Effective Detection of Moving Objects Using Complex Background Subtraction
- R.Kalai selvi
- D. Durai kumar
video surveillance, CWT,approximate median,sub band differencing, morphological operation
The detection of moving object is based on background subtraction under complex wavelet transform (CWT) domain for video surveillance system. Initially it starts with background initialization by choosing start frame or taking initial few frames with approximate median method. Then the CWT is applied to both low and high frequency sub bands. In order to remove some unwanted pixels, morphological operation is performed for object edge smoothness. The proposed approach has some advantages of background noise insensitiveness and invariant to varying illumination or lighting conditions. After the object detection, performance of method is measured through metrics such as sensitivity, accuracy, correlation. This object detection also helps to track detected object using connected component analysis.
R.Kalai selvi, D. Durai kumar. "Effective Detection of Moving Objects Using Complex Background Subtraction".INTERNATIONAL JOURNAL OF ENGINEERING DEVELOPMENT AND RESEARCH ISSN:2321-9939, Vol.3, Issue 2, pp.52-57, URL :https://rjwave.org/ijedr/papers/IJEDR1502011.pdf
Volume 3 Issue 2
Pages. 52-57
