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

A PSO-based cooperative salient target detection method of an RGBD graph

A target detection and salience technology, applied in the field of computer vision, can solve problems such as insufficient detection effectiveness

Active Publication Date: 2019-04-09
ANHUI UNIVERSITY
View PDF13 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The above existing technologies generally have the defects of insufficient detection effectiveness and effect. Accordingly, there is an urgent need for an efficient and effective collaborative salient object detection method based on PSO RGBD images.

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
  • A PSO-based cooperative salient target detection method of an RGBD graph
  • A PSO-based cooperative salient target detection method of an RGBD graph
  • A PSO-based cooperative salient target detection method of an RGBD graph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] see Figure 1-2 : the present embodiment a kind of synergistic salient object detection method based on the RGBD figure of PSO, this method comprises the following steps:

[0035] S1, input the RGB image, the depth image and the salient image of the corresponding RGBD single image; utilize the gPb-owt-ucm segmentation method to perform regional segmentation to obtain the candidate target area; the specific process of the step S1 is: generate the candidate target area, and give Set a group of RGBD pictures We first generate an initial saliency map based on the RGBD single image saliency method, and then use the gPb-owt-ucm segmentation method to perform superpixel segmentation on the original RGB image, and divide it into Q regions; each region is defined as And the saliency value of each region is the average value of the saliency values ​​of all pixels in the region; finally, by setting a threshold T, those superpixels whose saliency value is greater than T are sele...

Embodiment 2

[0044] A kind of synergistic salient object detection method based on PSO RGBD graph of this embodiment, its method is basically the same as embodiment 1, the main difference is: the maximum number of iterations we set as 40-60 times according to experimental experience, the most The optimal maximum number of iterations is 50.

[0045] Experimental detection: Through the collaborative significance detection with other methods on the public data set, the PR curve of the detection result is compared as Figure 4 As shown, the evaluation index histogram is compared as Figure 5 It can be seen that our method has achieved good detection results, which fully demonstrates the effectiveness and universality of this method.

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 PSO-based cooperative salient target detection method for an RGBD graph. The method comprises the following steps: S1, inputting an RGB image, a depth image and a salient image of a corresponding RGBD single image; Carrying out region segmentation by utilizing a gPb-owt-ucm segmentation method to obtain a candidate target region; S2, obtaining an optimal clustering centernumber and an optimal feature of each particle by using a PSO method; S3, performing clustering through the initial population to obtain a saliency map; And S4, selecting the optimal particles by using the clustering quality, and updating the saliency map. According to the invention, the PSO is utilized to extract the target characteristics so as to carry out subsequent significant detection; ThePSO can automatically determine the classification number and extract features at the same time so as to perform collaborative saliency detection, the PSO is firstly utilized to obtain features, andthe features are utilized to perform region classification so as to obtain a final multi-classification result and a collaborative saliency map. According to the invention, the effectiveness and obvious advantages in effect are proved through image library test comparison.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a collaborative salient target detection method based on a PSO RGBD image. Background technique [0002] Salient object detection in computer vision has attracted increasing attention in recent years. Numerous saliency detection models focus on detecting salient objects from a single image and achieve excellent performance. Co-saliency detection, as an emerging and challenging problem based on salient object detection, has received increasing attention in recent years. Different from traditional single saliency detection models, co-saliency detection methods focus on finding common salient objects in multiple images. Co-saliency detection is mostly used in public pattern recognition, image matching, and co-recognition. It can be regarded as a combination of similar object recognition and co-classification tasks, and it belongs to the basic research work in computer vision. [00...

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): G06K9/62G06N3/00G06T7/11G06T7/40G06T7/50G06T7/90
CPCG06N3/006G06T7/11G06T7/40G06T7/50G06T7/90G06F18/23213
Inventor 汪蕊刘政怡谢丰
Owner ANHUI UNIVERSITY
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