Anti-candid shot detection method based on deep learning

A deep learning and detection method technology, applied in the field of image processing and target detection, can solve the problems of low precision and inability to achieve real-time effects, achieve high accuracy rate, good detection effect, and overcome the effect of large amount of calculation

Active Publication Date: 2021-04-06
XIDIAN UNIV
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the above-mentioned deficiencies in the prior art, and propose an anti-candid detection method based on a deep learning network, which solves the problems of low precision and inability to achieve real-time effects when detecting candid behavior in videos

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
  • Anti-candid shot detection method based on deep learning
  • Anti-candid shot detection method based on deep learning
  • Anti-candid shot detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052] The present invention will be further described below in conjunction with the accompanying drawings.

[0053] refer to figure 1 , to further describe in detail the specific steps for realizing the present invention.

[0054] Step 1, build a deep learning target detection network.

[0055] Build a Yolov3 target detection network consisting of four modules;

[0056] The structure of the first module is as follows: input layer→1st convolutional layer→2nd convolutional layer→1st convolutional submodule→3rd convolutional layer→2nd convolutional submodule→4th convolutional layer→ The third convolutional submodule → the fifth convolutional layer → the fourth convolutional submodule → the sixth convolutional layer → the fifth convolutional submodule; the second convolutional submodule is composed of four sequentially connected first Composed of convolution units; the third convolution sub-module is composed of eight second convolution units connected in series; the fourth co...

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 an anti-sneak shot detection method based on deep learning, the steps of which are as follows: 1. Construct a deep learning target detection network; 2. Generate a training set; 3. Adopt multiple scale frame methods to detect the same sneak shot behavior in a picture Mark; 4. Train the deep learning network; 5. Detect candid behavior; 6. Enhance the image without candid behavior; 7. Re-detect the image after feature enhancement. The invention overcomes the problem of low detection accuracy caused by diversification of candid photography behaviors and actions by adopting a variety of scale frame methods when marking the data set, constructs a deep learning network and performs image enhancement processing on the humanoid area, ensuring In order to achieve real-time effects in the detection of candid behavior in surveillance video, it has a high accuracy rate.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a deep learning-based anti-sneak shooting detection method in the technical field of target detection. The invention can detect in real time the sneak shooting behavior of people captured in the video surveillance. Background technique [0002] Anti-surveillance detection in video surveillance is a very necessary behavior in many confidential institutions and units, which can prevent the internal confidential information of institutions or units from leaking to the outside world. However, in practice, it is time-consuming and labor-intensive to manually detect sneak shots in video surveillance, and it is difficult to achieve real-time detection. In order to solve the above problems, people usually design target detection methods, and use computers to detect sneak shots in video surveillance. [0003] Yulinian Electronics Nantong Co., Ltd. provides a computer visi...

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): G06K9/00G06N3/04G06N3/08
CPCG06N3/08G06V40/20G06N3/045
Inventor 张静胡锐周秦申枭李云松
Owner XIDIAN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap