WO2001017229A1 - Procede de compression d'image - Google Patents
Procede de compression d'image Download PDFInfo
- Publication number
- WO2001017229A1 WO2001017229A1 PCT/JP2000/004472 JP0004472W WO0117229A1 WO 2001017229 A1 WO2001017229 A1 WO 2001017229A1 JP 0004472 W JP0004472 W JP 0004472W WO 0117229 A1 WO0117229 A1 WO 0117229A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image data
- data
- luminance
- point
- image
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/41—Bandwidth or redundancy reduction
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
- H04N1/46—Colour picture communication systems
- H04N1/64—Systems for the transmission or the storage of the colour picture signal; Details therefor, e.g. coding or decoding means therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
- G06T9/005—Statistical coding, e.g. Huffman, run length coding
Definitions
- the present invention relates to an image processing technique for compressing digital image data such as a full-color image such as an additive three primary colors (RGB) or a subtractive three primary colors (CMY) or a black and white 256-tone (gray scale) image.
- RGB additive three primary colors
- CY subtractive three primary colors
- RGB black and white 256-tone
- the color image processing is performed using digital image data consisting of the luminance information of each color component in a color model such as RGB or CMY, that is, the color channel (color model).
- RGB or CMY that is, the color channel (color model).
- This digital image data is represented by one byte that expresses the luminance of each pixel, for example, for each RGB color in 256 levels from 0 to 255 (256 levels). (8 bits) of information is allocated, and accordingly, each pixel has 3 bytes of information. For this reason, the amount of data of high-definition color images in particular is enormous.
- technology that compresses image data is needed.
- Japanese Patent Application Laid-Open No. 10-32719 discloses a sufficient compression ratio and high image reproduction.
- color image data consisting of luminance information of three primary colors is converted into color model data consisting of one luminance information and two pieces of color difference information, and this is compressed by a method that can be reproduced with high quality.
- An image compression method and apparatus are disclosed.
- As a high-quality reproducible compression processing technology in the XZ plane, where the luminance distribution of each pixel in the XY image plane is represented by the size in the Z-axis direction, the luminance changes continuously within a certain allowable error range.
- an object of the present invention is to realize a high compression effect and a high degree of reproduction of a continuous change of a color tone even for a high-definition digital image data. Accordingly, it is an object of the present invention to provide an image compression method capable of reproducing a high-quality image that can be used as an image original for printing. Disclosure of the invention
- a bitmap digital image data having one or a plurality of color channels is converted into a two-dimensional vector using the horizontal position and luminance of each pixel for each horizontal column. And quantizing according to the magnitude of the brightness change with reference to the line segment connecting the start point and the end point of each column,
- the vectorized and quantized image data in the horizontal direction is transformed into a two-dimensional vector for each row in the vertical direction using the position and luminance in the vertical direction, and a line segment connecting the start point and the end point of each row.
- Quantization is performed according to the magnitude of the luminance change on the basis of, to convert the image data into compressed image data having a matrix structure based on the significance of the luminance information of each pixel.
- the luminance information of each pixel included in the image data is converted into a two-dimensional vector and quantized according to the magnitude of the luminance change. Since the evening is hierarchized and reconstructed, pixels with practically no significance Luminance information will be absorbed in the same vector. Therefore, the amount of original image data can be reduced without substantially losing the information contained in the original image data, and the most efficient data structure particularly suitable for compressing high-definition image data Can be created.
- the original bitmap digital image data is divided into unit image data of a predetermined size, and the vectorization and quantization in the horizontal direction are performed for each of the divided unit image data. Then, by performing vectorization and quantization in the vertical direction, it can be converted into unit compressed image data having a matrix structure, and finally, the unit compressed image data can be integrated. As a result, even when the amount of original image data is relatively large, efficient processing can be performed according to the processing capacity of the combination image and the required image quality.
- the original bitmap / digital image data is divided for each color channel, and the divided image data for each color channel is converted into compressed image data having a matrix structure. It can be integrated into one compressed image data. As a result, for example, image processing of RGB or CMY systems can be reduced to 1/3 the processing unit, so that efficient processing can be performed.
- each unit image data is further divided for each color channel and divided. After converting the unit image data for each color channel into unit compressed image data of matrix structure, and integrating them into one unit compressed image data, more efficient compression is achieved. Processing becomes possible.
- the compressed image data having the matrix structure generated as described above can be further arithmetically compressed.
- the image data can be compressed at a higher level without loss.
- the process of quantizing in the horizontal or vertical direction is based on the line connecting the starting point and the ending point of each column or row as a reference vector, and the maximum positive and / or negative deviation point in the section is defined.
- the process of calculating this maximum deviation point is described as The process is repeatedly performed until the deviation of the reference vector loses its significance as the desired image data, and each pixel is classified according to the magnitude of the deviation at each maximum deviation point. It consists of generating hierarchical data.
- the plurality of hierarchical data are stored in 8-bit, 6-bit, 4-bit, and 1-bit data, respectively. It is preferable that the data be composed of first to fourth hierarchical data.
- FIG. 1 is a flowchart schematically showing a preferred embodiment of the image compression method according to the present invention.
- FIG. 2 is a flowchart showing the process of step 3 in FIG. 1 for dividing image data into optimal processing units.
- FIG. 3 is a flowchart schematically showing the process of step 4 in FIG. 1 in which image data divided into optimal processing units is vectorized to generate matrix-structured image data.
- FIG. 4 is a flowchart showing a process of generating the first hierarchical data by vectorizing the image data in the horizontal direction.
- FIG. 5 is a flow chart showing a process of vectorizing the image data in the horizontal direction to generate the second layer data.
- FIG. 6 is a flowchart showing a process of generating the third hierarchical data by vectorizing the image data in the horizontal direction.
- FIG. 7 is a flowchart showing a process of generating the fourth hierarchical data by vectorizing the image data in the horizontal direction.
- FIG. 8 is a flowchart schematically showing a process of transforming an image data horizontally vectorized into vertical vectors after shaping processing.
- FIG. 9 is a flowchart showing a process of generating a V FZ image file of the present invention by post-processing image data that has been vectorized in the horizontal and vertical directions.
- FIG. 10 is a flowchart showing a process of reproducing the generated VFZ image file of the present invention.
- FIG. 11 is a diagram illustrating an example of the division of the image data.
- FIG. 12 is a diagram in which the image data divided in FIG. 11 is further divided into: R, G, and B color channels.
- Fig. 13A shows the image data of one column in the horizontal direction of the color channel R.
- Fig. 13B shows the luminance distribution, that is, the magnitude of the luminance with respect to the pixel position (length from the starting point).
- FIG. 3 is a diagram for explaining the concept of vectorization.
- Fig. 14A shows the first to fourth hierarchical data generated from one row of image data in the horizontal direction by vectorization.
- Fig. 14B shows the image data of all the columns vectorized in the horizontal direction. It is a figure showing the state where shaping processing was carried out.
- FIG. 15 is a conceptual diagram showing the distribution of image data that has been vectorized in the horizontal and vertical directions in the first to fourth layers.
- FIG. 16 is a diagram showing an array of image data vectorized in the horizontal and vertical directions.
- FIG. 1 is a flowchart schematically showing a process of generating a compressed image file from a full-color image original, that is, an original image, using a method according to an embodiment of the present invention.
- the desired image manuscript is electronically processed, for example, bitmap data having multiple color channels such as RGB or CMY, or one color channel expressed in grayscale only.
- image data is generated (step S1). Normally, such image data is obtained using various known means, such as an image scanner and a digital camera, at an arbitrary input resolution and a preset value of, for example, 12 bits or 8 bits.
- the image data is processed online on a computer or workstation via a recording medium such as a CD-R, MO disk, or DVD, or a network using Ethernet or the like. Entered (step S 2).
- the processing computer pre-processes the input image data and stores the pre-processed image data in a memory.
- Step S3 First, as shown in FIG. 2, the entire image data is expanded in the memory (step S21). Next, the information contained in the header label of the image data is analyzed to determine the size of the entire image (number of dots), the configuration of one color channel, and the amount of luminance information (number of bits) assigned to one pixel. After the confirmation, the image is divided into image sizes of the optimal processing unit determined in advance according to the analysis result, and each of the divided unit image data is reconstructed on the memory (step S22).
- This optimal processing unit can be set to a fixed image size in advance, or can be selected appropriately according to the processing speed of the processing unit, the capacity such as the memory capacity, and the image reproduction conditions such as the required image quality. Can be
- the RGB color image data having a resolution of 1,024 ⁇ 1024 dots is compressed.
- the optimal processing unit is 256 x 256 dots
- the image data in Fig. 11 is divided into four in the vertical and horizontal directions, and reconstructed into a total of 16 unit image data.
- each pixel has an information amount (brightness value) of 3 bytes in total for each color, R, G, B channel, 1 byte each.
- this unit image data is divided into unit image data for each channel of each color of R'G and B, and reconstructed on a memory.
- Step S23 Data compression processing is performed using the unit image data of each color channel as the minimum processing unit.
- each pixel has an information amount (luminance value) of one byte. Therefore, the data amount of the minimum processing unit processes the entire original image data. It decreases from 1024 2 X 3 2 bytes in the case, up to 256 x 25 6 by Bok.
- the unit image data is sequentially subjected to vector decomposition processing for each single color channel to generate image data converted into a matrix structure (step S4).
- the image data of the minimum processing unit is vectorized in the horizontal direction one column at a time for each color channel, and the size of the luminance information of each pixel (number of bits) ) To generate the first to fourth hierarchical data (Step S31).
- the image processing in the minimum processing unit consists of 256 horizontal columns. Each pixel has a luminance value expressed by the amount of information of one byte.
- Figure 13B shows the horizontal column, where the horizontal axis represents the position of each pixel with a length from 0 to 255, and the vertical axis represents the luminance value of each pixel in 256 steps from 0 to 255. The luminance distribution of one horizontal row of pixels is shown.
- the information of the luminance distribution and the continuous change of the luminance can be held.
- the vectorization of the image data is performed.
- a line segment connecting these is defined as a mask.
- the evening vector P 1 P2 is set to [x2—xl, y2-y1] (step S41).
- the luminance deviation is calculated based on the mass vector, and points D 1 (xll, yll) and D 2 where the maximum positive and negative deviations are obtained (xl2, yl2) is determined (step S42).
- step S44 it is determined whether or not the absolute value of the deviation at the maximum deviation points D1 and D2 is less than or equal to 64, that is, whether or not the information amount is less than or equal to 6 bits (step S44). If it is not less than or equal to 64, that is, if it is greater than this, this point D1 or D2 is evaluated as having a change in luminance with respect to the start point and the end point having an information amount to be represented by 8 bits. Conversely, if the value is 64 or less, this point D1 or D2 is evaluated as having the amount of information to be represented by 6 bits or less in the amount of change in luminance with respect to the start point and the end point.
- the generation process of the first layer data in the section between the start point and the end point ends, and the process moves to the next generation process of the second layer data.
- Step S45 the sub-vector generated between each two adjacent points including the point D1 and / or D2 is stored as the first layer data.
- Xy is similarly calculated between the two adjacent points.
- a two-dimensional subvector for example, P1D11, D11D12, D12D1, etc., is generated (step S43). Then, it is determined whether or not the absolute value of the maximum deviation is 64 or less (step S44). If the absolute value is 64 or less, the generation process of the first hierarchy data is terminated for that section, and the next second hierarchy data is generated. It shifts to the evening generation process. If it is larger than 64, the newly generated sub-vector is similarly added to the first layer data as two x and y quantities. Thus, the first layer data shown in FIG. 14A is generated in the memory.
- the absolute value of the maximum deviation with respect to the reference vector was calculated. It is determined whether it is 16 or less (step S51). If it is not less than 16 or greater, it is evaluated that the amount of change in luminance relative to the reference sub-vector has the amount of information to be represented in 6 bits. Conversely, if the value is 16 or less, the point is that the amount of change in luminance with respect to the reference subvector is evaluated to have the amount of information to be represented by 4 bits or less. Is completed, and the process proceeds to the next step of generating the third layer data.
- the sub-vector generated between each two adjacent points including those points in step S43 of the first layer The data is stored in the memory as a night (Step S52). These sub The vector is arranged in succession as shown in Fig. 14A, with two quantities x and y each having 6 bits of information as one data. Furthermore, in each section connecting these two adjacent points, the luminance deviation is calculated based on each sub-vector, and the point where the maximum positive and negative deviation is obtained is determined.
- the xy two-dimensional subvector is similarly calculated between each two adjacent points. Is generated (step S53). Then, for those new maximum deviation points, it is determined whether the absolute value of the maximum deviation is equal to or less than 16 (step S51). If the absolute value of the maximum deviation is larger than 16, the newly generated sub-vector is x, y Add to the second layer data as two quantities.
- Steps S51 to S53 are repeatedly executed for all newly determined maximum deviation points until the absolute value of the maximum deviation becomes 16 or less. Subvectors determined to be larger than 16 are sequentially added as second-layer data, and finally the second-layer data shown in FIG. 14A is generated on the memory.
- the process of generating the third hierarchical data is performed in the same manner as the process of generating the second hierarchical data, as shown in FIG. That is, for each point determined to have a maximum deviation of 16 or less in step S51 of the second layer data generation process, it is determined whether the absolute value of the maximum deviation with respect to its reference vector is 4 or less (step S61). If it is not less than 4 or larger, it is evaluated that the amount of change in luminance with respect to the reference sub-vector has the amount of information to be represented by 4 bits. Since the amount of change in the luminance with respect to the vector is evaluated to have the amount of information to be represented by 2 bits or less, the generation process of the third hierarchical data is completed, and the generation of the next fourth hierarchical data is completed. Move on to the process.
- the sub-vector generated between each two adjacent points including those maximum deviation points in step S53 of the second layer data generation process is It is stored in the memory as evening (step S62).
- These subvectors are arranged consecutively, as shown in FIG. 14A, with two x and y quantities each having a 4-bit information amount as one data item. Furthermore, in each section connecting these two adjacent points, each sub-vector is used as a reference. Calculate the luminance deviation and determine the point where the maximum positive and negative deviation is taken respectively.
- an Xy two-dimensional subvector is similarly calculated between each two adjacent points. Generate (Step S63). Then, for those new maximum deviation points, it is determined whether the absolute value of the maximum deviation is less than or equal to 4 (step S61), and if it is greater than 4, the newly generated subvector is x, y2 Add it to the third layer as overnight.
- Steps S61 to S63 are repeated for all the newly determined maximum deviation points until the absolute value of the maximum deviation becomes 4 or less.
- the sub-vectors determined to be larger than 4 are sequentially added as third-layer data, and finally the third-layer data shown in FIG. 14A is generated in the memory.
- the maximum deviation from the reference vector was calculated. It is determined whether the absolute value is 1 or 0 (step S71). If it is either 1 or 0, the point is evaluated as having the amount of change in luminance with respect to the reference sub-vector having the amount of information to be represented by 1 bit.
- the inter-vector is stored in the memory as the fourth layer data using two quantities x and y as shown in FIG. 14A (step S72).
- the point is evaluated as having a change in luminance relative to the reference vector that has more than one bit of information.
- the luminance deviation is further calculated based on the sub-vector generated between each two adjacent points including the maximum deviation point in step S63 of the third hierarchical data generation process.
- the point at which the maximum positive and negative deviation is taken is determined (step S73).
- an Xy two-dimensional subvector is similarly generated between each two adjacent points thereof ( In step S74), it is determined again whether the absolute value of the maximum deviation is 1 or 0 (step S71).
- steps S71, 73 and 74 show that for all the maximum deviation points generated in this way, the amount of change in luminance with respect to the reference vector is 1 bit. It is evaluated as having the amount of information to be represented, and is repeatedly executed until it is stored in the memory as fourth-layer data. As a result, the data of all points remaining after the generation of the third layer data is converted into a vector with two quantities x and y expressed by 1 bit, and the fourth layer Generated as data.
- the luminance change of each pixel is subjected to the vector decomposition processing, and the first to fourth hierarchical data having different numbers of bits depending on the size are processed. It is generated and arranged on the memory in order from the first column to the 256th column.
- information on the start point P1 and the end point P2 of the original image data that is, the horizontal length and the luminance value are arranged at the head of each column.
- data indicating the boundary is inserted in each layer data overnight.
- the image data divided into the minimum processing units is further processed independently for each column. Therefore, the size of the data to be processed in one round is small. Also, if this data is reconstructed after being expanded and processed in a certain area of the memory, the original data becomes unnecessary, so that those memory areas can be released and used for the next processing. . For this reason, in the present invention, the memory of the processing console can be used more effectively than in the past, and depending on the processing capacity, the above-described horizontal columns can be simultaneously or in parallel with a certain time delay. Because processing can be performed, processing speed is greatly improved. Further, even a processing computer having relatively low capacity can process large-size image data.
- the length of the first to fourth layers is different for each column. Therefore, in this embodiment, for each layer of data, the one with the longest data length is selected from all columns and is used as the reference data length. If the data length is shorter than that, In each case, a value of “0” is entered in the shaded data length area in Fig. 14B.
- the image data of the minimum processing unit is such that each hierarchical data has the same data length from the first column to the 256th column, that is, each row in the vertical direction has the same 256 It is shaped so that it consists of continuous data and reconstructed in memory (Step S32) 0
- step S33 The vectorization of each row is performed in the same manner as the horizontal vectorization described above according to the process shown in FIG. 8, and thereby the first to fourth hierarchy data are generated for each row.
- a luminance deviation is calculated with respect to the other pixels on the basis of a mask vector, and a point V1 (z11, yll) at which a maximum positive and / or negative deviation is obtained.
- yl, y2, yll, and y12 used for each point of Ql, Q2, VI, and V2 are the points P1, P2, Dl, and D2 of each point used in the horizontal vectorization. Different from yl, y2, yll, yl2.
- the sub-vectors represented by the two quantities z and y each having an 8-bit information amount are stored in the memory as the first layer data. Then, based on the sub-vectors, new maximum points of positive and negative deviations are determined, and as long as the absolute value of the deviation exceeds 64, those sub-vectors are added to the memory. Generate the first layer data.
- Step S 8 When the absolute value of the maximum deviation of all newly determined maximum deviation points becomes 64 or less, the process similarly proceeds to the generation process of the second hierarchy data, and the same as described above with reference to FIG.
- Vectorization is performed in accordance with the process of (1), and a second layer data consisting of only data having a 6-bit information amount is generated and stored in a memory.
- vectorization is performed in the same manner as described above with reference to FIG. 6 to generate a third-layer data consisting of only data having a 4-bit information amount and store it in the memory.
- the fourth layer data composed of only 1-bit data is generated by the same vectorization processing as described above with reference to FIG. 7, and is similarly stored in the memory. (Step S 8 4) o
- each step of generating the first to fourth hierarchical data is performed for each row with respect to the small-sized data. Independently performed for each row, the original memory area of the reconstructed data is sequentially opened and used for the next processing, thereby enabling effective use of the memory, and simultaneous or time-delay of each row. Parallel processing can significantly improve processing speed. Also, even a relatively large-sized image data can be processed by a processing computer having a relatively low capacity.
- the image data processed by the vector decomposition in the vertical direction is divided into the start point and end point and the first to fourth layer data in advance as shown in Fig. 14B. From the first layer data, all of the first to fourth layer data is generated, whereas the second to fourth layer data is generated from the second layer data, and the third to fourth layer data is generated from the third layer data. In addition, only the fourth-level data is generated from the fourth-level data and from the fourth-level data. Actually, when the inventor of the present invention executed the image compression processing of this embodiment for some full-color test images, the first to fourth hierarchical data roughly showed the area a and the area b + c shown in FIG. , Area d + e + f, area g + h + i + j
- these data groups a to j are arranged in the order of the first to fourth hierarchical data, that is, arranged in ascending order of the number of bits, shaped and reconstructed on the memory (step S). 8 5).
- compressed image data having a desired matrix structure is obtained from the minimum processing unit, that is, the unit image data of a single color channel (step S4).
- the luminance information of each pixel is vectorized and the data is hierarchized and reconstructed on the basis of the degree of significance thereof, the luminance information of the pixel having substantially no significance is represented by 1 Can be absorbed in one vector. Therefore, by simply hierarchizing in this way, the data amount can be reduced without substantially losing the information contained in the original file, and especially for compressing high-definition image data with a large amount of information. Thus, the most efficient data structure can be created.
- the process of layering and reconstructing the image data that has been converted into a two-dimensional vector in the horizontal and vertical directions in this way. Can be referred to as quantization because it is hierarchized by dividing the data based on a certain number of bits.
- This image data is subjected to compression processing by a conventional technique (step S5).
- the compression processing is performed on each of the first to fourth hierarchical layers by using, for example, arithmetic compression combining known publicly known run-length compression processing and Huffman compression processing.
- a compressed image according to the present invention is generated for unit image data of a single color channel.
- the image data is hierarchized in this manner, the effect of arithmetic compression is higher than in the past.
- arithmetic compression in addition to layering of the image data, it is possible to compress the image data at a high compression ratio as a whole, despite losslessness.
- the vector decomposition processing is performed for each of them, and the image data is converted into a matrix structure image data, and the compression processing is performed. Generate compressed data according to the invention.
- step S6 the image compression data of the three color channels generated separately in this way are integrated into one to generate compressed RGB image data of an optimal processing unit and stored in the memory.
- Step S91). 1 When the generation of the compressed image data for all six unit image data is completed, these are integrated into one to generate the VFZ image file of the present invention in which the entire original image data is compressed.
- Step S92 At this time, a header label describing the file size, data structure, etc. is inserted at the beginning of the VFZ image file.
- the VFZ image file is output from a processing computer to a reproduction computer or workstation, usually online via a network or using a recording medium such as a CD-R (step). S7).
- playback conditions such as output resolution and image size are input.
- the VFZ image file is decompressed, and the image data having the above-described matrix structure is generated based on the reproduction conditions.
- the image data is directly output from a playback device to a predetermined screen of a display device or the like, stored in a storage device of a server, or transferred online.
- the output image data has the matrix structure described above with reference to FIG. Therefore, when the first-level data that is output earlier is input, the entire image is displayed on the screen, and as the data at the second and lower levels are input, a more detailed image is displayed. When all the data is input to, the original high-definition color image is reproduced.
- the case of compressing RGB image data has been described.However, the present invention is applicable to color data of CMY system and other formats, or image data having a single color channel such as gray scale. Can be applied as well. Also, as will be apparent to those skilled in the art, the present invention can be implemented by adding various changes and modifications to the above-described embodiment within the technical scope thereof.
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Compression Of Band Width Or Redundancy In Fax (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
- Color Television Systems (AREA)
- Image Processing (AREA)
Description
Claims
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2001521051A JP3829217B2 (ja) | 1999-08-27 | 2000-07-05 | 画像圧縮方法 |
US10/069,227 US7031514B1 (en) | 1999-08-27 | 2000-07-05 | Image compression method |
EP00944269A EP1251686A1 (en) | 1999-08-27 | 2000-07-05 | Image compressing method |
CA002383053A CA2383053A1 (en) | 1999-08-27 | 2000-07-05 | Image compressing method |
KR1020027002551A KR20020042816A (ko) | 1999-08-27 | 2000-07-05 | 화상압축방법 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP11/283296 | 1999-08-27 | ||
JP28329699 | 1999-08-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2001017229A1 true WO2001017229A1 (fr) | 2001-03-08 |
Family
ID=17663623
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2000/004472 WO2001017229A1 (fr) | 1999-08-27 | 2000-07-05 | Procede de compression d'image |
Country Status (7)
Country | Link |
---|---|
US (1) | US7031514B1 (ja) |
EP (1) | EP1251686A1 (ja) |
JP (1) | JP3829217B2 (ja) |
KR (1) | KR20020042816A (ja) |
CN (1) | CN1387722A (ja) |
CA (1) | CA2383053A1 (ja) |
WO (1) | WO2001017229A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007178166A (ja) * | 2005-12-27 | 2007-07-12 | Rkc Instrument Inc | ヒータ断線検知方法 |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006166355A (ja) * | 2004-12-10 | 2006-06-22 | Ricoh Co Ltd | 画像処理装置、画像処理方法、画像処理プログラム及び記録媒体 |
EP1670257B1 (en) | 2004-12-10 | 2018-09-26 | Ricoh Company, Ltd. | Compressing a mutlivalue image with control of memory space requirement |
CN101123668B (zh) * | 2006-08-08 | 2010-12-15 | 上海西门子医疗器械有限公司 | Ct图像的压缩方法 |
US8320694B2 (en) * | 2009-03-27 | 2012-11-27 | Sharp Laboratories Of America, Inc. | Surround error diffusion |
US8274705B2 (en) * | 2009-03-27 | 2012-09-25 | Sharp Laboratories Of America, Inc. | Multi-level surround error diffusion |
CN105938624A (zh) * | 2016-04-15 | 2016-09-14 | 张志华 | 一种犯规监控装置 |
CN107452039B (zh) * | 2016-06-01 | 2023-03-31 | 上海东方传媒技术有限公司 | 压缩rgb颜色空间的方法及装置 |
CN111462258B (zh) * | 2020-03-31 | 2023-09-15 | 上海大学 | 一种针对制作印刷菲林片的纹理线条图像矢量化方法与系统 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04579A (ja) * | 1990-04-17 | 1992-01-06 | Matsushita Electric Ind Co Ltd | 図形の特微点抽出方法 |
JPH09298747A (ja) * | 1996-04-30 | 1997-11-18 | D & I Syst Kk | 多階調画像データ圧縮・復元方法 |
JPH1032719A (ja) * | 1996-07-18 | 1998-02-03 | Benetsuse Corp:Kk | 画像圧縮方法および画像圧縮装置 |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4205341A (en) * | 1978-01-24 | 1980-05-27 | Nippon Telegraph And Telephone Public Corporation | Picture signal coding apparatus |
JPS57150275A (en) * | 1981-03-11 | 1982-09-17 | Fuji Photo Film Co Ltd | Adaptive quantizer |
US5072290A (en) * | 1986-09-19 | 1991-12-10 | Canon Kabushiki Kaisha | Color image signal encoding device |
US5828784A (en) * | 1992-07-13 | 1998-10-27 | Hitachi Denshi Kabushiki Kaisha | Data coding method and apparatus using a plurality of blocks of different length |
JPH08194816A (ja) | 1995-01-17 | 1996-07-30 | Pesupu Purosu Kk | 線分近似方法および線分近似方式 |
JP3957915B2 (ja) * | 1999-03-08 | 2007-08-15 | パイオニア株式会社 | フェード検出装置及び情報符号化装置 |
US6868190B1 (en) * | 2000-10-19 | 2005-03-15 | Eastman Kodak Company | Methods for automatically and semi-automatically transforming digital image data to provide a desired image look |
JP2004515134A (ja) * | 2000-11-22 | 2004-05-20 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | 映像信号処理 |
FI112424B (fi) * | 2001-10-30 | 2003-11-28 | Oplayo Oy | Koodausmenetelmä ja -järjestely |
-
2000
- 2000-07-05 JP JP2001521051A patent/JP3829217B2/ja not_active Expired - Fee Related
- 2000-07-05 CA CA002383053A patent/CA2383053A1/en not_active Abandoned
- 2000-07-05 WO PCT/JP2000/004472 patent/WO2001017229A1/ja not_active Application Discontinuation
- 2000-07-05 KR KR1020027002551A patent/KR20020042816A/ko not_active Application Discontinuation
- 2000-07-05 CN CN00814924A patent/CN1387722A/zh active Pending
- 2000-07-05 US US10/069,227 patent/US7031514B1/en not_active Expired - Fee Related
- 2000-07-05 EP EP00944269A patent/EP1251686A1/en not_active Withdrawn
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH04579A (ja) * | 1990-04-17 | 1992-01-06 | Matsushita Electric Ind Co Ltd | 図形の特微点抽出方法 |
JPH09298747A (ja) * | 1996-04-30 | 1997-11-18 | D & I Syst Kk | 多階調画像データ圧縮・復元方法 |
JPH1032719A (ja) * | 1996-07-18 | 1998-02-03 | Benetsuse Corp:Kk | 画像圧縮方法および画像圧縮装置 |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007178166A (ja) * | 2005-12-27 | 2007-07-12 | Rkc Instrument Inc | ヒータ断線検知方法 |
Also Published As
Publication number | Publication date |
---|---|
EP1251686A1 (en) | 2002-10-23 |
US7031514B1 (en) | 2006-04-18 |
JP3829217B2 (ja) | 2006-10-04 |
CN1387722A (zh) | 2002-12-25 |
KR20020042816A (ko) | 2002-06-07 |
CA2383053A1 (en) | 2001-03-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP2914226B2 (ja) | 可逆変換を可能にするディジタル信号の変換符号化方式 | |
US5701369A (en) | Fractal image compression device and method | |
US6978048B1 (en) | Encoding method and apparatus | |
JP3829217B2 (ja) | 画像圧縮方法 | |
WO2020115471A1 (en) | Bit plane encoding of data arrays | |
US7248744B2 (en) | Vector quantization of images | |
JP3764765B2 (ja) | ディジタル画像の処理方法及びシステム | |
JPH06232764A (ja) | データ圧縮装置および方法 | |
CN112118449B (zh) | 一种图像压缩与解压缩的方法及装置 | |
US6118898A (en) | Methods and apparatus for compressing and decompressing multiple data sets and for managing memory storage of compressed data sets | |
JPH07177502A (ja) | 画像情報圧縮方法、圧縮画像情報記録媒体、圧縮画像情報再生装置 | |
US20060235915A1 (en) | Method and apparatus for converting data, method and apparatus for inverse converting data, and recording medium | |
US5999112A (en) | Data compression apparatus and method, data expansion apparatus and method, and recording medium | |
JPH06217144A (ja) | 画像の量子化ディジタル表現から信号レベルの解像度を増すための方法及び装置 | |
JPWO2009122463A1 (ja) | 画像圧縮装置、復元装置、圧縮方法、復元方法及びプログラム | |
JP2001326935A (ja) | 画像符号/復号方法及びその装置並びにそのプログラムを記録した記録媒体 | |
WO1998058328A1 (en) | Digital signal converting method and device, transformation matrix generating method and device, and supply medium | |
JP3777649B2 (ja) | 画像圧縮方法及び画像伸張方法 | |
JP3862418B2 (ja) | 画像復号化方法および装置並びに画像復号化方法をコンピュータに実行させるためのプログラムを記録したコンピュータ読取り可能な記録媒体 | |
JP3209396B2 (ja) | 画像データの圧縮方法及び装置 | |
CN1547162A (zh) | 一种基于金字塔四叉树的纹理映射方法 | |
JPH05260311A (ja) | 圧縮画像デ−タの伸長方法 | |
WO2001016762A1 (fr) | Procede et systeme d'acheminement d'images | |
JPH0621828A (ja) | ベクトル量子化復号化器 | |
WO2002047396A1 (en) | Image format |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA CN JP KR US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AM AZ BY KG KZ MD RU TJ TM AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2383053 Country of ref document: CA |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1020027002551 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2000944269 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 008149240 Country of ref document: CN |
|
WWP | Wipo information: published in national office |
Ref document number: 1020027002551 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: 10069227 Country of ref document: US |
|
WWP | Wipo information: published in national office |
Ref document number: 2000944269 Country of ref document: EP |
|
WWR | Wipo information: refused in national office |
Ref document number: 1020027002551 Country of ref document: KR |
|
WWW | Wipo information: withdrawn in national office |
Ref document number: 2000944269 Country of ref document: EP |