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

Image scene type discrimination method based on covariance features

A discriminant method and covariance technology, applied in character and pattern recognition, instruments, computer components, etc., can solve the problems of insufficient generalization ability of SIFT scene classifiers, unstable number of SIFT points, uneven distribution, etc., to achieve Effects of Scalability on Strong Sorting Capabilities

Active Publication Date: 2014-03-26
PLA UNIV OF SCI & TECH +2
View PDF3 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But it still uses SIFT as the basic feature, and the amount of calculation has not changed.
[0005] The scene classification method based on SIFT points is also affected by unfavorable factors such as unstable number of SIFT points and uneven distribution.
In particular, SIFT points are low-level image features and do not have semantic recognition capabilities, so SIFT-like scene classifiers still have the problem of insufficient generalization ability

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
  • Image scene type discrimination method based on covariance features
  • Image scene type discrimination method based on covariance features
  • Image scene type discrimination method based on covariance features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention includes three parts: 1) In the dictionary formation stage, use the image segmentation tool to manually complete the extraction of interest targets with semantic functions in the training image, calculate the covariance features of the interest targets, and perform Sigma points on the covariance features of the interest targets Decompose, convert to Euclidean space vectors, and then cluster the vectors of interest objects extracted from all training images to obtain the feature dictionary of interest objects. 2) In the classifier training stage, the training image is evenly divided into blocks, and the Euclidean space vector of each block is calculated to form a histogram with the feature dictionary as an element. The histogram of the feature dictionary and the image scene category are used as input to classify the training scene. Support Vector Machines. 3) In the stage of scene recognition, the feature dictionary histogram of the image to be reco...

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 an image scene type discrimination method based on covariance features, wherein based on local Sigma point features of a covariance matrix of an object of interest, a semantic model is established for scenes, and the scenes are classified. The method comprises that on the basis of the interactive image segmentation, the object of interest, which is a typical object when a user selects such a scene, is extracted from the scene; the covariance matrix which represents the object of interest is formed by combining pixel positions, colors, Gabor features and LBP features; and the covariance matrix is converted into Sigma point features in an European-style space, and learning and discrimination of a scene classifier is completed via a support vector machine.

Description

technical field [0001] The invention relates to a method for discriminating image scene types, in particular to a method for classifying and recognizing image scenes by using covariance features. Background technique [0002] Image scene classification is an important issue in the field of computer vision. The scene category not only includes people's overall understanding of the image, but also has great significance for the detection and recognition of objects of interest in the scene, video surveillance and other computer vision applications. In image retrieval 1. The field of remote control device navigation has been widely used, so it has received extensive attention. At present, the research on scene classification has achieved certain results, but due to the complexity of the scene itself, illumination, occlusion and other changing factors, scene classification is still a challenging problem. [0003] The early scene classification method mainly uses the global featu...

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): G06K9/62
Inventor 吴泽民张凤霞邱正伦付毅田畅曾明勇
Owner PLA UNIV OF SCI & TECH
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com