Unlock instant, AI-driven research and patent intelligence for your innovation.

Real-time video identification method based on track prediction and size decision

A trajectory prediction, real-time video technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problem of reporting to the upper computer or business management server, to meet real-time requirements, improve accuracy, and reduce errors the effect of the possibility

Active Publication Date: 2019-08-23
ZHEJIANG UNIVERSITY OF MEDIA AND COMMUNICATIONS
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the limitation of storage resources and communication resources of video surveillance equipment, the equipment will not report all the results to the host computer or business management server, and can only selectively report the recognition results according to the preset parameters

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
  • Real-time video identification method based on track prediction and size decision
  • Real-time video identification method based on track prediction and size decision
  • Real-time video identification method based on track prediction and size decision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039]A real-time video recognition method based on trajectory prediction and size decision-making in this embodiment can be applied to a variety of different intelligent video surveillance front ends, including license plate recognition systems based on embedded cameras, vehicle face recognition systems, and vehicle feature recognition systems , Pedestrian feature recognition system, etc. The present invention will be further described in detail below by taking license plate recognition as an example with reference to the accompanying drawings.

[0040] Such as figure 1 Shown is an embodiment of a real-time video recognition method based on trajectory prediction and size decision-making on an embedded license plate recognition camera. Each blue rectangular frame in the figure represents the position where the corresponding license plate was recognized once, and the leftmost rectangular frame in the figure is the first recognition; the car enters the screen from the upper lef...

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 provides a real-time video identification method based on track prediction and size decision, and the method comprises the steps: employing a video identification algorithm to identify the characteristics of an object captured by a video shooting tool, and outputting the identified target information; checking whether the identified target information meets a preset confidence coefficient requirement or not; performing stability inspection; then predicting the motion track of the target object; and when the motion track of the target object, the cache number and the confidence coefficient of the identification result accord with preset service triggering conditions, integrating the current identification result and the result cached in the system, and optimally selecting a final stable identification result as output. On the basis that the video recognition core algorithm is not changed, the stability and accuracy of the recognition result are improved through the optimization method of the recognition result for multiple times, and the high efficiency and accuracy of video monitoring and entrance and exit management are improved.

Description

technical field [0001] The invention relates to the optimization of output results of a real-time video recognition system, in particular to a method for real-time video recognition based on track prediction and size decision-making. Background technique [0002] With the rapid development of artificial intelligence-related technologies, more and more intelligent embedded video front-ends have appeared in the traditional video surveillance industry. These video front-ends are not only responsible for video data collection, but also begin to play the role of object perception and intelligent recognition. In an embedded video surveillance device, it usually includes an algorithm module for object recognition, a cache management module for recognition results, a local storage module for recognition results, and a communication module that supports submission of recognition results. Among these modules, the algorithm module of target object recognition is responsible for analyz...

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): G06K9/00G06K9/32
CPCG06V20/41G06V20/46G06V20/63G06V20/625
Inventor 王忠
Owner ZHEJIANG UNIVERSITY OF MEDIA AND COMMUNICATIONS