CN1477601A - Image information coding method - Google Patents
Image information coding method Download PDFInfo
- Publication number
- CN1477601A CN1477601A CNA021298947A CN02129894A CN1477601A CN 1477601 A CN1477601 A CN 1477601A CN A021298947 A CNA021298947 A CN A021298947A CN 02129894 A CN02129894 A CN 02129894A CN 1477601 A CN1477601 A CN 1477601A
- Authority
- CN
- China
- Prior art keywords
- data
- image
- picture
- image document
- coding method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The image data coding method for transmission of continuous image includes the following steps: fetching continuous image, using fuzzy comparison method to compare difference of continuous image, analyzing difference quantity of image, and selecting proper coding algolithm according to the difference quantity, for example JPEG or variable length and triangular transform coding, the make image compression. According to the different image data said invention can select various coding methods, such as directly-coding method, variable length coding method and triangular transform coding method.
Description
Technical field
The present invention relates to a kind of method of compression of images, particularly a kind of difference that has according to image document, the method for compressing image of adjustment compress mode.
Background technology
Image Compression is very important on tele-control system.When tele-control system carries out remote computer control, can utilize the direct transmission of image, make client computer in the distance carry out the control of computing machine by network or the like information transmission media.JPEG is a kind of algorithm of compression of images, and it is by (the International Organization forStandardization of International Standards Organization; ISO) and (the InternationalTelegraph and Telephone Consultative Committee of Consultative Committee on International Telephone and Telegraph (CCITT); CCITT) a digital picture compression standard of being set up mainly is to be used for still image compression aspect.When the algorithm that uses JPEG carried out compression of images, digital picture will be according to the requirement of jpeg algorithm, and elder generation is a plurality of blocks (Block) with image division, carries out the compression and the transmission of each block again.Generally speaking, for requiring preferable image quality, even the pixel change of part is only arranged in each block, but whole block all can be required compression and transmit.
Fig. 1 is existing tele-control system, utilizes the technology of compression of images, makes the client computer can be by other computing machine of Long-distance Control.As shown in FIG., this tele-control system has controller 120, computer switch (Keyboard-Video-Mouse Switch; KVMSwitch) 104, with four computing machines of being controlled by computer switch 104 1～4102.These remote controllers 120 are by network 122 and client computer 124 contacts.And these remote controllers 120 comprise, A/D converter 106, first in first out controller (First IN First Out; FIFO) 108, output load module 110, CPU (central processing unit) (Central Processing Unit; CPU) 112, Memory Controller 114, storer 116 and network card 118.
After if A/D converter 106 carries out the sampling of image document, producing 200 pixels (Pixel) changes, and it is distributed in 200 blocks, and each block is when having the situation of 16*16 pixel, in order to compress and transmit these image documents, system must have powerful compression and reading capability, finishing the processing of these image documents in the reasonable time interval.
And after these image documents are finished compression, further use network to carry out the transmission of compressed images data if desire, then under the limited available bandwidth of network, even the compressed image document of finishing, desire to carry out the real-time Transmission of so huge image document, will cause network to bear greatly.Generally speaking, utilize compress technique effectively to express and separate the bottleneck that data transmission in network forms, but because image document transmits in network in a large number at present, existing network is bearing a large amount of image documents, under limited frequency range, desire to carry out realtime graphic transmission and computer control, will be subject to the transmittability of network.
How effectively continuous the and huge image document of compression to utilize limited network bandwidth, carries out the image real-time Transmission, and carries out the control of remote computer, is network and the common target of server managers in fact.
Summary of the invention
In above-mentioned background of invention, finish even continuous and huge image document has compressed, but utilize finite element network frequency range and transmittability, desire to carry out the control of continuous images real-time Transmission and remote computer, will be subject to the restriction of network capacity.How effectively to compress the consecutive image data to alleviate the burden of network, real is user's common hope.
One of purpose of the present invention is to utilize a kind of method of transmitting the image document of continuous compression via network effectively.
Another purpose of the present invention is to utilize more present image block and previous image block, calculates wherein different pixel quantities, calculates the algorithm of the compressed image data of required use automatically, carries out the compression and the transmission of image document.
A further object of the present invention, it is the algorithm that utilizes the proper compression image document, required according to the different images situation, carry out different compression calculations, effectively reduce the demand of network bandwidth and transfer rate, make via Network Transmission realtime graphic and control computer equipment more easy and time delays thereby reduction.
According to above-described purpose, the invention provides a kind of image information coding method, be used in the consecutive image transmission, can reduce coding back image document size, thereby reduce demand network bandwidth.Be particularly suitable for computer switch (Keyboard-Video-Mouse Switch; KVM Switch) via network, the needs of the image transmission during the management remote computer.The method comprises, read consecutive image, use fuzzy relative method, compare the preceding image of consecutive image and the image document of each pixel of the same block of back image, the discrepant quantity of analysis image, the heteromerism amount of being on duty used the JPEG encryption algorithm to carry out the compression of image more than or equal to 32 o'clock, be on duty the heteromerism amount less than 32 o'clock, use change length and three angle modulation encryption algorithms to carry out compression of images.
Wherein fuzzy relative method, need to set earlier a fuzzy comparison range, the consecutive image data of same position is separated red data, green data and the blue data of forming, the absolute value of the red data of an image after the red data of last image deducts, the blue data of previous image deducts the absolute value of the blue data of back one image, and the absolute value of green data that the green data of previous image deducts back one image is considered as this image document identical during all smaller or equal to this fuzzy comparison range; Otherwise if the front and back difference of arbitrary color during greater than this fuzzy comparison range, then is considered as this image document inequality.
Before bluring relative method, can down to the data difference that connects comparison front and back image image identical position in front and back be got rid of earlier.
The present invention further rearranges the nonlinear data of different image document becomes linear data, uses direct coding algorithm, three angle modulation encryption algorithms or change length coding algorithm to carry out the coding of different image document according to the discrepant quantity decision again.And when the identical image data, if the length of data behind the coding that is produced through three angle modulation encryption algorithms and change length coding algorithm behind the coding that is produced greater than the direct coding algorithm during length of data, then uses the direct coding algorithm to encode.
Change length coding algorithm of the present invention also comprises, and the first change length coding algorithm has the ability of handling consecutive identical byte, for example " AAAAAA "; The second change length coding algorithm has the ability of handling the consecutive identical byte of interlocking, for example " ABABAB ".
The method of improving compression of images and transmission of the present invention according to different demands, determines required compression method, makes full use of the advantage of each compression method, effectively reduces the demand of image transfer rate and network bandwidth.
Description of drawings
Preferred embodiment of the present invention will be aided with following figure and do more detailed elaboration in comment backward, wherein:
Fig. 1 is existing tele-control system;
Fig. 2 is the schematic flow sheet of the main allocator of compression of images of the present invention;
Fig. 3 is the form that the pixel format conversion of the RGB555 of standard is separated into red (R), green (G) and blue (B);
Fig. 4 is the schematic flow sheet of fuzzy comparison program;
Fig. 5 is the coded program of preferred embodiment RLEDMP algorithm of the present invention;
Fig. 6 is the coded format of the described direct coding program of step 614 among Fig. 5;
Fig. 7 is the schematic flow sheet of the direct coding program of step 614 among Fig. 5;
Fig. 8 is the schematic flow sheet of the capable pattern pointer coding of step 604 among Fig. 5;
Fig. 9 is the switching circulation synoptic diagram of RLEDM coding mode; And
Figure 10 A to Figure 10 C is the schematic flow sheet of the RLEDM coding of step 610 among Fig. 5.
[figure number explanation]
102 computing machines 1～4 104 computer switch
106 A/D converters, 108 first in first out controllers
110 output load modules, 112 CPU (central processing unit)
114 Memory Controllers, 116 storeies
118 network cards, 120 controllers
122 networks, 124 client computers
202～1434 steps
Embodiment
By in the above-mentioned background of invention as can be known, the continuous images data, though compressed, under the finite element network frequency range, desire to carry out continuous images real-time Transmission and remote computer control, will be subject to the transmittability of network.If can more effectively compress the consecutive image data, just can alleviate the burden of network, and the transmission of realtime graphic and the time delays of remote computer control are significantly reduced.
Below will be with accompanying drawing and detailed description, clear spirit of the present invention described, as the person skilled in the art after understanding preferred embodiment of the present invention, when can be by the technology of teachings of the present invention, change and modification, it does not break away from spirit of the present invention and scope.The invention provides a kind of improved change length coding (Run Length Encoding) and three angle modulation compiling methods (DeltaModulation Encoding), carrying out the coding of computer picture, and the transmission of the consecutive image after compressing by network.See also Fig. 2 to Figure 10 C, in conjunction with the following description, so that spirit of the present invention and scope to be described in detail in detail.
We claim this coded program to be change length coding and position encoded (the RunLength Encoding and Delta Modulation Position of three angle modulations at this; RLEDMP).Wherein changing length coding and represent with RLE, is a kind of undistorted algorithm (Lossless Algorithm), and it represents, the data that repeats so can effectively reduce the quantity of data with source book and multiplicity.And three angle modulations are represented with DM, are a kind of difference signal coding (Differential PulseCode Modulation that belongs to; DPCM) technology.Use this kind method to have only a few-bit to be used for the place inequality of encoding, determine a numerical value earlier after, express other numerical value in the mode of utilizing increment and decrement.P represents position encoded (Position Coding), is used for writing down the data of capable pattern.
According to the requirement of Joint Photographic Experts Group, original image document at first is split into the matrix of a plurality of 8*8, and each matrix is represented 64 pixels.Because JPEG compresses, at first RGB to be converted to the form of brightness and colourity.Generally speaking, be converted into Y and represent brightness, Cb represents chroma blue, and Cr represents red color.Therefore our matrix that defines 16*16 is the requirement unit of a minimum.When the data that part is only arranged was inequality, it was unworthy using jpeg algorithm.In such cases, we will adopt the algorithm of above-mentioned RLEDMP to replace original jpeg algorithm.
It among Fig. 2 the schematic flow sheet of the main allocator of compression of images of the present invention.Utilize the automatic converting data of distribution function to enter specific flow process.At first, with the image cutting, to form the block of 16*16 pixel, step 202.Then, prepare data to blur the relatively program of (FuzzyComparison), step 204.More existing then block and previous block are to obtain whole discrepant quantities (Total Difference Number; TDN), step 206.According to relatively result, allocator carries out different compressions with image document again, and for example in the present embodiment, using predetermined difference number is 32, and the heteromerism of being on duty is not less than 1 but less than 32 o'clock, enter step 210, uses the algorithm of RLEDMP to compress.Be on duty heteromerism greater than 32 o'clock, then enter step 212, use the algorithm of JPEG to compress.Step 214 stores the data after encoding.Though in this embodiment, select for use 32 to be a selection threshold that uses JPEG or RLEDMP, the present invention does not limit this numerical value, and generally speaking, in the matrix of 16*16, the numerical value less than 128 all can use.The present invention utilizes a predetermined values of disparity to carry out the judgment criterion of the compression algorithm of each image block use, make each block according to the comparison of previous image, select the proper compression algorithm, compress, so can produce the compression of images kind algorithm of the most suitable this block, can effectively reduce the data size after the compression of images, thereby save image and transmit the needed time.
Fig. 3 is the form that the pixel format conversion of the RGB555 of standard is separated into red (R), green (G) and blue (B).Existing pixel format uses the form of RGB555, it is the pixel format 400 among the figure, for conveniently comparing, the present invention is separated into red pixel format 4 01, green pixel format 4 02 and blue pixel format 4 03, and each color pixel form uses 8 positions to carry out the storage of data, as shown in FIG., the data of each color is all less than 0x20, so definition 0x20 is termination character (End ofbuffer character; EOBC).
Fig. 4 is the schematic flow sheet of fuzzy comparison program.Because via the unsettled situation of A/D converter 106 generation data samplings, use when comparing at the predetermined numerical value of this definition earlier.This predetermined numerical value is with fuzzy comparison range (Fuzzy Comparison Range; FCR) represent.When the data that is sent by A/D converter 106, as 0x10,0x11 or 0x12 all can be regarded as identical data, the unsettled phenomenon when overcoming A/D converter 106 capturing images, that is to say when difference value smaller or equal to 2 the time, it can be considered as identical.Following formula one uses when fuzzy control is calculated for carrying out:
Result＝Abs(CR-FR)＜＝FCR.and.Abs(CG-FG)＜＝FCR.and.Abs(CB-FB)＜＝FCR
Above-mentioned CR, CG and CB represent the present R, the G that separate in the pixel of present image and the color data of B respectively, and FR, FG and FB represent the previous R, the G that separate in the pixel of previous image and the color data of B respectively.
When above-mentioned Result was zero, the pixel of representing previous image was not identical pixel with the pixel of present image, and it is identical pixels that other situation is represented these two pixels.
At first, step 502 is fuzzy initial setting value relatively, sets the predetermined value of FCR, and with linage-counter (Line Counter; LC), some counter (Dot Counter; DC), go pattern (Line Pattern; LP) and whole different quantity (Total Difference Number; TDN) all make zero earlier.Step 504 is obtained the image document of desiring movement images 1 and image 2.Step 506, whether the image document of every bit is identical in the movement images bit by bit, when the image document of finding this point when being identical, directly enters step 514, and when the image document of this point is also inequality, then enters the program of fuzzy comparison of the present invention.Step 508 with the image document of this point, is separated into the data of RGB color according to the separation method among Fig. 3, and the image document of this point of image 1 is divided into R1, B1 and G1, and the image document of the same position of image 2 is divided into R2, G2 and B2.Whether step 510, the image document that utilizes fuzzy relatively this point of formula one be in can received scope, and promptly whether the rgb value of this point can be regarded as identically, and is accepted as identical data.If be judged to be the identical step 514 that then directly enters, if be regarded as the step 512 that then enters inequality.Step 512 is set at 1 with the LP of this position, and the numerical value of TDN is added 1, represents whole different quantity to increase a bit again.Then enter step 514, with the position of LP, be moved to the left a position, and DC is added 1.Step 516 is whether the image document of judging these row relatively finishes, and is example at this with 16 picture point of every row, so DC judges whether more than or equal to 16.If relatively do not finish as yet, promptly DC gets back to step 504 and continues this more next location point less than 16, if relatively finish, promptly DC then enters step 518 more than or equal to 16.Step 518 is that the LP value that above-mentioned comparison is write down deposits LP impact damper (LP Buffer in; LPBUF) in, and LP and DC be reset to zero and LC is added 1, just enter in the next line, proceed the work of image blurring comparison,, finish the fuzzy of this image and relatively work up to step 520.
Use the program of image blurring comparison of the present invention, after relatively finishing, produced the data of one group of row pattern simultaneously, below will comparative approach of the present invention and comparative result be described in detail in detail with table one, table two and table three.The image document that table one representative is previous, the data of the capable pattern that image document that table two representative is present and table three representative are produced after comparing.
Fuzzy comparison program of the present invention is the pixel that utilizes each row in comparison sheet one and the table two and each row, to produce the data of the capable pattern in the table three.
Table one: the data of previous image.
| 0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0000 | ?0000 | ?0000 | ?0000 | ?0001 | ?0001 | ?0001 | ?0000 | ?0000 | ?0001 | ?0001 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0000 | ?0001 | ?0001 | ?0000 | ?0020 | ?0020 | ?0020 | ?0000 | ?0000 | ?0020 | ?0020 | ?0020 | ?0001 | ?0000 | ?0000 | ?0000 |
| 0000 | ?0020 | ?0020 | ?0000 | ?0021 | ?0021 | ?0021 | ?0000 | ?0000 | ?0021 | ?0021 | ?0021 | ?0020 | ?0000 | ?0000 | ?0000 |
| 0000 | ?0021 | ?0021 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0021 | ?0001 | ?0001 | ?0000 |
| 0000 | ?0000 | ?0000 | ?0001 | ?0400 | ?0400 | ?0400 | ?0001 | ?0001 | ?0001 | ?0400 | ?0400 | ?0000 | ?0020 | ?0020 | ?0001 |
| 0000 | ?0400 | ?0400 | ?0020 | ?0020 | ?0001 | ?0020 | ?0020 | ?0020 | ?0020 | ?0020 | ?0020 | ?0400 | ?0021 | ?0021 | ?0020 |
| 0000 | ?0020 | ?0020 | ?0021 | ?0001 | ?0020 | ?0001 | ?0021 | ?0021 | ?0021 | ?0001 | ?0001 | ?0020 | ?0000 | ?0000 | ?0021 |
| 0000 | ?0001 | ?0001 | ?0000 | ?0000 | ?0021 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0001 | ?0400 | ?0400 | ?0000 |
| 0000 | ?0020 | ?0000 | ?0400 | ?0000 | ?0000 | ?0020 | ?0400 | ?0400 | ?0400 | ?0000 | ?0000 | ?0000 | ?0020 | ?0020 | ?0400 |
| 0000 | ?0021 | ?0000 | ?0020 | ?0000 | ?0400 | ?0001 | ?0020 | ?0020 | ?0020 | ?0000 | ?0000 | ?0000 | ?0001 | ?0001 | ?0020 |
| 0000 | ?0000 | ?0000 | ?0001 | ?0000 | ?0020 | ?0000 | ?0001 | ?0001 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0001 |
| 0000 | ?0400 | ?0000 | ?0000 | ?0000 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0020 | ?0000 |
| 0000 | ?0020 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0001 | ?0000 |
| 0000 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0000 | ?0000 | ?0000 | 0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0001 | 0000 | 0000 | 0000 | ?0000 | 0000 | 7000 | ?7000 | ?0000 | ?0000 | ?0000 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0020 | 0000 | 0001 | 0000 | ?0000 | 0000 | 0000 | ?7000 | ?0000 | ?0001 | ?0000 | ?0020 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0021 | 0000 | 0020 | 0000 | ?0000 | 0000 | 0001 | ?7000 | ?0001 | ?0020 | ?0001 | ?0021 | ?0001 | ?0001 | ?0001 | ?0001 |
| 0000 | 0001 | 0021 | 0000 | ?0001 | 0000 | 0020 | ?7000 | ?0020 | ?0021 | ?0020 | ?0000 | ?0020 | ?0020 | ?0020 | ?0020 |
| 0400 | 0020 | 0000 | 0000 | ?0020 | 0001 | 0021 | ?7000 | ?0021 | ?0000 | ?0021 | ?0400 | ?0021 | ?0021 | ?0021 | ?0021 |
| 0020 | 0021 | 0400 | 0000 | ?0021 | 0020 | 0000 | ?7000 | ?0000 | ?0400 | ?0000 | ?0020 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0001 | 0000 | 0020 | 0001 | ?0000 | 0021 | 0400 | ?7000 | ?0001 | ?0020 | ?0400 | ?0001 | ?0400 | ?0400 | ?0400 | ?0400 |
| 0000 | 0400 | 0001 | 0020 | ?0400 | 0000 | 0020 | ?7000 | ?0020 | ?0001 | ?0001 | ?0020 | ?0020 | ?0020 | ?0020 | ?0020 |
| 0000 | 0020 | 0000 | 0021 | ?0001 | 0001 | 0001 | ?7001 | ?0021 | ?0001 | ?0020 | ?0021 | ?0001 | ?0001 | ?0001 | ?0001 |
| 0000 | 0001 | 0000 | 0000 | ?0020 | 0020 | 0000 | ?7000 | ?0000 | ?0020 | ?0021 | ?0000 | ?0020 | ?0000 | ?0000 | ?0020 |
| 0000 | 0000 | 0000 | 0400 | ?0021 | 0021 | 0400 | ?7000 | ?0400 | ?0021 | ?0000 | ?0400 | ?0021 | ?0000 | ?0000 | ?0021 |
| 0000 | 0000 | 0000 | 0020 | ?0000 | 0000 | 0020 | ?7000 | ?0020 | ?0000 | ?0400 | ?0020 | ?0000 | ?0000 | ?0000 | ?0000 |
| 0000 | 0000 | 0000 | 0001 | ?0400 | 0400 | 0001 | ?7000 | ?0001 | ?0400 | ?0020 | ?0001 | ?0400 | ?0000 | ?0000 | ?0400 |
| 0000 | 0000 | 0000 | 0000 | ?0020 | 0020 | 0000 | ?7000 | ?0000 | ?0020 | ?0001 | ?0000 | ?0020 | ?0000 | ?0000 | ?0020 |
| 0000 | 0000 | 0000 | 0000 | ?0001 | 0001 | 0000 | ?7000 | ?0000 | ?0001 | ?0000 | ?0000 | ?0001 | ?0000 | ?0000 | ?0001 |
| 0000 | 0000 | 0000 | 0000 | ?0000 | 0000 | 7001 | ?7000 | ?0001 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 | ?0000 |
Table two: the data of present image.
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
| ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??1 | ??1 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 | ??0 |
Table three: the data of the capable pattern that after comparing, is produced.
After relatively, write the data among the LPBUF:
0x0300，0x0100，0x0100，0x0100，0x0100，0x0100，0x0100，0x0100，
0x0100，0x0100，0x0100，0x0100，0x0100，0x0100，0x0100，0x0300，
Be the data of the capable pattern that launches in the table three.
Because whole different quantity are 18 points, so according to judgment mode illustrated in fig. 2 and predetermined TDN threshold, above-mentioned image comparative result will adopt the coded system of RLEDMP to carry out the compress mode of this image block.
Fig. 5 illustrates the coded program of preferred embodiment RLEDMP algorithm of the present invention.Step 602 is at first judged the quantity of TDN, if the quantity of TDN directly enters step 614 less than 4, otherwise if the quantity of TDN enters next procedure 604 more than or equal to 4.Step 604 is to step 606, is P method (P Method) what this claimed, and wherein step 604 is gone the coding of pattern pointer (Line Pattern Index), and step 606 is then carried out the coding of LPBUF.In next procedure 608, the pixel information that LP is indicated variant image slices vegetarian refreshments is separated into the color data of R, G and B.Following step 610 is then carried out the coding of RLEDM.Whether the size after relatively encoding again is greater than TDN*3.If the size behind the coding is then carried out the program of the direct coding of step 614 greater than TDN*3, but if the size behind the coding, is then used the data after RLEDM encodes smaller or equal to TDN*3.
For more clearly demonstrating method of the present invention, please refer to Fig. 6, Fig. 6 is the coded format of the described direct coding program of step 614 among Fig. 5.As shown in FIG., the character of front is used for storing the data of row 702, and the character that is right after is used for writing down remaining data of recording pixel 706 then of data of row 704.
Fig. 7 is the schematic flow sheet according to the direct coding program of step 614 among preferred embodiment Fig. 5 of the present invention.Step 802 is for reading initial data, with the address (LPBAddress of row pattern impact damper; LPBA) with frame buffer address (Image Buffer Address; IBA) read in.Step 804, the initial value of setting linage-counter are zero.Step 806 reads the data of LP according to LPBA, and LPBA adds 2, sets linage-counter (Column Counter simultaneously; CC) be zero.Step 808 judges that whether LP is zero, if be zero directly to enter step 818, if non-vanishing, then enters step 810, and the position data of judging most LP whether zero is if zero enter step 816, as if the non-vanishing step 812 that enters.Step 812 writes present column counter and column counter coding in the output string (Output Stream).Step 814 writes required pixel data in the output string.Step 816 adds 1 with LP to move to left a position and CC.Flow process returns step 808, when LP=0, enters step 818, and LC is added 1.Whether step 820 judges LC more than or equal to 16, if get back to step 806 less than 16, if more than or equal to 16, then finishes this program.
The RLEDMP algorithm is converted into linear data with nonlinear data, and affords this responsibility by the P method.The P method is at first set up 16 capable pattern pointer (Line PatternIndex).A capable pattern is represented in each position, and when the row pattern is non-vanishing, corresponding position will be set to 1, comprise data in this journey to represent.And the P method and the pattern buffer of will going compress.
Fig. 8 is the schematic flow sheet according to the capable pattern pointer coding of step 604 among preferred embodiment Fig. 5 of the present invention.At first, step 902 reads capable pattern buffer address.Step 904 will be gone pattern pointer (Line Pattern Index; LPI) and the initial value of a counter be set at zero.Step 906 reads the data of the LP among the LPB.Step 908 judges whether the data of LP is zero.When it is non-vanishing, enter step 910, the significance bit of setting the minimum of row pattern pointer is 1.When the data of LP is zero or after step 910 finishes, enter step 912, row pattern pointer is to the character that moves to left, and counter and row pattern buffer address all increase.Step 916 is if counter less than 16, is then got back to step 906.After counter accumulative total is to 16, enter step 918, will go pattern pointer and whole different quantity storages.
Below will describe the capable pattern buffer coding method of step 606 among preferred embodiment Fig. 5 of the present invention in detail.Because row pattern pointer has comprised the data of row, only need handle these LP in this P method and be not equal to zero situation.The P method is divided into four nibbles (Nibble) with each row pattern.When if nibble is non-vanishing, with the content of carry-out bit 1 and nibble to output string.
| 1 | ??X | ??X | ??X | ??X |
When if nibble is zero, when still remaining nibble was also non-vanishing, carry-out bit 0 and position 1 were to output string.
| 0 | 1 |
If when remaining nibble is zero, export two positions 0 to output string.
| 0 | ????0 |
The P method is also encoded the positional information of present block, so can carry out the processing of these separate datas.The RLEDMP algorithm is assembled these data by row pattern data, and tests the position of each row pattern impact damper.If find the position of non-zero, will obtain pixel information in the image buffer according to current position.Pixel information in image buffer is separated into the data of R, G and B at once by RGB555.Whole R data can be stored in the RGB sampling buffer (RGB Sample Buffer), whole G data then can be stored in the RGB sampling buffer (RGB Sample Buffer) and according to the TDN displacement, same, whole B data then can be stored in the RGB sampling buffer (RGB Sample Buffer) and according to the TDN*2 displacement.
For example, following image document is handled:
| P1 | P2 | P3 | … | |
| P4 | P5 | … | ||
| P6 | … | |||
| P7 | P8 | … | ||
| … | … | … | … | … |
Wherein, P1～P8 uses the form of RGB555, comprises R respectively
P1G
P1B
P1～R
P8G
P8B
P8And its value is all less than 0x20.
Above-mentioned block has 8 pixels inequality altogether, after its separation and conversion, can get following linear data:
R
P1R
P2R
P3R
P4R
P5R
P6R
P7R
P8G
P1G
P2G
P3G
P4G
P5G
P6G
P7G
P8B
P1B
P2B
P3B
P4B
P5B
P6B
P7B
P8
With following linear data is example:
0x08，0x09，0x09，0x08，0x08，0x08，0x09，0x08
After the RLE compression, the data buffer after the compression only need use 5 bytes (40), writes down these data, and is as follows:
0x40，0x49，0x42，0x48，0x40
If use the algorithm of DM to compress, then can get following data:
0x42，0x45，0x00，0xD8
As mentioned above, after the algorithm of use DM compresses, only need 4 bytes (actual is 31), can store and transmit these data, so significantly explanation is combined the advantage that is produced with RLE with the algorithm of DM.
As shown in Figure 9, the switching for RLEDM coding mode of the present invention circulates.The method of improving compression of images and transmission of the present invention, will be according to the minimum code principle, after the data of receiving, RLEDM of the present invention will switch according to received character string forms, at EM_NORMAL1101, switch between EM_PRE_DM 1102 and the EM_DM 1103.
RLE algorithm of the present invention also comprises other two kinds of treatment mechanisms.This treatment mechanism of two kinds is respectively, a RLE encryption algorithm, this claim for RLE01, it is handled at the identical characters string format, for example: " AAAAAA "; And the 2nd RLE encryption algorithm, this claim for RLE02, it is handled at staggered identical string format, for example: " ABABAB ".
The RLE01 Methods for Coding is used in following table explanation, when the quantity of multiplicity less than 5 the time, RLE01 utilizes following coding method to encode.
| Data (Base Value) | Flag (Flag) | Quantity (Unit Amount) | |||||
| X | X | X | X | X | 1 | N | ?N |
When source book is: during 0x06,
Data after compression is: 6 (5), 1 (1), 0 (2).
When source book is: 0x06,0x06,0x06, during 0x06,
Data after compression is: 6 (5), 1 (1), 3 (2).
Because multiplicity can't be less than zero, so in amount field, with 0 representative once, increase in regular turn, and flag value is 1 expression compressed format that data is then inserted with source book.
But when multiplicity greater than four times and during less than 129 times, the RLE01 Methods for Coding utilizes following method to encode.
| Data (Base Value) | Flag (Flag) | Quantity (Unit Amount) | |||||||||||
| X | X | X | X | ? | 0 | ??0 | N | ?N | ?N | ?N | ?N | ?N | ?N |
When source book is: 0x06,0x06,0x06,0x06,0x06,0x06, during 0x06,
Data after compression is: 6 (5), 00 (2), 6 (7).
Though utilizing quantity at this is that 7 positions are example, it can use in the situation of any amount greater than 2 positions certainly.
Next carry out the explanation of RLE02 coding method again.
| Data (Base Value) | Flag (Flag) | Data (Base Vaue) | Quantity (Unit Amount) | ||||||||||||||||||
| A | ?A | ?A | ?A | ?A | ?0 | ?1 | ?1 | ?0 | ?0 | ?B | ?B | B | ?B | B | ?N | ?N | ?N | ?N | ?N | ?N | ?N |
When source book is: 0x11,0x06,0x11,0x06,0x11, during 0x06,
Data after compression is: 11 (5), 01 (2), 100 (3), 6 (5), 2 (7).
Though usage quantity is 7 positions is example, it can use in any amount of position situation.
If the variation range of data 3 between-3 in, the DM coding method will be used following coding method.
| Data difference (Value Difference) | Data after the compression (Compressed Code) |
| 0 | 000 |
| 1 | 001 |
| 2 | 010 |
| 3 | 011 |
| -1 | 101 |
| -2 | 110 |
| -3 | 111 |
| EOF | 100 |
| Data (Base Value) | Flag (F1ag) | Data after the compression | Finish flag (End Flag) | ||||||||
| A | ?A | ?A | ?A | ?A | 0 | ????1 | ????0 | … | ????1 | ????0 | ????0 |
When source book is: 0x06,0x07,0x05,0x08, during 0x05,
Data after compression is: 6 (5), 010 (3), 001 (3), 110 (3), 011 (3), 111 (3), 100 (3).
Consult Figure 10 A, Figure 10 B and Figure 10 C, it is the schematic flow sheet according to the RLEDM coding of step 610 among preferred embodiment Fig. 5 of the present invention.Figure 10 A mainly describes the setting of initial state, RLE01 coding and process ends.
Step 1202, initial conditions are set, and read the start address (RSBA) of RGB Sample Buffer, calculate and the end address (RGBEA) of RGB Sample Buffer, for RSBA adds TDN*3, and with termination character (End of Buffer Character; EOBC) be added in after the RGB Sample Buffer.The preset value of setting coding mode (Encode Mode) is Normal, and the coding mode of the EM Normal among Fig. 9 is used in its representative, and sets previous three an angle modulation difference value (Previous Delta-modulation Difference Value; PrevDM) be 0.At last, read the byte of RGB Sample Buffer to set previous character (PreviousCharacter; PrevCH) initial value, and with RSBA increase by 1.
Step 1204 judges that whether RSBA is more than or equal to RSBEA.As RSBA during more than or equal to RSBEA, connect step 1206, judge again whether Encode Mode equals EM PRE DM, if equate, then enter step 1208, send 01 to output string.Step 1212, three positions of output PrevDM with step 1214, send 100 to output string to output string.Then, the data that will encode in the buffer of step 1216 is removed.If but in the step 1206, when Encode Mode and EM_PRE_DM are also inequality, then enter step 1210, whether identical to judge Encode Mode with EM_DM, if the identical step 1212 that enters, if but it is also inequality, then enter step 1216.
Wherein in the step 1204, if RSBA during less than RSBEA, then enters step 1218, with multiplicity (Repeat Counter; RC) be set at 1.This flow process will determine the number of times of the repetition of a previous character (PrevCh).Step 1220, existing character (Current Character; Ch), and RSBA is added 1 by reading among the RGB Sample Buffer.Step 1222 judges whether Ch is identical with PrevCh, if the identical step 1224 that enters adds 1 with RC, is used for calculating continuous identical character quantity, and returns step 1220.If Ch and PrevCh are also inequality, then enter step 1226, judge whether RC equals 1.If equal 1, then enter the step 1302 among Figure 10 B, and if be not equal to 1, then this flow process will be according to following explanation mode, which coding flow process decision uses comparatively suitable.
Which kind of coded system how this flow process to determine to use carry out the coding of the data of these repetitions.At first, step 1230 judges whether Encode Mode equals EM_NORMAL.If Encode Mode equals EM_NORMAL, enter step 1254, judge that whether RC is smaller or equal to 4.If RC smaller or equal to 4 then enter step 1256, subtracts 1 with RC.Step 1258, with the value of PrevCh to moving to left three positions and it is added the value of RC.Step 1260 sends 8 PrevCh to output string.Step 1270 equals the PrevCh setting to Ch, to follow the trail of next continuation character.Flow process returns in the step 1204.If in the step 1254, RC then enters step 1262 greater than 4.Step 1262 subtracts 1 with RC.Step 1264 then, with PrevCh to two positions that move to left.Step 1266 is sent the PrevCh of seven positions to output string.Step 1268 is sent the RC of seven positions to output string.Then, step 1270 equals the PrevCh setting to Ch, and flow process is also got back in the step 1204.
And in the step 1230 wherein, when being not equal to EM_NORMAL, enter step 1232 as if Encode Mode, judge that whether RC is greater than seven.If RC greater than seven, then enters step 1252, send 100 to output string, and to reset Encode Mode be EM_NORMAL.And when RC is not more than seven, then enter step 1234.Step 1234, whether relatively PrevCh deducts the absolute value of Ch, greater than three angle modulation disparity range (DeltaModulation Difference Range; DMRANGE).If absolute value (PrevCh-Ch) is during greater than DMRANGE, and the RC in the step 1236 is more than or equal to 3 o'clock, and flow process is got back to step 1252.And when the RC in the step 1236 less than 3 the time, enter step 1238, judge whether EncodeMode equals EM_PRE_DM.If Encode Mode equals EM_PRE_DM, enter step 1272 and send to output string, and enter step 1274 10, three positions, 000 and 100 that this flow process is sent PrevDM and are set PrevCh and are equaled Ch to output string.And when existing coding mode is not EM_PRE_DM, directly enter step 1274.When the absolute value in the step 1234 (PrevCh-Ch) when being not more than DMRANGE, this flow process enters step 1240, judges whether existing coding mode is EM_PRE_DM.When existing coding mode is not during for EM_PRE_DM, enter step 1244, to output string, and RC subtracts 1 with three positions sending PrevDM.And when existing coding mode is EM_PRE_DM, enter step 1242, send 10 to output string, and to set Encode Mode be EM_DM, and then enter step 1244.After step 1244, flow process enters step 1246, judges whether RC equals zero.When RC is not when equalling zero, enter step 1248, this flow process sends 000 to output string with repeating, and RC subtracted 1, when RC equals zero, enter step 1250, PrevDM is set at Ch and deducts PrevCh, and PrevCh set again equals Ch, and then flow process is got back in the step 1204.
Figure 10 B mainly illustrates the RLE02 coding.At first, step 1302 stores the existing position RSBA of input of character string, then enters step 1304, reads two characters by RSBA, and first character is written as next character (Next Character; NextCh), and with RSBA add 1.Step 1306 sends second character to compare string (Comparison Character; CCh), and with RSBA add 1 again.Step 1308 judges whether PrevCh equals NextCh, if PrevCh equals NextCh, and step 1310, judge whether Ch equals CCh.If arbitrary invalid situation in step 1308 or the step 1310 enters step 1312, RSBA is subtracted two.If enter step 1314 under the situation that both all set up, be that RC is set at two.Step 1316,1318,1322,1324 and 1326, more next two of RSBA characters and PrevCh and Ch are to calculate the character of continuous repetition.After calculating was finished, if when having any difference to take place, flow process entered step 1332.Step 1332, relatively whether RC is greater than 2, if RC greater than 2, next enters step 1338, whether more present coding mode is EM_NORMAL, if this flow process will be sent the rule of relevant data according to RLE02 to output string.In step 1332, RC is not more than at 2 o'clock, enters step 1334, and relatively whether absolute value (PrevCh-Ch) is greater than DMRANGE.When absolute value (PrevCh-Ch) during greater than DMRANGE, enter step 1336, store the position of input of character string, and enter among Figure 10 C.If in the step 1338, when present coding mode is not EM_NORMAL, send 100 to output string, step 1340.
According to the RLE02 coding rule, at first enter step 1342, relatively whether RC equals zero, if when RC equals zero, enters step 1360.Step 1360 by obtaining data in the input of character string, and is stored to PrevCh, and RSBA being added 1 is EM_NORMAL with setting present coding mode again.In step 1342, when RC is not equal to zero, and step 1344, RC set counter and equals 127 greater than 128 o'clock.In step 1344, RC is not more than at 128 o'clock, and counter equals RC and subtracts 1.Then enter step 1350,1352,1354 and 1356, send five positions of five positions, 10 and 100, Ch of PrevCh and seven positions of counter, to output string.Step 1358 is set at RC with RC and deducts counter and deduct 1, and gets back to step 1342, equals zero up to RC.
Figure 10 C mainly illustrates the coding of DM.At first, step 1402, relatively whether absolute value (PrevCh-Ch) is greater than DMRANGE.When absolute value (PrevCh-Ch) when being not more than DMRANGE, enter step 1404, whether more present coding mode is EM_NORMAL.When present coding mode is EM_NORMAL, enter step 1406, Encode Mode is set at EM_PRE_DM.Step 1408, five positions sending PrevCh are to output string, and step 1418 is set at Ch with PrevDM and deducts PrevCh, and step 1434, and PrevCh is set at Ch.This flow process is got back to the step 1204 among Figure 10 A then.
If in the step 1404, when present coding mode is not EM_NORMAL, enter step 1410, whether more present coding mode is EM_PRE_DM.If, enter step 1412, Encode Mode is set at DM, step 1414 and step 1416, send 10 with three positions of PrevDM to output string.If in the step 1410, when present coding mode was not EM_PRE_DM, flow process directly entered step 1416, then entered step 1418.
If in the step 1402, absolute value (PrevCh-Ch) enters step 1418 during greater than DMRANGE, and whether more present coding mode is EM_PRE_DM.If, enter step 1420, step 1422 and step 1424, send 10, three positions of PrevDM and 100 are to output string.Enter step 1432, Encode Mode is set at EM_NORMAL, then flow process enters step 1434.If in the step 1418, when present coding mode was not EM_PRE_DM, flow process entered step 1426, and whether more present coding mode is EM_DM.If EM_DM enters step 1422.If not EM_DM, enter step 1428, with three positions of value displacement of PrevCh and add 100, enter step 1430 then, eight positions of PrevCh are sent to output string.Flow process then enters step 1432.
Tele-control system as shown in Figure 1 when using the method for improving compression of images and transmission of the present invention, has many advantages.If the image pixel that tendency to develop send a computer picture to have 800*600, and each pixel is used 16 positions.Suppose the ratio of compression that the use jpeg algorithm had 15: 1, image document after compression is about 64,000 character boundaries.If there are 2000 pixels to be dispensed in the block of 200 16*16,, need 6726 bytes approximately if only use jpeg algorithm.If when using RLEDMP, even only utilize the direct coding method to compress, size of data after compression only is 3000 bytes approximately.If the compressibility of RLEDMP is, then only need 1500 bytes at 2: 1 o'clock.So use the method for improving compression of images and transmission of the present invention, can significantly reduce demand for image signal transmission speed and network bandwidth, making on network effectively, the transmission compressed image becomes more efficient.
The invention provides a kind of algorithm, a kind of feasible pattern in conjunction with two kinds of codings more is provided in conjunction with change length coding and three angle modulations coding.The method of improving compression of images and transmission of the present invention can determine required compression method according to different demands, makes full use of the advantage of each compression method, effectively reduces the demand of image transfer rate and network bandwidth.
Understand as the person skilled in the art, the above only is preferred embodiment of the present invention, is not in order to limit protection scope of the present invention; All other do not break away from the equivalence finished under the disclosed spirit and changes or modify, and all should be included within the scope that claim of the present invention protects.
Claims (26)
1. an image information coding method is used in the consecutive image transmission, has the function that reduces coding back image document size, it is characterized in that this image information coding method comprises at least:
(a) read Fig. 1 picture and Fig. 2 picture;
(b) use fuzzy relative method, analyze this Fig. 1 picture and this Fig. 2 picture;
(c), determine the encryption algorithm of this Fig. 2 picture according to analysis result; And
(d) carry out the coding of this Fig. 2 picture.
2. image information coding method as claimed in claim 1 is characterized in that: above-mentioned step (a) comprises at least:
(a1) with this Fig. 1 picture and this Fig. 2 picture, divide into a plurality of blocks, wherein each this block also comprises a plurality of pixels; And
(a2) read the image document of these pixels of this Fig. 1 picture and this Fig. 2 picture in regular turn.
3. image information coding method as claimed in claim 2 is characterized in that, above-mentioned fuzzy relative method comprises:
(b1) set a fuzzy comparison range;
(b2) separate this Fig. 1 picture and this Fig. 2 image document as corresponding pixel, making becomes the first red data, the first green data and the first blue data, and the second red data, the second green data and the second blue data;
(b3) deduct the absolute value of this second red data smaller or equal to this fuzzy comparison range when this first red data, this first green data deducts the absolute value of this second green data smaller or equal to this fuzzy comparison range, and the absolute value that this first blue data deducts this second blue data is considered as identical with the image document of this pixel of this Fig. 1 picture this Fig. 2 picture during smaller or equal to this fuzzy comparison range; And
(b4) deduct the absolute value of this second red data when this first red data, this first green data deducts the absolute value of this second green data, and/or the absolute value that this first blue data deducts this second blue data is during greater than this fuzzy comparison range, is considered as the image document of this Fig. 2 picture and this pixel of this Fig. 1 picture inequality.
4. image information coding method as claimed in claim 3 is characterized in that: above-mentioned step (b1) comprises:
(b1.1) directly relatively this Fig. 1 picture and this Fig. 2 obtain a comparative result as the image document of corresponding pixel;
(b1.2) when this comparative result when being identical, the image document of this pixel is considered as identical; And
(b1.3) when this comparative result when being inequality, execution in step (b1) is to (b4).
5. image information coding method as claimed in claim 3 is characterized in that: above-mentioned step (b) also comprises:
(b5) image document that writes down this pixel is considered as identical position;
(b6) image document that writes down this pixel is considered as position inequality, and writes down the image document of this pixel; And
(b7) store a delegation's pattern pointer and a different image document, wherein on behalf of the image document of this pixel, this row pattern pointer be considered as identical position with 0, represent the image document of this pixel to be considered as position inequality with 1, this different image document, the image document that writes down this pixel is considered as the image document of position inequality.
6. image information coding method as claimed in claim 5 is characterized in that: above-mentioned step (c) also comprises:
(c1) add up these and be considered as position inequality;
(c2) relatively these are considered as the quantity and a predetermined threshold of position inequality;
(c3) when this quantity during, use the image document coding of jpeg algorithm with this block of this Fig. 2 picture more than or equal to this predetermined threshold; And
(c4) when this quantity during, use change length and three angle modulation encryption algorithms that the different image document of this row pattern pointer and this is encoded less than this predetermined threshold.
7. image information coding method as claimed in claim 6 is characterized in that: above-mentioned step (c4) also comprises:
This different image document is separated into the image document of three kinds of colors of RGB; And
Rearrange the image document of these red bluish-green three kinds of colors, the red image data is arranged in together, the blue image data is arranged in together, the green image data is arranged in together, and form a linear data, wherein this linear data has a plurality of bytes, represents these red image data, blue image data and green image data respectively.
8. image information coding method as claimed in claim 7 is characterized in that: above-mentioned change length and three angle modulation encryption algorithms comprise:
The direct coding algorithm uses during less than direct coded gates sill in this quantity;
Three angle modulation encryption algorithms, less than this predetermined threshold but more than or equal to this direct coding threshold, and the variable quantity of the byte of this linear data uses during smaller or equal to the scope of a predetermined change in this quantity; And
Change length coding algorithm, less than this predetermined threshold but more than or equal to this direct coding threshold, and the variable quantity of the byte of this linear data uses during greater than the scope of a predetermined change in this quantity,
Wherein when the identical image data, the length of data behind the coding that this three angle modulations encryption algorithm of process and this change length coding algorithm are produced as if the length of data behind the coding that is produced greater than this direct coding algorithm, then uses the direct coding algorithm to encode.
9. image information coding method as claimed in claim 8 is characterized in that: above-mentioned direct coding threshold is 4.
10. image information coding method as claimed in claim 8 is characterized in that: the scope of above-mentioned predetermined change is-3 to+3.
11. image information coding method as claimed in claim 8 is characterized in that: above-mentioned change length coding algorithm also comprises, and the first change length coding algorithm has the ability of handling consecutive identical byte.
12. image information coding method as claimed in claim 8 is characterized in that: above-mentioned change length coding algorithm also comprises, and the second change length coding algorithm has the ability of handling the consecutive identical byte of interlocking.
13. image information coding method as claimed in claim 6 is characterized in that: above-mentioned predetermined threshold is less than 128, when the pixel of each block is 16*16.
14. image information coding method as claimed in claim 9 is characterized in that: above-mentioned predetermined threshold is 32.
15. image information coding method as claimed in claim 3 is characterized in that: the preset value of above-mentioned fuzzy comparison range is 2.
16. image information coding method as claimed in claim 1 is characterized in that: above-mentioned image information coding method is at computer switch (Keyboard-Video-Mouse Switch; Image transmission during KVM Switch) via network, management remote computer is gone up and is used, and the required network bandwidth of this image transmission is reduced.
17. an image information coding method is at a computer switch (Keyboard-Video-Mouse Switch; Consecutive image transmission during KVM Switch) via network, management remote computer is gone up and is used, has the function that reduces coding back image document size and reduce the required network bandwidth of this image transmission, it is characterized in that this image information coding method comprises at least:
The Fig. 1 that reads same position as block and Fig. 2 as block;
Use fuzzy relative method, analyze this Fig. 1 as block and this Fig. 2 as block;
According to analysis result, determine the encryption algorithm of this Fig. 2 picture; And
Carry out the coding of this Fig. 2 picture.
18. image information coding method as claimed in claim 17 is characterized in that: above-mentioned fuzzy relative method comprises:
Set a fuzzy comparison range;
Directly relatively this Fig. 1 as block and this Fig. 2 as the image document of block, obtain a comparative result;
When this comparative result when being identical, the image document of this pixel is considered as identical; And
When this comparative result when being inequality,
Separate this Fig. 1 picture and this Fig. 2 image document as corresponding pixel, making becomes the first red data, the first green data and the first blue data, reaches the second red data, the second green data second blue data;
When this first red data deducts the absolute value of this second red data smaller or equal to this fuzzy comparison range, this first green data deducts the absolute value of this second green data smaller or equal to this fuzzy comparison range, and the absolute value that this first blue data deducts this second blue data is considered as identical with the image document of this pixel of this Fig. 1 picture this Fig. 2 picture during smaller or equal to this fuzzy comparison range; And
Deduct the absolute value of this second red data when this first red data, this first green data deducts the absolute value of this second green data, and/or the absolute value that this first blue data deducts this second blue data is during greater than this fuzzy comparison range, is considered as the image document of this Fig. 2 picture and this pixel of this Fig. 1 picture inequality.
19. image information coding method as claimed in claim 18 is characterized in that: the preset value of above-mentioned fuzzy comparison range is 2.
20. image information coding method as claimed in claim 18 is characterized in that: above-mentioned fuzzy relative method also comprises:
The image document that writes down this pixel is considered as identical position;
The image document that writes down this pixel is considered as position inequality, and writes down the image document of this pixel; And
Store a delegation's pattern pointer and a different image document, wherein on behalf of the image document of this pixel, this row pattern pointer be considered as identical position with 0, represent the image document of this pixel to be considered as position inequality with 1, this different image document, the image document that writes down this pixel is considered as the image document of position inequality.
21. image information coding method as claimed in claim 20 is characterized in that: above-mentioned according to analysis result, determine the encryption algorithm of this Fig. 2 picture also to comprise:
Add up these and be considered as position inequality;
Relatively these are considered as the quantity and a predetermined threshold of position inequality;
When this quantity during, use the image document coding of jpeg algorithm with this block of this Fig. 2 picture more than or equal to this predetermined threshold; And
When this quantity during, use change length and three angle modulation encryption algorithms that the different image document of this row pattern pointer and this is encoded less than this predetermined threshold.
22. image information coding method as claimed in claim 21 is characterized in that: above-mentioned change length and three angle modulation encryption algorithms comprise:
This different image document is separated into the image document of three kinds of colors of RGB; And
Rearrange the image document of red bluish-green three kinds of colors, the red image data is arranged in together, the blue image data is arranged in together, the green image data is arranged in together, and form a linear data, wherein this linear data has a plurality of bytes, represents these red image data, blue image data and green image data respectively.
23. image information coding method as claimed in claim 22 is characterized in that: above-mentioned change length and three angle modulation encryption algorithms also comprise:
The direct coding algorithm uses less than 4 o'clock in this quantity;
Three angle modulation encryption algorithms, less than 32 but more than or equal to 4, and the variable quantity of these bytes of this linear data uses during smaller or equal to the scope of a predetermined change in this quantity; And
Change length coding algorithm is used in this quantity less than 32 but more than or equal to 4, and the variable quantity of the byte of this linear data uses during greater than the scope of a predetermined change,
Wherein when the identical image data, the length of data behind the coding that this three angle modulations encryption algorithm of process and this change length coding algorithm are produced as if the length of data behind the coding that is produced greater than this direct coding algorithm, then uses the direct coding algorithm to encode.
24. image information coding method as claimed in claim 23 is characterized in that: the scope of above-mentioned predetermined change is-3 to+3.
25. image information coding method as claimed in claim 23 is characterized in that: above-mentioned change length coding algorithm also comprises, and the first change length coding algorithm has the ability of handling consecutive identical byte.
26. image information coding method as claimed in claim 23 is characterized in that: above-mentioned change length coding algorithm also comprises, and the second change length coding algorithm has the ability of handling the consecutive identical byte of interlocking.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021298947A CN1295652C (en) | 2002-08-20 | 2002-08-20 | Image information coding method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB021298947A CN1295652C (en) | 2002-08-20 | 2002-08-20 | Image information coding method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1477601A true CN1477601A (en) | 2004-02-25 |
| CN1295652C CN1295652C (en) | 2007-01-17 |
Family
ID=34144322
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB021298947A Expired - Lifetime CN1295652C (en) | 2002-08-20 | 2002-08-20 | Image information coding method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1295652C (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101036151B (en) * | 2004-10-08 | 2010-05-26 | 辉达公司 | Methods and systems for rate control in image compression |
| CN101620842B (en) * | 2008-06-30 | 2014-04-16 | 瑞萨电子株式会社 | Image processing circuit, and display panel driver and display device mounting the circuit |
| CN104217445A (en) * | 2013-05-31 | 2014-12-17 | 精英电脑(苏州工业园区)有限公司 | A method for distortionless coding and decoding of desktop images of computers |
| TWI466095B (en) * | 2010-03-31 | 2014-12-21 | Apple Inc | Reduced-power communications within an electronic display |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5883976A (en) * | 1994-12-28 | 1999-03-16 | Canon Kabushiki Kaisha | Selectively utilizing multiple encoding methods |
| EP0817121A3 (en) * | 1996-06-06 | 1999-12-22 | Matsushita Electric Industrial Co., Ltd. | Image coding method and system |
| FR2782879B1 (en) * | 1998-08-28 | 2000-11-03 | Thomson Multimedia Sa | IMAGE COMPRESSION PROCESS |
-
2002
- 2002-08-20 CN CNB021298947A patent/CN1295652C/en not_active Expired - Lifetime
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101036151B (en) * | 2004-10-08 | 2010-05-26 | 辉达公司 | Methods and systems for rate control in image compression |
| CN101620842B (en) * | 2008-06-30 | 2014-04-16 | 瑞萨电子株式会社 | Image processing circuit, and display panel driver and display device mounting the circuit |
| TWI466095B (en) * | 2010-03-31 | 2014-12-21 | Apple Inc | Reduced-power communications within an electronic display |
| CN104217445A (en) * | 2013-05-31 | 2014-12-17 | 精英电脑(苏州工业园区)有限公司 | A method for distortionless coding and decoding of desktop images of computers |
| CN104217445B (en) * | 2013-05-31 | 2017-05-17 | 精英电脑(苏州工业园区)有限公司 | A method for distortionless coding and decoding of desktop images of computers |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1295652C (en) | 2007-01-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1271926B1 (en) | Image processing method, apparatus and computer program for compression-encoding | |
| CN1071526C (en) | Adaptive variable-length coding and decoding methods for image data | |
| CN1198466C (en) | Method for producing qunantization matrix | |
| CN1303820C (en) | Quality based image compression | |
| US11074723B2 (en) | Lossless compression of fragmented image data | |
| AU2012332327A1 (en) | Method and apparatus for image compression storing encoding parameters in 2D matrices | |
| CN103581678B (en) | To improve the method and system of decoder capabilities by using multiple decoder channels | |
| CN1550104A (en) | DCT compression using Golomb-Rice coding | |
| US8351693B2 (en) | Image compression apparatus and computer-readable recording medium recorded with image compression program | |
| RU2567988C2 (en) | Encoder, method of encoding data, decoder, method of decoding data, system for transmitting data, method of transmitting data and programme product | |
| CN101420614A (en) | Method for compressing image and device that a kind of hybrid coding and dictionary encoding are integrated | |
| CN103841424A (en) | System And Method For Randomly Accessing Compressed Data From Memory | |
| CN1295652C (en) | Image information coding method | |
| US7046387B1 (en) | Efficient coding of color images including monochrome images | |
| CN1129870C (en) | Method and device for estimating motion in digitized image with pixels | |
| JP2003209698A (en) | Image processing device and method therefor, computer program, and storage medium | |
| US20020001414A1 (en) | System for building a data compression encoder | |
| CN101998154B (en) | Multi-computer switcher for transmitting image signals to a plurality of far-end control computers and method thereof | |
| CN1280745A (en) | Method for controlling the compression rate of digital images | |
| US6697525B1 (en) | System method and apparatus for performing a transform on a digital image | |
| KR20210113681A (en) | Compression for multiple images | |
| Yang et al. | A fast and efficient codec for multimedia applications in wireless thin-client computing | |
| JP7063925B2 (en) | Encoder, encoding system, and encoding method | |
| EP4117289A1 (en) | Image processing method and image processing device | |
| CN100421465C (en) | Method of compression and recompression image data for image acquiring apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CX01 | Expiry of patent term | ||
| CX01 | Expiry of patent term |
Granted publication date: 20070117 |