Abstract—In this paper, a new algorithm to detect a moving target (foreground moving target) in an image with variable and complex background is presented. It should be mentioned that background in this paper is an image containing moving objects. In this method, first the edges of all objects in the image are extracted using Canny filter and GVF active contour then the target is detected using wavelet transform. The boundary of target is extracted using a new boundary detection method. As a result, the background is removed and the target is extracted. To detect the type of motion (mode of the motion) and also to intellectualize the algorithm the neural network is used. Finally the algorithm detects the type of motions. Multiplicity and diversity of complexities in the background of the images which the algorithm is applied on them, variability of the background, using Wavelet Transform to detect the desired target, proper using of GVF Active Contour, presenting and applying a new method for boundary detection and, generally the novelty and ability of our algorithm in detection of target in many different images, and also new application of all stages of this algorithm , are all reasons that discriminate our algorithm from all the methods presented in the same area.
Index Terms—Boundary detection, background removal, canny filter, GVF, motion detection, wavelet transform.
Farid Jafarian was with Department of Electrical Engineering, Sepahan Isfahan Institute of Science and Technology, Isfahan, Iran. He is now with the Department of Electronics and Communications Engineering, University of Birjand, Birjand, Iran. (e-mail: email@example.com).
Raheleh Kafieh is with the Department of Biomedical Engineering, Isfahan University of Medical Science, Isfahan, Iran. (e-mail: address:firstname.lastname@example.org).
Cite: Farid Jafarian and Raheleh Kafieh, "New Algorithm to Detect Moving Target in an Image with Variable and Complex Background Using Wavelet Transform," International Journal of Computer Theory and Engineering vol. 5, no. 1, pp. 71-76, 2013.