Method and device for compression of ADPCM image

An image compression and pixel technology, applied in image communication, image coding, image data processing, etc., can solve the problems of ineffective improvement of image quality, unsatisfactory quantization effect, complicated encoding and decoding process, etc. Improve the effect, improve the objective quality and subjective effect, the effect of simplifying the encoding and decoding process

Active Publication Date: 2008-10-01
BEIJING VIMICRO ARTIFICIAL INTELLIGENCE CHIP TECH CO LTD
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0002] In a computer, an image is composed of a large number of pixels, and each pixel is represented by a set of binary numbers, and the higher the accuracy of the binary numbers, the more vivid the color of the entire image and the clearer the image, however In the process of image transmission, it is usually necessary to compress and encode the image into a code stream, and restore the original color and precision after transmission. Most of the existing technologies use ADPCM (Adaptive Differential Pulse Code Modulation), that is, Adaptive Differential Pulse Code Modulation. Modulation is used to achieve image compression, including prediction, quantization, and entropy coding processes. However, the coding used in the original ADPCM needs to be filled with complex line headers, strip headers, and pseudo-start codes. There are many padding words, and the encoding and decoding process is complicated. high complexity
In addition, the quantization process simply divides the quantization level according to the quantization interval. When the prediction error generated after the prediction step is large, the quantization effect is not very ideal. When the bit rate increases, the image quality cannot be effectively improved.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for compression of ADPCM image
  • Method and device for compression of ADPCM image
  • Method and device for compression of ADPCM image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] Embodiment one is the ADPCM image compression method that the present invention proposes, and its specific implementation process is as follows:

[0036] 1. First, use the reconstructed adjacent pixels A, B and C to calculate the current original pixel X 0 The predicted value X' of .

[0037] Each image is composed of pixels, and the value of the unknown pixel can be roughly judged by the value of the known pixel, such as figure 1 As shown in , assuming that the values ​​of the three pixels of A, B, and C are known, since the input image is a Bayer image, the effective bits of each pixel are 10 bits or 8 bits, then the fourth pixel point X 0 The predicted value X' of is:

[0038] When the effective bit of the pixel is 10 bits, X'=(C+B) / 2

[0039] When the effective bit of the pixel is 8 bits, X'=(3*C-2*A+3*B) / 4

[0040] where A, B, C and X 0 They are the color difference components of the same nature of adjacent pixels, if A, B or C does not exist, set its value to...

Embodiment 2

[0070] Embodiment 2 is an image compression device, such as figure 2 as shown in

[0071] A prediction unit, configured to use the reconstructed values ​​A, B, and C of adjacent pixels to calculate the predicted value X' of the current pixel;

[0072] Quantization unit, including three quantizers Q 0 , Q 1 and Q 2 and the quantizer selection subunit, the relationship between the three groups of quantizers is as follows:

[0073] Q 0 =0.5Q 1

[0074] Q 2 =1.75Q 1

[0075] Q 1 =[-D2, -D1, -D0, 0, D0, D1, D2]*QP

[0076] The optimal quantizer is selected from the three quantizers by the quantizer selection subunit for the prediction error (ie X 0 -X') is quantified;

[0077] Coding unit, the user performs entropy coding on the quantization result, and uses overflow code coding in the quantization interval at both ends;

[0078] An inverse quantization unit, configured to inverse quantize the quantization result output by the quantization unit;

[0079] The reconst...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to an ADPCM image compressing method and apparatus. According to the invention, an overflow code encoding strategy is added when a predicted error is big, external quality and active effect of encoded images are markedly improved without increasing encoding rate. In addition, the encoding / decoding process according to the invention are simplified, the compressing complexity of original images is reduced and the encoding performance is improved, thereby, the quality and active effect of the encoded images are markedly improved, the reduction of the complexity of an encoder / decoder is good for application of the algorism in a software ISP.

Description

technical field [0001] The invention relates to an ADPCM (Adaptive Differential Pulse Code Modulation) image compression method and device. Background technique [0002] In a computer, an image is composed of a large number of pixels, and each pixel is represented by a set of binary numbers, and the higher the accuracy of the binary numbers, the more vivid the color of the entire image and the clearer the image, however In the process of image transmission, it is usually necessary to compress and encode the image into a code stream, and restore the original color and precision after transmission. Most of the existing technologies use ADPCM (Adaptive Differential Pulse Code Modulation), that is, Adaptive Differential Pulse Code Modulation. Modulation is used to achieve image compression, including prediction, quantization, and entropy coding processes. However, the coding used in the original ADPCM needs to be filled with complex line headers, strip headers, and pseudo-start ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06T9/00H04N7/26H04N7/32H04N19/124
Inventor 王耀辉季鹏飞
Owner BEIJING VIMICRO ARTIFICIAL INTELLIGENCE CHIP TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap