Moving Objects Detection And Tracking Techniques Based Optical Flow

Moving objects detection and tracking is an important topic in the field of computer vision, image and video processing and pattern recognition, its main objective is to obtain the motion parameters of moving targets (such as position, velocity, acceleration, etc.) as well as the moving trajectory in order to carry out further processing and analysis, to understand the behavior of moving objects in order to complete higher level tasks. Moving objects detection and tracking technology based optical flow can direct access to the motion parameters of moving targets with high precision, but has not been able to in the real-time progress needs. In this paper, movement of vehicles on the road is moving target for the in-depth study.This paper researches optical flow method based on the moving target detection and tracking technology. There major parts: image preprocessing; moving target detection technology based on optical flow; tracking of moving targets based on optical flow. Innovation and research are as follows:First of all, in the image pre-processing module, this paper introduces a removing noise method based on optical flow and the experimental results show that this new removing noise method is more effective in veracity.Secondly, in the moving target detection module, the core of this paper, propose two new adaptive fuzzy clustering algorithms. The image segmentation and moving target detection experiment results show that the two adaptive fuzzy clustering algorithms have high accuracy and effectiveness.Finally, in moving object tracking module, this paper combines optical flow-based model and kalman filter method. Experimental results show that algorithm meets the premise of real-time and has a certain degree of accuracy. But there is undetected, flase and multi-target overlap and so on. How to improve the detection and tracking algorithm is the subject of further study.