Unlock instant, AI-driven research and patent intelligence for your innovation.

Post selection prediction method in bandwidth compression

A prediction method and bandwidth compression technology, applied in the multimedia field, can solve problems such as the deterioration of image effects and the inability of prediction methods to obtain the prediction effect, and achieve the effect of optimizing the prediction effect.

Inactive Publication Date: 2019-04-16
XIAN CREATION KEJI CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In areas with complex image textures, a single prediction method often cannot obtain the best prediction results, resulting in poor image effects

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
  • Post selection prediction method in bandwidth compression
  • Post selection prediction method in bandwidth compression
  • Post selection prediction method in bandwidth compression

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0020] See figure 1 , figure 1 It is a flowchart of a post-selection prediction method in bandwidth compression provided by an embodiment of the present invention. The method comprises the steps of:

[0021] Step 1. Divide the image into multiple MBs with a size of m×n; wherein, m and n are the row resolution and column resolution of each MB respectively;

[0022] Step 2. Using the adaptive template prediction method and the adaptive texture gradient prediction method to predict the current MB to obtain the corresponding residual subjective sum;

[0023] Step 3. Determine the final prediction residual for each pixel in the current MB according to the subjective sum of the residual.

[0024] Specifically, step 2 may include the following steps:

[0025] Step 21. Using an adaptive template prediction method to predict the current MB to obtain the first residual subjective sum;

[0026] Step 22: Predict the current MB by using an adaptive texture gradient prediction method t...

Embodiment 2

[0056] This embodiment focuses on the description of the adaptive template prediction method on the basis of the foregoing embodiments. Specifically, the adaptive template prediction method provided by the present invention includes the following steps:

[0057] Step 1. Update the adaptive template corresponding to the current MB;

[0058] Step 2. Obtain the prediction residual of the current MB according to the updated adaptive template;

[0059] Step 3. Judging whether all MB prediction residuals have been obtained, if yes, the prediction ends; otherwise, jump to step 1.

[0060] Among them, before step 1, it may also include:

[0061] Step X1, determining the number of epitopes in the adaptive template list and the sequence number of the epitopes;

[0062] Step X2, initializing and filling the adaptive template.

[0063] Among them, step 1 may include:

[0064] Step 11, detecting the MB reconstruction value of the adjacent reference direction of the current MB;

[006...

Embodiment 3

[0086] This embodiment focuses on the detailed description of the adaptive template in the adaptive template prediction method based on the foregoing embodiments.

[0087] See figure 2 , figure 2 It is a schematic diagram of an adaptive template provided by an embodiment of the present invention. The establishment of the adaptive template includes the following steps:

[0088] Step 1. Define the number of epitopes and the sequence number of the epitopes of the adaptive template

[0089] Preferably, the number of epitopes in the self-adaptive template can be defined as 4, 8, 16 or 32; in this embodiment, the number of epitopes is 16 as an example, and the same is true for other numbers of epitopes. The number of epitopes in the adaptive template is 16, and the epitope numbers are arranged in order from 0 to 15. The smaller the number, the higher the priority. Each epitope records a set of reconstruction values ​​of one MB. The MB size can be set, and this embodiment takes...

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 a post selection prediction method in bandwidth compression. The method comprises the following steps: dividing an image into several m*n MB, wherein m and n are respectivelyrow resolution and column resolution of each MB; respectively using an adaptive template prediction method and an adaptive texture gradient prediction method to predict the current MB to obtain corresponding residual error subjective sum; and determining final prediction residual error of each pixel in the current MB according to the residual error subjective sum. In the post selection predictionmethod in bandwidth compression provided by the invention, based on the adaptive template prediction method and the adaptive texture gradient prediction method, an optimal prediction method can be selected through a prediction selection algorithm, and prediction effects for a complex texture image are optimized further.

Description

technical field [0001] The invention relates to the field of multimedia technology, in particular to a post-selection prediction method in bandwidth compression. Background technique [0002] The bandwidth compression technology mainly consists of four parts, including: prediction module, quantization module, code control module and entropy coding module. The prediction module, as an important module, uses the spatial redundancy between adjacent pixels to predict the current pixel value according to the adjacent pixel information. The standard deviation of the predicted difference is much smaller than the standard deviation of the original image data. Encoding the predicted difference is more conducive to minimizing the theoretical entropy of the image data and achieving the purpose of improving compression efficiency. At present, the algorithms of the prediction module are mainly divided into two categories, texture-related prediction and pixel value-related prediction. ...

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): H04N19/192H04N19/50
CPCH04N19/192H04N19/50
Inventor 张莹罗瑜
Owner XIAN CREATION KEJI CO LTD