 'is a complete number). The method for compressing moving image information according to one of claims 1 to 13, further comprising performing the intra-frame compression while changing the size of the divided block within the same frame, before it starts the inter-frame compression procedure. 15. A method for compressing moving image information which: compares adjacent pixels spatially within a frame or compares adjacent pixels temporarily between frames by outputting the resulting difference information for pixels; it stores in a bitmap the information regarding whether the information of the difference that came out is larger or not than a given parameter (threshold), and compresses the information that is stored in the bitmap that is greater than the parameter ( threshold); reducing redundant information by the same; the method comprising performing intra-frame compression while changing the size of the divided block within the same frame, before the inter-frame compression procedure is initiated. 16. A method for compressing moving image information according to claim 15, wherein intra-frame compression is performed by comparing the pixels within each block, while changing the size of the divided block, giving output to the information of the resulting difference for the pixels, and using a smaller block size for the portion that includes the difference information if the difference information is greater than the parameter (threshold) P. 17. A method for compressing moving picture information according to either claim 15 or 16, wherein if the difference information between the pixels is larger than the (threshold) parameter P, a block size is repeatedly used. even smaller. 18. A method for compressing moving image information according to one of claims 15 to 17, further comprising dividing an image into a block frame, and approximating (replacing) each block with a single plane that is represented by At least three components for the pixels within each block. 19. The method for compressing moving image information according to one of claims 15 to 18, wherein the intra-frame compression is performed with the plane that is represented by three pieces of data: intensity of a pixel within a block, the intensity gradient within the block in the X direction, and the intensity gradient within the block in the Y direction. The method for compressing motion picture information according to one of claims 15 to 19, wherein intra-frame compression is performed by compressing the entire image in a block unit of nxm pixels (n and m are integers, respectively) using an intra-frame compression method, which compares the pixels between the image original and the image that was expanded after being compressed, outputting the information of the difference resulting from each pixel, and if there is a pixel that caused a dif larger than the parameter (threshold) P, a much smaller block size is used repeatedly for a portion or a surrounding area that includes this pixel, until a designated minimum block size is reached. 21. The method for compressing motion picture information according to one of claims 15 to 20, wherein when there is no change in the size of the block as a result of the intra-frame compression, inter-frame compression is performed. -picture . 22. The method for compressing moving picture information according to one of claims 15 to 20, wherein when the size of the block changes to a larger size, the data in the block comes out as is, without calculating the difference in the data . The method for compressing motion picture information according to one of claims 15 to 20, wherein when the size of the block changes to a larger size, the difference is calculated from the previous expanded data in each portion. and it is compressed in terms of the unit of block size. 24. The method for compressing moving picture information according to one of claims 1 to 23, further comprising using an intra-frame coded frame I (i.e., a frame of reference only from which it can be reconstructed). an image), which spatially divides the I frame into blocks I, and scatter the I blocks between each frame along the time axis. 25. The method for compressing moving image information according to claim 24, wherein the scattering of the blocks I between each frame along the time axis is performed in such a way that no block I is inserted in any block within the table that has been updated, because the difference between the tables is greater than a given parameter (threshold) within a specific time period. 26. The method for compressing image information in motion, which previously divides an image into a box in blocks, approximating (substituting) each of the blocks divided with a single plane that is represented by three pieces of data: intensity of a pixel within each block, the gradient of each block in the X direction, and the gradient of each block in the Y direction, using an intra-frame coded table I, that is, a reference frame only from which can reconstruct an image), and inserting the I frame in a series of pictures; the method comprising spatially dividing frame I into blocks I, and without inserting any block I into any block within the frame that has been updated because the difference between frames is greater than a given parameter (threshold) within a specific period of time when the I blocks are scattered between each frame along the time axis. 27. A method for compressing moving image information, which compares the adjacent pixels spatially within a frame with one another, to output the difference information of the pixel values; it stores in a bitmap the information regarding whether the information of the difference that came out is larger or not than a given parameter (threshold), and compresses the information that is stored in the bitmap that is greater than the parameter ( threshold), thereby reducing redundant information, the method comprising using an intra-frame encoded taI, that is, a reference frame only from which an image can be reconstructed), which divides the frame spatially I in blocks I, and that does not insert any block I in any block within the tathat has been updated, because the difference between the ta is greater than a given parameter (threshold) within a specific period of time when it disperses the I blocks between each frame along the time axis. 28. A system for compressing moving image information which comprises a bitmap information recoding element for comparing adjacent pixels spatially within a frame with each other or comparing adjacent adjacent pixels temporarily between the ta, outputting the information of the resulting difference for the pixels, and storing in a bitmap the information with respect to whether the information of the difference that came out is greater or not than a given parameter (threshold), and information compression element for compressing the information that is stored in the bitmap that is greater than the parameter (threshold), thereby reducing the redundant information, the system comprising a block approach element for dividing an image within a block frame, before the inter-frame compression procedure is started and approaches (replaces) each block It has a single plane that is represented by at least three components for the pixels within each block. 29. The system for compressing moving image information according to claim 28, wherein the information compression element processes (removes) the information that is stored in the bitmap that is larger than the parameter (threshold), as a pixel without changes. 30. The system for compressing moving image information in accordance with either claim 28 or 29, wherein the block approach element uses an average or the least square approach method, so as to configure a single plane that it is represented by at least three components for the pixels. 31. The system for compressing moving image information according to one of claims 28 to 30, wherein in the block approach element the plane is represented by three pieces of data: intensity of a pixel within a block, the intensity gradient inside the block in the X direction, and the intensity gradient inside the block in the Y direction. The system for compressing moving image information according to one of claims 28 to 31, wherein the information that is stored in the bitmap information recoding element is compressed using at least one binary image coding method that is selected from the group consisting of length execution encoding, modified READ coding (MR, MMR), modified Huffman coding (MH), and JBIG coding. 33. The system for compressing moving image information according to one of claims 28 to 32, wherein the information compression element, which compresses the information greater than the parameter (threshold), performs the customizable Huffman encoding. , which uses as many Huffman tables as the expected number of pieces of information. 34. The system for compressing moving image information according to one of claims 28 to 32, further comprising the entropy coding element, which can reduce the redundant information between the frames; where this entropy coding element performs either the adaptive Huffman coding, which it encodes using a table that is selected from as many Huffman tables as an expected number of pieces of information, or the adaptive arithmetic coding, which encode using a table that is selected from as many arithmetic tables as an expected number of pieces of information. 35. The system for compressing moving picture information according to one of claims 28 to 34, wherein the difference information that is stored by the bit map information recoding element is the difference that comes out through of the comparison of the pixel t and the pixel t - 1 between the frames, where a block is configured by the pixels of nxm (n and m are integers of 2 or more) within a frame. 36. The system for compressing moving image information according to one of claims 28 to 35, wherein the intra-frame compression element performs intra-frame compression in such a manner that the entire image is compressed in a block unit of nxm pixels (n and m are integers, respectively), using an intra-frame compression method, compares the pixels between the original image and the image that expands after being compressed, outputs the information from the resulting difference for the pixels, and if there is a pixel that has caused a difference greater than the parameter (threshold), it repeatedly uses a much smaller block size for a portion or a surrounding area that includes that pixel, until a designated minimum block size is reached. 37. The system for compressing moving image information according to one of claims 28 to 36, wherein when there is no change in the block size as a result of the intra-frame compression that was performed by the element of block approach, interframe compression is performed. 38. The system for compressing moving image information according to one of claims 28 to 37, wherein when the block size is changed to a larger one as a result of the intra-frame compression that was performed by the block approach element, no additional calculation is made for the data difference within the block and it is output as is. 39. The system for compressing moving image information according to one of claims 28 to 37, wherein when the block size changes to a larger size as a result of the intra-frame compression that was performed by the block approach element, the difference is calculated and compressed from the previous expanded data within each portion in terms of the block size unit. 40. The system for compressing moving picture information according to one of claims 28 to 39, further comprising the insertion element of the block I, which uses an intra-frame coded frame I (i.e., a reference frame) only from which an image can be reconstructed), which spatially divides the I frame into blocks I, and which scatters the I blocks between each frame along the time axis. 41. The system for compressing moving image information according to claim 40, wherein the insertion element of the block I does not insert any block I into any block within the frame that has been updated because the difference between the frames is larger than the parameter (threshold) within a specific time period. 42. A system for compressing moving image information, which comprises a block approach element, by means of which an image within the frame is previously divided, and all blocks that were previously divided approach (replace) each one with a only plane that is represented by the intensity of a pixel in each block, the inclination of intensities in each block in the X direction, and the inclination of intensities in each block in the Y direction; the system further comprising the block degeneration element I for spatially dividing an intra-frame coded frame I into blocks I, and the block insertion element I for inserting a block I in a portion, except for the block within of the table that has been updated because the difference between the tables is greater than a given parameter (threshold) within a specific time period, when the blocks I are scattering between each frame along the time axis.