Dictionary coding method based on Shannon coding, image processing method and processing equipment

A technology of image processing and coding methods, applied in the field of super-resolution images, can solve the problems of occupying storage space, large vocabulary, and long time to read into memory, etc., and achieve the effect of reducing model complexity and simplifying learning tasks

Pending Publication Date: 2022-04-29
SHENZHEN HONGDIAN TECH CORP
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The purpose of the present invention is to provide a super-resolution dictionary encoding method and image processing method based on Shannon encoding, to solve the problem in the prior art that the dictionary based on sparse representation usually has a large vocabulary and takes up a large amount of storage space, causing it to be read into The memory time is too long, which increases the technical problem of the preloading time of the super-resolution dictionary model in the embedded system

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
  • Dictionary coding method based on Shannon coding, image processing method and processing equipment
  • Dictionary coding method based on Shannon coding, image processing method and processing equipment
  • Dictionary coding method based on Shannon coding, image processing method and processing equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] Such as figure 1 As shown, the present invention provides a dictionary encoding method based on Shannon encoding, and the dictionary encoding method adopts Shannon encoding, including the following steps. S10: Sort the q numerical symbols in the sparse representation dictionary in the order of decreasing occurrence probability. The sum of the probability of occurrence of all numerical symbols in the dictionary is 1, but the probability of each numerical symbol is different. After sorting, the first numerical value The probability of occurrence of the symbol is the highest, followed by the occurrence probability of the second numerical symbol, and so on, the probability of occurrence of the qth numerical symbol is the lowest. S20: Pass -logP(S i )≤l i i ) inequality, calculate the integer code length l of each numerical symbol i , that is, for l i The obtained value range is rounded, wherein i=1, 2, ... q, P(S i ) is the probability of the i-th numerical symbol appe...

Embodiment 2

[0033] An image processing method based on Shannon coding, the image processing method performs super-resolution processing through the super-resolution dictionary provided in Embodiment 1, such as figure 2 As shown, the following steps are included: S100: Input the low-resolution image to be processed, and perform block segmentation to obtain multiple low-resolution segmentation blocks, and the size of the block segmentation is based on the size of the low-resolution image and the required high resolution Image OK. S200: Perform feature extraction on all low-resolution segmentation blocks to obtain a feature block of each low-resolution segmentation block. The features mainly include information such as brightness, edge, texture, and color of the low-resolution segmentation block. S300: Find the primary atom closest to the feature block in the super-resolution dictionary, subtract the feature block and the atom, and obtain a primary residual. An atom is an entry in a super-...

Embodiment 3

[0039] A processing device comprising: one or more processors; a memory for storing one or more computer programs, and one or more processors for executing the one or more computer programs stored in the memory so that one or more The processor executes the Shannon coding-based dictionary coding method of Embodiment 1, and / or the Shannon coding-based image processing method of Embodiment 2. The processing device enables the compression of the super-resolution dictionary, reduces the storage space, and shortens the time to read it into the memory, thereby reducing the preloading time of the super-resolution dictionary model in the embedded system and improving the efficiency of image super-resolution processing .

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 discloses a dictionary coding method based on Shannon coding, an image processing method and processing equipment, relates to the technical field of super-resolution images, and solves the technical problems that in the prior art, a sparse representation dictionary is generally large in vocabulary quantity and occupies a large amount of storage space, and the preloading time is increased. The coding method comprises the following steps: sorting q numerical symbols in a sparse representation dictionary according to an occurrence probability decreasing sequence; -logP (Si) is less than or equal to project; calculating the integer code length li of each numerical symbol according to the 1, 1-logP (Si) inequality; calculating an accumulation probability Gi of the ith numerical value symbol according to the sequence of the numerical value symbols; the i bit after the decimal point of the binary number corresponding to the accumulation probability Gi is taken to form a binary code word of the ith numerical value symbol; and binary code words of all numerical symbols are calculated to obtain a super-resolution dictionary. According to the method, the super-resolution dictionary is compressed, the storage space is reduced, the time for reading the super-resolution dictionary into the memory is shortened, and the image super-resolution processing efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of super-resolution images, in particular to a dictionary encoding method based on Shannon encoding, an image processing method and a processing device. Background technique [0002] Find a suitable dictionary for the samples of ordinary dense expressions, and convert the samples into suitable sparse expressions, so that the learning task is simplified and the complexity of the model is reduced. It is usually called "dictionary learning", also known as "sparse coding". "(sparse coding). [0003] At present, the dictionary learning based on sparse representation in the existing image super-resolution technology is usually: [0004] [0005] Among them, the matrix B (d*k order matrix) is a dictionary matrix, k is called the vocabulary of the dictionary, usually specified by the user, α i is sample X i sparse representation of . The first term in the formula is the hope α i Can refactor X nicely i , an...

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
IPC IPC(8): G06T9/00G06T3/40G06T5/50G06T7/11G06V10/44
CPCG06T9/00G06T3/4053G06T5/50G06T7/11G06T2207/20021G06T2207/20224
Inventor张小虎左绍舟龚潇刘文强丁靖
OwnerSHENZHEN HONGDIAN TECH CORP