Research Article
Research on Real Time Tracking Method of Multiple Moving Objects Based on Machine Vision
@INPROCEEDINGS{10.1007/978-3-031-50552-2_11, author={Yuan Wang}, title={Research on Real Time Tracking Method of Multiple Moving Objects Based on Machine Vision}, proceedings={Advanced Hybrid Information Processing. 7th EAI International Conference, ADHIP 2023, Harbin, China, September 22-24, 2023, Proceedings, Part IV}, proceedings_a={ADHIP PART 4}, year={2024}, month={3}, keywords={Machine Vision Multiple Moving Targets Real-Time Tracking Machine Vision Acquisition Unit Hsv Color Characteristics Apparent Characteristics Km Bipartite Graph Matching Algorithm}, doi={10.1007/978-3-031-50552-2_11} }- Yuan Wang
Year: 2024
Research on Real Time Tracking Method of Multiple Moving Objects Based on Machine Vision
ADHIP PART 4
Springer
DOI: 10.1007/978-3-031-50552-2_11
Abstract
In the process of real-time tracking of multiple moving targets, there is a big gap between the tracking effect and the ideal effect due to the influence of the objective environment state. Therefore, a real-time tracking method of multiple moving targets based on machine vision is proposed. A machine vision acquisition unit including lighting device, camera, image acquisition card, processing system and control actuator is constructed. At the same time, in order to avoid the impact of outdoor extreme weather on image acquisition effect, MF-DSC03 with waterproof function and automatic light compensation function is used as the camera head device. In the phase of real-time tracking of multiple moving targets, based on HSV color features and apparent features in real-time video image data of multiple moving targets captured by machine vision, iterative tracking of the state of multiple moving targets is realized by means of KM bipartite graph matching algorithm. In the test results, the design method outperforms the control group in multi-target tracking performance on different data sets, showing an ideal tracking effect.
