Object Motion Detection and Data Processing in Large-Scale Particle Image Velocimetry

Abstract

Large-Scale Particle Image Velocimetry (LSPIV) is a non-intrusive imaging measurement method of the river surface flow velocity. However, in natural environments with the impact of the shadow and strong light, small objects on the water surface can only be seen as tiny points with little pixels, which degrade the performance of existing PIV methods in the lab in the view of the requirements of continuous tracer measurement and tracking. We propose an algorithm for tracer particles detection and image data processing of complex water surfaces. A combination of Top-Hat transform and adaptive threshold segmentation is utilized to detect the floating small objects first followed by a pre-matching based on the shape features of a single particle. Finally, a fine matching is carried out based on the similarity among the particles and motion distance without error vectors. The experimental results show that the proposed method has a higher detection rate for the small target detection in the river environment and can further improve the estimation accuracy of the tracer particle motion vector. The proposed method can solve the problem of the detection and estimation of motion vector of tiny targets under complex surface optical environment.

KeyWords:

• Target detection
• top-hat transform
• motion vector estimation
• particle tracking velocimetry

Acknowledgement

This work is supported by the National Natural Science Foundation of China (No. 61273251, 61401195, 61374019). Meanwhile, this paper is got support from Lan Yao, the graduate student of Hohai University. Thanks for her help very much.