Dynamic capturing method and device for video key points

A dynamic capture, key point technology, applied in image analysis, image enhancement, instruments, etc., can solve problems such as target occlusion, and achieve the effect of improving quality, efficiently reading and displaying, and reducing the amount of calculation.

Pending Publication Date: 2022-05-10
TIANJIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In addition, there are still many technical difficulties in tracking under complex backgrounds, real-time tracking of multiple targets, and occlusion of targets during tracking, and researchers need to continue to study in depth

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
  • Dynamic capturing method and device for video key points
  • Dynamic capturing method and device for video key points
  • Dynamic capturing method and device for video key points

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0040] The embodiment of the present invention provides a kind of dynamic capture method of video key point, see figure 1 and figure 2 , the method includes the following steps:

[0041] 101: Perform processor operations such as grayscale processing and filtering processing on the image, and obtain the preprocessed image;

[0042]102: Modeling the background in the preprocessed image, and counting the gray value of the preprocessed image without moving objects into the interval, and then obtaining a statistically significant distribution interval of the initial background gray value;

[0043] Subsequent histogram equalization can be performed based on the distribution interval of the initial background gray value.

[0044] 103: Perform foreground detection on the image processed in step 102, that is, make a difference between the processed image and the background, obtain a background difference map, distinguish the moving foreground and background, and realize image segmen...

Embodiment 2

[0050] The preprocessing part in Embodiment 1, ie step 101, will be further introduced below in conjunction with specific examples, see the description below for details.

[0051] First, the original video frame

[0052] For the image preprocessing problem, the original video is firstly "divided into frames". Assuming that the total time of the acquired video is t, a key frame is extracted every time t', representing the motion state within t. After the key frame is extracted, the obtained n=t / t′ frame images are reassembled into a section of video and displayed in the software window. This process can be specifically described as the following steps:

[0053] 1) Manually obtain the target video and data format;

[0054] 2) Use the visual computing library function to divide the target video into frames, select a frame at a specific time interval, and finally obtain multiple video key frames;

[0055] Wherein, the visual computing library function is a built-in function of ...

Embodiment 3

[0097] The image background modeling in Embodiment 1, that is, step 102, will be further introduced in combination with specific examples, see the description below for details.

[0098] Because the background in the video is more complex, which affects the trajectory prediction effect, the background difference method is firstly used to determine the approximate range of the trajectory. Background subtraction method is a very commonly used moving target detection algorithm. It is mainly used to extract the foreground without changing the background. Its principle is to perform a difference operation between the current frame and the background, then set an appropriate threshold, and perform binarization through the threshold. The first step of the background subtraction method is to obtain the background image and model the background. Then the difference between the image of the current frame and the background is performed, and then an appropriate threshold is set for bin...

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 dynamic capture method and device for video key points, and the method comprises the steps: carrying out the modeling of a background in a preprocessed image, carrying out the preprocessing of an image without a moving target, obtaining the gray value of the image, carrying out the statistics according to intervals, and obtaining the distribution interval of the gray value of an initial background image; when a moving target appears, obtaining a background difference image through subtraction by using an early-stage background image, distinguishing a moving foreground and a moving background, and realizing image segmentation; based on an image segmentation result, a multi-scale regional convolutional neural network is added in the matching process of the original key points to improve the matching precision of the key points in the previous frame and the next frame of the video; and fitting the motion trails of the key points by using a least square method to realize dynamic capture of the key points in the video. The device comprises a processor and a memory. According to the invention, dynamic prediction of the key point motion trail of the oil pipeline is realized, the motion state of the oil pipeline is fitted, and real-time monitoring of the operation state of the oil pipeline is realized.

Description

technical field [0001] The invention relates to the technical field of moving object detection, in particular to a method and device for dynamic capture of video key points based on background difference method and least square method. Background technique [0002] With the development and progress of the times, especially the advent of the information age, video surveillance has basically been applied in various fields, such as: traffic information monitoring, public safety monitoring, key public place monitoring and security and other fields. [0003] At present, surveillance cameras are widely used, but most surveillance cameras cannot automatically complete the monitoring work, and only play the role of recording process. Specific monitoring tasks are still done by humans rather than machines. When the video surveillance work is carried out manually, the work pressure is high, and the concentration of the staff and the ability to handle special events are high. Relevan...

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): G06T7/246G06T7/254G06T7/90G06T7/194G06T5/00G06T5/20G06T5/40G06T7/11
CPCG06T7/246G06T7/254G06T7/11G06T7/194G06T7/90G06T5/002G06T5/007G06T5/40G06T5/20G06T2207/10016
Inventor 聂为之赵岳余杨苏育挺
Owner TIANJIN 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