Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

An Online Image Collection Compression Method Preserving Classification Accuracy

A compression method and precision maintenance technology, applied in the field of image processing, can solve the problem that the quantitative influence of image classification accuracy is not clear, and achieve the effect of reducing time, reducing storage capacity requirements, and strengthening the robust performance of classification

Active Publication Date: 2022-05-06
DONGHUA UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Different image compression parameters generate compressed image sets with different storage capacities, but the quantitative impact of compression parameters on image classification accuracy is not yet clear

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
  • An Online Image Collection Compression Method Preserving Classification Accuracy
  • An Online Image Collection Compression Method Preserving Classification Accuracy
  • An Online Image Collection Compression Method Preserving Classification Accuracy

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0017] Below in conjunction with specific embodiment, further illustrate the present invention. It should be understood that these examples are only used to illustrate the present invention and are not intended to limit the scope of the present invention. In addition, it should be understood that after reading the teachings of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

[0018] The embodiment of the present invention relates to an online image set compression method with classification precision maintenance, comprising the following steps:

[0019] Step 1. Initialization process. The first image set needs to work in training mode and test all compression parameters to obtain the optimal compression method F * 1 , the corresponding compression parameter is {Q * (I 1 ),S * (I 1 )}, the cla...

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 on-line image set compression method with classification precision maintenance, which includes: properly compressing two compression parameters of the image set: quality factor Q and resolution S. Based on the convolutional neural network classifier, the image sets obtained under different compression parameters are classified under the classifier, and the classification accuracy is compared and analyzed to obtain a data set compression method with accuracy preservation. The optimal compression method provides a reference for the selection of a two-parameter compression method that maintains the classification accuracy of subsequent image sets. The present invention can quickly and accurately find the optimal compression method while maintaining the classification accuracy of the online image collection, and greatly reduces the time required for the optimal compression while maintaining the classification accuracy of the online image collection.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an online image collection compression method with classification precision maintained. Background technique [0002] As an important pattern recognition problem, image classification needs to classify the image set according to certain standards according to the characteristics of the image, and it has received more and more attention in the military and civilian fields. In recent years, the framework of "feature extraction + classifier" mode has become a classic architecture in the field of pattern recognition. Convolution Neural Network (CNN) has been widely used in the field of image classification. The CNN framework first extracts the features of the image set, and sends the extracted image features of the dataset to the classifier for classification, and finally obtains the image classification results. Compared with the traditional image classification, the CNN ...

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 Patents(China)
IPC IPC(8): G06T9/00
CPCG06T9/002
Inventor 吴乐明刘浩魏国林孙嘉曈陈根龙刘洋黄震况奇刚
Owner DONGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products