Single change detection-based moving object segmentation by using Daubechies complex wavelet transform

Research in motion analysis is a challenging field and it has a variety of video surveillance applications. For any video surveillance application, background detection and removal plays an important role in segmentation of the moving objects. This study proposes a new method for segmentation of the moving object, which is based on single change detection applied on Daubechies complex wavelet coefficients of two consecutive frames. The authors have chosen Daubechies complex wavelet transform as it is shift invariant and has a better directional selectivity as compared with real-valued wavelet transforms. Single change detection is a method to obtain video object plane by inter-frame difference of two consecutive frames, and it provides automatic detection of appearances of new objects. The proposed method does not require any other parameter except wavelet coefficients. Segmentation results of the moving objects after applying the proposed method are compared with those obtained after applying other spatial and wavelet domain segmentation methods in terms of visual performance and a number of quantitative measures viz misclassification penalty, relative position-based measure, structural content, normalised absolute error and average difference and the proposed method is found better than the other methods.