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

Target detector and construction method and application thereof

A construction method and detector technology, which is applied in the field of target detection, can solve problems such as low classification accuracy and target detection is susceptible to background interference, and achieve the effects of improving sensitivity, increasing discrimination, and improving classification accuracy

Active Publication Date: 2020-01-21
HUAZHONG UNIV OF SCI & TECH
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a target detector and its construction method and application, which are used to solve the technical problem that the target detection in the existing target detector is susceptible to background interference and the classification accuracy is not high

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
  • Target detector and construction method and application thereof
  • Target detector and construction method and application thereof
  • Target detector and construction method and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] A construction method 100 of a target detector, such as figure 1 shown, including:

[0044] Step 110, building a Faster R-CNN target detection model framework, which includes a region proposal network module and multiple cascaded multi-core and multi-background detection structures;

[0045] Step 120, using the region suggestion network module to generate a training sample set;

[0046] Step 130, based on the training sample set and its weight distribution, using a loss function to iteratively train multiple cascaded multi-core and multi-background detection structures to obtain a Faster R-CNN target detection model;

[0047] Among them, during each iterative training, after each multi-core and multi-background detection structure is trained, the weight distribution is updated and the training sample with a large loss function value has a large weight, and is generated based on the updated weight distribution and the current multi-core and multi-background detection st...

Embodiment 2

[0081] A target detector, which is constructed by using any one of the above target detector construction methods. The target detector constructed by the above construction method gives full play to the respective advantages of traditional filtering algorithms and machine learning algorithms, avoids manual selection of thresholds, has good robustness in different cloud backgrounds, and has high classification accuracy .

[0082] The relevant technical solutions are the same as those in Embodiment 1, and will not be repeated here.

Embodiment 3

[0084] A storage medium, wherein instructions are stored in the storage medium, and when a computer reads the instructions, the computer is made to execute any method for constructing an object detector as described in Embodiment 1 above.

[0085] The relevant technical solutions are the same as those in Embodiment 1, and will not be repeated here.

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 target detector and a construction method and application thereof, and the method comprises the steps: building a Faster R-CNN target detection model framework which comprises a regional suggestion network module RPN and a plurality of cascaded multi-core multi-background detection structures; adopting RPN to generate a training sample set; based on the training sample set and the weight distribution thereof, iteratively training a plurality of cascaded multi-core multi-background detection structures by adopting a loss function to obtain a Faster R-CNN target detection model; wherein after each multi-core multi-background detection structure is trained in each iterative training, the weight distribution is updated, the weight of the training sample with a large loss function value is large, and the cascaded next multi-core multi-background detection structure is trained based on the updated weight distribution and the regression sample generated by the current multi-core multi-background detection structure. According to the invention, a plurality of cascaded multi-core multi-background detection structures are introduced into the Faster R-CNN, and training is carried out based on weight distribution and updating thereof, so that the classification precision of the whole detector is improved, and the detector has relatively good detection performancein a complex background.

Description

technical field [0001] The invention belongs to the field of target detection, and more specifically relates to a target detector and its construction method and application. Background technique [0002] Object detection is a complex problem that needs to solve two main tasks, first the detector must solve the problem of recognition, distinguish the object from the background, and assign it a corresponding class label. Second, the detector must solve the localization problem (aka regression) to assign accurate bounding boxes to objects. These two tasks are particularly difficult because there are many suspected objects in the image that are very close to the target features, which will affect the judgment of the detector. Therefore, detectors must find real objects while suppressing these suspected objects. [0003] When the target (especially UAV) is in the complex background, it is difficult for the detector to detect the target correctly, and the performance of the det...

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): G06F9/46G06K9/62G06N3/04
CPCG06F9/46G06N3/045G06F18/24Y02T10/40
Inventor 胡静熊涛卢鑫鑫蒋侃高翔张旭阳康愫愫
Owner HUAZHONG 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