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

Privacy camera system based on deep learning

A deep learning and camera system technology, applied in neural learning methods, image enhancement, image analysis, etc., can solve problems such as low processing efficiency, long protection processing time, cumbersome video content protection processing, etc., to improve processing efficiency, The effect of protecting video privacy

Pending Publication Date: 2022-03-01
AEROSPACE HI TECH HLDG GROUP
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the current privacy protection method for video content cannot be directly implemented on the camera side, resulting in cumbersome protection processing of video content, long protection processing time and low processing efficiency per unit time, and proposes Privacy camera system based on deep learning

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
  • Privacy camera system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0019] Specific Embodiment 1: The privacy camera system based on deep learning in this embodiment includes: an input module, a preprocessing module, a gesture judgment module, a human body contour segmentation module, a face recognition module, and an output module; the flow chart of the present invention is as follows: figure 1 .

[0020] Described input module is used for obtaining the video that needs privacy protection;

[0021] The preprocessing module is used to perform noise filtering processing on the input video;

[0022] The gesture judgment module adopts MTCNN-P network, which is used to judge whether the video after noise filtering is protected according to the gesture;

[0023] The human body contour segmentation module adopts an improved U-NET network, which is used to extract the human body foreground and background information in the video verified by the gesture judgment module, and to block the background part;

[0024] The improved U-Net network is a netwo...

specific Embodiment approach 2

[0030] Specific implementation mode two: the preprocessing module is used to perform noise filtering processing on the input video, including the following steps:

[0031] Step 1. Determine whether the input video is a daytime video or a nighttime video;

[0032] Step 2: Perform median filter processing on each frame image in the night video to obtain a noise-filtered video.

[0033] Beneficial effects of this embodiment: noise specifically refers to noise, a type of interference signal, especially at night when there are many image noises. The median filtering method is used to remove noise, because this method can not only remove noise, but also protect image edge information. Before the image adopts the median filter algorithm, it is judged whether the image belongs to the daytime image or the night image, and only the night image is processed, because the night image has more noise.

specific Embodiment approach 3

[0034] Specific embodiment three: in the step 1, judging whether the video input belongs to a daytime image or a night video is realized by binarizing each frame of image in the input video;

[0035] The fixed threshold value of the binarized image is 127.

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 privacy camera system based on deep learning, and relates to the field of multimedia processing. The invention aims to solve the problem of low processing efficiency in unit time caused by tedious video content protection processing process and long processing time of the existing video content privacy protection method. The system comprises an input module, a preprocessing module, a gesture judgment module, a human body contour segmentation module, a face recognition module and an output module. The input module is used for acquiring a video needing to be protected; the preprocessing module is used for carrying out noise filtering processing on an input video; the gesture judgment module is used for judging whether videos can be stored or read according to gestures; the human body contour segmentation module is used for extracting human body foreground and background information in the video and shielding a background part; the face recognition module is used for recognizing whether a face image is a set person or not before the shielded video is output; and the output module is used for outputting the video after privacy protection. The method and the device are used for privacy protection of video contents.

Description

technical field [0001] The invention belongs to the field of multimedia processing, in particular to a privacy camera system based on deep learning. Background technique [0002] In modern society, security and privacy are problems that everyone and even every enterprise has to face. Issues related to security and privacy can be seen everywhere in life, such as education, medical care, transportation and so on. In recent years, with the development of network technology, video has many advantages as a carrier of information transmission, and the important equipment of video source - camera has attracted much attention. How to collect video safely and controllably has become a problem that users have to face. [0003] The existing technical solutions are of a great variety due to different emphases. At present, the privacy protection of video content is divided into three parts, namely: face recognition, privacy area protection, and privacy access protection. At present, f...

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): G06T7/194G06T5/00G06V20/40G06V40/16G06V40/20G06N3/04G06N3/08
CPCG06T7/194G06N3/08G06T2207/10016G06T2207/20032G06T2207/20081G06T2207/20084G06T2207/20112G06T2207/30196G06N3/045G06T5/70
Inventor 李秋生周谷春阳谭思齐
Owner AEROSPACE HI TECH HLDG GROUP
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