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

Image analysis multi-target tracking method

A multi-target tracking and image analysis technology, applied in the field of target tracking, can solve the problems of location proximity principle tracking method failure, tracking interruption, etc.

Active Publication Date: 2019-10-08
南昌嘉研科技有限公司
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Obviously, the greater the distance the object moves between frames, the more limited this method will be
Additionally, tracking will necessarily be interrupted when objects are occluded
For example: a ball rolls under the arm of a billiard player and is blocked for 0.5 seconds, at this time, the position approach principle tracking method will definitely fail

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
  • Image analysis multi-target tracking method
  • Image analysis multi-target tracking method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0042] A kind of image analysis multi-target tracking method, see figure 1 ( figure 1 In , the circle means the starting point of the process, figure 1 The five processes in the middle and vertical directions each work on an independent thread, and the solid line arrows are the respective workflows of each thread, and the dotted line arrows are the message communication between each thread), including the following steps:

[0043] S1: Obtain the current original image of the target to be tracked; specifically include:

[0044] Use OpenCV's VideoCapture function to obtain the current BGR three-channel Mat image of the target to be tracked;

[0045] Or, use FFmpeg to read the current streaming media of the target to be tracked (such as the real-time image output by the network camera), and decompress the streaming media to obtain an AVFrame image of YUV three channels, and convert the AVFrame image into a Mat image.

[0046]Specifically, the method can be implemented on the O...

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 an image analysis multi-target tracking method. The method comprises the steps of obtaining a current original image of a to-be-tracked target; inputting the original image intoa preset background segmentation model, and calculating a background mask according to the original image; establishing a pyramid model according to the background mask, performing analyzing throughthe pyramid model to obtain moving objects in the background mask, and recording a position area of each moving object as a Spot object; and matching the Spot object in the background mask with a Target object library to obtain a target position of the Spot object, and finishing target tracking of the to-be-tracked target. According to the method, the same moving object can be identified in a series of continuous images, a plurality of moving objects in a picture are tracked at the same time, and the situation that the objects are temporarily shielded can be resisted.

Description

technical field [0001] The invention belongs to the technical field of target tracking, and in particular relates to an image analysis multi-target tracking method. Background technique [0002] Object recognition and tracking are common requirements in computer vision applications. The background segmentation method in the current computer vision library can use the changing rules of the image to identify moving objects. The basic working principle is: in a series of continuous images, the part where the pixel changes is the moving object (foreground), and the pixel remains unchanged. is the static object (background). However, when there are multiple objects moving in the image, the background segmentation method cannot recognize the same object in each image. For example, if a video of billiards is input into the background segmentation method, this method can effectively identify the rolling ball in each frame, but it cannot answer where a certain ball in the previous ...

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
IPC IPC(8): G06T7/246G06T5/00
CPCG06T7/246G06T2207/20016G06T5/70
Inventor 林嘉
Owner 南昌嘉研科技有限公司