A multi-modal security monitoring method based on deep learning image processing

An image processing and deep learning technology, applied in neural learning methods, genetic models, character and pattern recognition, etc., can solve the problems of high equipment requirements, increased cost and overhead, and poor applicability, and achieve high tracking monitoring, The effect of increasing the detection blind area and enriching the amount of data

Active Publication Date: 2021-06-18
CHENGDU UNIV
View PDF18 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In this case, if multiple cameras are monitored online at the same time, the overhead is huge, the requirements for equipment are quite high, and the applicability will be much worse. At the same time, for the diversification of monitoring , the rich content of the monitoring algorithm will also increase the cost and overhead. If the increased overhead of the algorithm is superimposed on the monitoring of multiple cameras, the overhead will increase exponentially.

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 multi-modal security monitoring method based on deep learning image processing
  • A multi-modal security monitoring method based on deep learning image processing
  • A multi-modal security monitoring method based on deep learning image processing

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] This embodiment proposes a multi-mode security monitoring method based on deep learning image processing, such as figure 1 , figure 2 shown, including the following steps.

[0047] Step 1: Build a global map model; specifically, build a global map model of the monitored site.

[0048] Step 2: Divide sub-areas and install surveillance cameras; specifically, divide multiple different security monitoring sub-areas under the constructed global map model, divide different security monitoring sub-areas into different monitoring levels, and monitor each security monitoring sub-area. Corresponding surveillance cameras are installed in the monitoring sub-area and run for a period of time.

[0049] Step 3: Prepare the pre-training set; specifically, select the image database as the pre-training set for pre-training, obtain the pre-training model, and retrieve all the monitoring images of the historical surveillance cameras in the security monitoring sub-area for enriching the ...

Embodiment 2

[0062] In this embodiment, on the basis of the above-mentioned embodiment 1, in order to better realize the present invention, further, in the step 4, the specific operation of sorting in the monitoring display queue according to the high and low intervals of the monitoring intensity level is: set the first queue and the second queue, first add the two surveillance cameras with the highest monitoring intensity level to the first queue and the second queue respectively, and then add the two surveillance cameras with the lowest monitoring intensity level to the first queue and the second queue respectively Then add the two surveillance cameras with the highest monitoring intensity level among the unsorted surveillance cameras to the third place in the first queue and the second queue respectively, and then add the two surveillance cameras with the lowest monitoring intensity level among the unsorted surveillance cameras Surveillance cameras are respectively added to the fourth of...

Embodiment 3

[0066] In this embodiment, on the basis of any one of the above-mentioned embodiments 1-2, in order to better realize the present invention, further, in the step 4, the preloading time D is set, when a certain monitoring camera is on the monitoring screen When there is still time D left in the display duration of the monitoring sequence, the images collected by the surveillance cameras ranked last in the monitoring sorting queue will be preloaded. When the display duration of a surveillance camera on the surveillance screen is exhausted, the preloaded images will be used for Replace display.

[0067] Working principle: as Figure 5 As shown in , assuming that three surveillance cameras in the detection display queue are used for display at one time, in Figure 5 in the first Figure 4 No. 1, No. 11, and No. 3 surveillance cameras in the detection display queue are displayed, and then when there is time D left for the display time of one of the surveillance cameras No. 1, 11,...

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 proposes a multi-mode security monitoring method based on deep learning image processing, which divides the scenes that need to be monitored according to the degree of importance, and regularly adjusts them according to the follow-up machine learning results, so as to realize the application of limited monitoring resources in the most needed monitoring At the same time, focus on the content that needs to be monitored. By setting the dangerous ID and the sub-dangerous ID, the people who are most likely to be monitored will be screened out. Focus on the people who need to be monitored the most. At the same time, by collecting the actual scenes that need to be monitored and pasting a large amount of external training data, the data of the pre-training set that meets the scenes to be monitored is enriched, thereby improving the accuracy of the final recognition. At the same time, through the training results, it is also possible to adjust and supervise the areas most likely to send abnormal behavior. Realize video online monitoring with lower equipment cost and overhead to achieve maximum efficiency.

Description

technical field [0001] The invention belongs to the technical field of computer image processing and monitoring, and in particular relates to a multi-mode security monitoring method based on deep learning image processing. Background technique [0002] With the rapid development of computer technology, image processing technology based on machine learning has become more mature and applied in various industries. Such as fast payment for face recognition, password lock for face recognition, fast intelligent recognition for image recognition, etc. But it is mainly applied in the image recognition level. Because the image database is quite rich, the accuracy of image recognition is quite high when supported by huge image data resources; however, the data volume of video databases in the world today is far lower than that of image data Therefore, the direct recognition of video is limited by database resources, and its technical maturity and accuracy are far lower than that of...

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): G06K9/00G06K9/32G06K9/62G06N3/04G06N3/12G06N3/08
CPCG06N3/08G06N3/126G06V40/166G06V20/52G06V10/25G06N3/045G06F18/23213G06F18/214
Inventor 古沐松范文杰孙珮凌游磊苗放
Owner CHENGDU 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