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

A target detection method, system and terminal equipment

A target detection and target technology, applied in image analysis, image enhancement, instruments, etc., can solve problems such as low target detection efficiency and accuracy, and achieve the effect of improving efficiency and accuracy

Active Publication Date: 2020-08-14
SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a target detection method, system and terminal equipment to solve the problem of low efficiency and accuracy of target detection in the prior art

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 target detection method, system and terminal equipment
  • A target detection method, system and terminal equipment
  • A target detection method, system and terminal equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] figure 2 It shows the implementation flowchart of the target detection method provided by an embodiment of the present invention. For the convenience of description, only the parts related to the embodiment of the present invention are shown, and the details are as follows:

[0039] Such as figure 2 As shown, a target detection method provided by an embodiment of the present invention includes:

[0040] Step S101, using a convolutional neural network to extract several targets from the image under test using detection frames.

[0041] Step S102, acquiring feature attributes of the target, where the feature attributes include spatial features and explicit features.

[0042] Step S103, according to the feature attribute, calculate the correlation feature between objects based on the relationship calculation model.

[0043] Step S104, using the associated features to integrate the feature attributes of the target to obtain aggregated features.

[0044] Step S105, sub...

Embodiment 2

[0118] An embodiment of the present invention also provides a target detection system for performing figure 2 The method step in the corresponding embodiment, it comprises:

[0119] The target extraction module is used to use the convolutional neural network to extract several targets from the image under test using the detection frame.

[0120] The feature acquisition module is used to acquire feature attributes of the target, and the feature attributes include spatial features and explicit features.

[0121] The association calculation module is used to calculate the association features between objects based on the relationship calculation model according to the feature attributes.

[0122] An integration module, configured to use the associated features to integrate the feature attributes of the target to obtain aggregated features.

[0123] A prediction module is used for substituting the aggregated features into a retrospective relational sub-network composed of the f...

Embodiment 3

[0129] Figure 6 It is a schematic diagram of a terminal device provided by an embodiment of the present invention. Such as Figure 6 As shown, the terminal device 6 of this embodiment includes: a processor 60 , a memory 61 , and a computer program 62 stored in the memory 61 and operable on the processor 60 . When the processor 60 executes the computer program 62, it realizes the steps in each embodiment as described in Embodiment 1, for example figure 2 Steps S101 to S107 are shown. Alternatively, when the processor 60 executes the computer program 62, functions of the modules / units in the system embodiments described in Embodiment 2 are implemented.

[0130] The terminal device 6 refers to a terminal with data processing capabilities, including but not limited to computers, workstations, servers, and even smart phones with excellent performance, palmtop computers, tablet computers, personal digital assistants (PDAs), smart TVs (Smart TVs) TV) etc. Operating systems are...

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 present invention provides a target detection method, system and terminal equipment, wherein the method includes: using a convolutional neural network to extract several targets from a tested image using a detection frame; acquiring feature attributes of the target, the feature attributes include Spatial features and explicit features; according to the feature attributes, based on the relational calculation model, calculate the associated features between the targets; use the associated features to integrate the feature attributes of the targets to obtain aggregated features; substitute the aggregated features into The retrospective relationship sub-network composed of the fully connected layer of the convolutional neural network and the relationship calculation model calculates the quasi-prediction score; calculates the classification score according to the quasi-prediction score and the detection frame; according to the classification score Classify and detect objects. The invention greatly improves the efficiency and accuracy of target detection.

Description

technical field [0001] The invention belongs to the technical field of target detection, and in particular relates to a target detection method, system and terminal equipment. Background technique [0002] Target detection has important application value in many fields, such as intelligent deployment and control security, industrial information applications, and automobile assisted driving. However, various appearance features, complex and diverse background environments, dynamically changing scenes between pedestrians and cameras, and strict requirements for real-time performance and stability of the system pose great challenges to the target detection problem. The current target detection method based on deep learning has too many redundant calculations, and cannot accurately identify small and dense targets in the scene, and there are problems such as insufficient use of effective information in the scene. [0003] To sum up, there is a problem of low efficiency and accu...

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): G06T7/00G06N3/04
CPCG06T7/0002G06T2207/10004G06T2207/20081G06N3/045
Inventor 张维桐田艳玲张锲石程俊
Owner SHENZHEN INST OF ADVANCED TECH CHINESE ACAD OF SCI
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