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

A method of object region detection based on deep learning

A target area and deep learning technology, applied in neural learning methods, image analysis, instruments, etc., can solve problems such as difficult target detection and confusion

Active Publication Date: 2021-09-17
BEIJING UNIV OF TECH
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Target detection is not difficult for humans. It is easy to locate and classify the target objects through the perception of different color modules in the picture, but for computers, it is difficult to get directly from the image because of the RGB pixel matrix Abstract concepts such as dogs and cats and their positions are located, and sometimes multiple objects and cluttered backgrounds are mixed together, making target detection more difficult
The core problems to be solved by target detection are: 1. The target may appear anywhere in the image
3. Targets may come in various shapes

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 method of object region detection based on deep learning
  • A method of object region detection based on deep learning
  • A method of object region detection based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] The present invention will be described in further detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0042] The used hardware equipment of the present invention has 1 PC machine, 1 nvidia1080 graphics card;

[0043] Such as figure 1 As shown, the present invention provides a target area detection method based on deep learning, and the following is the specific content of the experiment under coco2017. Specifically include the following steps:

[0044] Step 1, get the coco2017 image dataset. and clean these data. Since the 2017coco dataset is a relatively clean public dataset, the pictures were not deleted.

[0045] Step 2, image preprocessing, because each image in the coco dataset is labeled, so all the data are image enhanced. Images undergo data augmentation with 50% probability. The data enhancement used mainly includes rotation, flip, contrast enhancement, cropping, brightness, and affine transformation....

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

A target area detection method based on deep learning belongs to the technical field of computer vision, and the method mainly uses a retinanet detection network. RetinaNet is essentially a network structure composed of resnet+FPN+two FCN subnetworks. Here I replace the previous resnet with ResNeXt50 and densenet169 for the backbone. And the FPN layer and loss function of the retnanet network were modified, and finally the fusion of the models was carried out. This target detection method combines the advantages of the current mainstream target detection methods, and has solved a series of practical problems. This algorithm was tested under coco2017, and the performance is very good. It is better than the single model under retinanet and the result when the model is not improved. In addition, it also has better performance on other data sets.

Description

technical field [0001] The invention belongs to the technical field of computer vision, mainly relates to the improvement of deep learning image detection methods, and involves some traditional image processing. Background technique [0002] With the development of artificial intelligence, the application of computer vision has also been vigorously developed. In computer vision applications, image detection is an important branch, and image target detection is of great significance in fields such as face recognition, unmanned driving, unmanned retail, and intelligent medical care. [0003] Image target detection is an important research direction in computer vision. With the development of deep learning, target detection technology has made great progress. Target detection is not difficult for humans. It is easy to locate and classify the target objects through the perception of different color modules in the picture, but for computers, it is difficult to get directly from ...

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): G06T7/00G06N3/04G06N3/08
Inventor 张涛郝兵冯宇婷
Owner BEIJING UNIV OF 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