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

Salient target extraction method based on feature perception

A target extraction and salient technology, applied in the field of image processing, can solve the problems of reducing the effect of salient objects, insufficient salient attributes, etc., to achieve the effect of wide application, high accuracy, and improved accuracy

Active Publication Date: 2022-02-08
LOW SPEED AERODYNAMIC INST OF CHINESE AERODYNAMIC RES & DEV CENT
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the salient attributes considered in the salient area are not sufficient, and it is easy to generate highlight errors, and once the error is made, the effect of salient object extraction will be greatly reduced

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
  • Salient target extraction method based on feature perception
  • Salient target extraction method based on feature perception
  • Salient target extraction method based on feature perception

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0064] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

[0065] The present embodiment provides a method for extracting salient objects based on feature perception, comprising the following steps:

[0066] S1: Process the original image through the characteristic MSCN factor, characteristic image entropy, characteristic dark channel, H channel of characteristic HSV channel and S channel of characteristic HSV channel to obtain 5 feature layers, and obtain the feature matrix ; During processing, such as figure 1 As shown, it is the original picture. After the original picture is imported, 5 feature layers with the same size as the original picture are generated, in which the MSCN factor represents the texture feature, and the feature layer obtains the feature matrix , the image entropy represents the de...

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 belongs to the technical field of image processing, and particularly relates to a salient target extraction method based on feature perception. The method comprises the steps: processing the original picture through the feature MSCN factor, the feature image entropy, the feature dark channel, the H channel of the feature HSV channel and the S channel of the feature HSV channel to obtain five feature layers, and obtaining a feature matrix; performing quadruple down-sampling on the feature matrix to obtain a down-sampling feature matrix; performing normalization processing on the down-sampling feature matrix to obtain a normalized down-sampling feature matrix; performing single-feature salient target extraction on the normalized down-sampling feature matrix to obtain a feature matrix; carrying out weight fusion on the single-feature salient target to obtain a feature matrix, wherein the position of the feature matrix in the original image is a first salient target. According to the method, multi-source complementary information of different levels such as features and decisions is fused, and the accuracy of detection of salient target extraction is improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a salient target extraction method based on feature perception. Background technique [0002] Since the subject of machine vision aroused people's extensive and close attention in the 1960s, technicians have been unanimously committed to researching projects that computers can have the same ability to adapt to the external environment as humans. In order to enable the computer to perform high-level semantic analysis and understanding, the object extraction link in the image must first be solved. Because object extraction has become a basic problem in computer vision, and the research results based on the computational model of visual attention based on saliency have also made great progress. [0003] How to efficiently acquire salient objects of an image. According to the calculated object saliency calculation, it can be divided into gaze point-based salien...

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): G06V10/80G06K9/62
CPCG06F18/253
Inventor 左承林何苗易贤熊浩赵荣
Owner LOW SPEED AERODYNAMIC INST OF CHINESE AERODYNAMIC RES & DEV CENT