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

BP neural network image compression method based on improved image block classification algorithm

A BP neural network and classification algorithm technology, applied in the field of image data compression of modern communication networks, can solve problems such as poor compression performance, achieve good compression performance, solve poor compression performance, and improve accuracy.

Pending Publication Date: 2020-11-27
ELECTRIC POWER RES INST STATE GRID SHANXI ELECTRIC POWER
View PDF0 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of poor compression performance of the existing communication network image data compression technology, the present invention provides a BP neural network image compression method based on an improved image block classification algorithm

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
  • BP neural network image compression method based on improved image block classification algorithm
  • BP neural network image compression method based on improved image block classification algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] A kind of BP neural network image compression method based on improved image block classification algorithm, this method is to adopt following steps to realize:

[0019] Step 1: Calculate the mean square error SMSE of each image block and the mean square error FIMSE of the entire image according to the mean value SIM of the image sub-block;

[0020] Step 2: Use the improved image block classification algorithm to reasonably select the classification threshold δ of the image block to divide the image block into three types of image sub-blocks, namely image smooth block, target block, and edge block;

[0021] Step 3: Normalize the classified image sub-blocks, so that the pixel value of each pixel is suitable for the requirements between 0 and 1 of the input data set of the BP neural network;

[0022] Step 4: Use the Levenberg-Marquardt algorithm with the largest memory requirement and the fastest convergence speed to train the BP neural network to obtain a compressed data...

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 modern communication network image data compression technology, and in particular relates to a BP neural network image compression method based on an improved image block classification algorithm. According to the invention, the problem of poor compression performance of the existing communication network image data compression technology is solved. The BP neural networkimage compression method based on the improved image block classification algorithm comprises the steps of 1 calculating a mean square error SMSE of each image block and a mean square error FIMSE ofa whole image; 2 dividing the image block into three kinds of image sub-blocks; 3 performing normalization processing on the classified image sub-blocks; 4 training a BP neural network to obtain a compressed data set, namely completing an image data encoding process; 5 respectively compressing the image sub-blocks classified in the step 2 to different degrees; and 6 reconstructing a decoded image.The method is suitable for complex image recognition and feature analysis in complex, special and severe engineering fields of modern communication network environment.

Description

technical field [0001] The invention relates to a modern communication network image data compression technology, in particular to a BP neural network image compression method based on an improved image block classification algorithm. Background technique [0002] Image transmission technology has become an important communication technology in many information communication methods. However, due to the huge amount of image information, the transmission bandwidth is extremely occupied during data transmission, which eventually leads to many difficulties in image transmission and storage. Therefore, in order to effectively utilize the space of modern communication networks and storage devices, whether it is image data transmission or storage, it is necessary to compress image data. [0003] The current image compression is mainly divided into two categories: one is the lossless compression that restores the original data of the image without causing distortion and can complet...

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): H04N19/42H04N19/176G06N3/08G06N3/04G06K9/62
CPCH04N19/42H04N19/176G06N3/084G06N3/044G06F18/241
Inventor 俞华黄纯德刘永鑫赵亚宁孟晓凯杨虹白洋韩钰田贇
Owner ELECTRIC POWER RES INST STATE GRID SHANXI ELECTRIC POWER