Moving Object Detection and Tracking based on Correlation and Wavelet Transform Techniques to Optimize Processing Time

Moving object detection and tracking has attracted significant research interest in recent years. It has many application such as traffic monitoring, military, medicine and biological sciences etc. detection and tracking of moving object in video sequences can offer significant benefit to motion analysis.

In this paper, two algorithms for moving object detection and tracking are proposed. In the first algorithm cross correlation is used and in second algorithm, Wavelet transform based technique is used for detecting and tracking of the moving object. Cross Correlation is applied to each sub frame after taking the difference between the two frames. The minimum value of Cross Correlation indicates the presence of moving object. Location of the moving object is obtained by performing component connected analysis and morphological processing. After that the centroid calculation is used to track the moving object. The second algorithm is based on wavelet decomposition (i.e. multi resolution analysis) for the detection of moving object and then centroid calculation is used to track that object.

Qualitative and quantitative results in terms of Detection Rate (DR), False Alarm Rate (FAR) and average processing time per frame are given. The proposed algorithms are compared with the established methods based on simple difference and background subtraction. Comparison shows that methods based on cross correlation and wavelet decomposition outperform the previous methods in Success Rate. Also, it is observed that processing time of cross correlation based method is better than wavelet decomposition based method.