Modification of medium transfer detector for target tracking with variable radiation pattern
Subject Areas : GeneralPayman Moallem 1 , عليرضا معمارمقدم 2 , جواد عباس پور 3 , masoud kavoshtehrani 4
1 -
2 -
3 - دانشکده فنی و مهندسی دانشگاه اصفهان
4 -
Keywords:
Abstract :
One of the conventional methods in the field of image tracking of non-rigid targets is to use a repetitive procedure called average transfer in determining the central mode position of the target. The display of the target in the average transfer tracker is based on the histogram of spatial interpolation feature with a direction-independent kernel. The most critical challenge in the medium transfer detector is the kernel scaling. So far, no efficient and perfect method to adjust or adapt the kernel dimensions when the target dimensions change has been presented. Another problem of the average transmission detector occurs when facing a target with a variable radiation pattern. In this article, with the approach of solving these problems, the average transmission tracking algorithm with strong adaptive scaling is presented, while it solves the problem of the average transmission algorithm in the face of changes in the radiation pattern of the target by adapting the target model in each frame. In the proposed method, the dimensions of the window in the next frame are set first by using the power calculation method resulting from the time-space derivatives of the intensity of the image pixels. Then, the results of the window scaling in the next frame are applied to the average transfer detector. The results show that the use of the proposed algorithm, while reducing the target positioning error in comparison with the standard average transfer algorithm, also shows a significant efficiency against the changes of contrast 2 and target radiation pattern.
مراجع
[1]. H. Tahvilian, P. Moallem, A. Monadjemi, Balloon Energy Based on Parametric Active Contour and Directional Walsh-Hadamard Transform and its Application in Tracking of Texture Object in Texture Background, EURASIP Journal on Advances in Signal Processing, Springer, 2012:253, Dec. 2012.
[2]. A.I. Comport, E. Marchand, F. Chaumette,Efficient model-based tracking for robot vision, Advanced Robotics, Vol.19, No. 10,pp. 1097-1113, 2005.
[3]. P. Salembier and F. Marques, Region-based representations of image and video: Segmentation tools for multimedia services, IEEE Transactionson Circuits, Systems and Video Technology,Vol. 9,pp. 1147-1169, 1999.
[4]. M. Kass, A. Witkin, and D. Terzopoulos, Snakes: Active contour models, in 1stInternational Conference on Computer Vision, UK, pp. 259-269, 1987.
[5]. J. Verestoy and D. Chetverikov, Comparative performance evaluation offour feature point tracking techniques, 22nd workshop of the Austrian pattern recognition group, Austria, pp. 255-263, 1998.
]6[. ع. معمارمقدم، پ. معلم، رديابي سريع و كارآمد هدف در رشته تصاوير ويدئويي به كمك انباره سازي خطاهاي تطبيق¬دهي نقاط گوشه¬اي آن، چهارمین کنفرانس ماشین بینایی و پردازش تصویر ایران، جلد دوم، صفحات 981 الی 990، دانشگاه فردوسی مشهد، 1385.
[7]. D. Comaniciu, V. Ramesh, P. Meer, Kernel-based object tracking, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, No. 5,pp. 564-577, 2003.
[8]. D.Comaniciu, P. Meer, Mean shift: A robust approach toward feature space analysis, IEEETransactions on Pattern Analysis and Machine Intelligence, Vol. 24, No.5, pp.603-619, 2002.
[9]. N.M. Artner, W. Burger, A Comparison of mean shift Tracking Methods, 12thCentral European Seminar on Computer Graphics, Austrian, pp. 197-204, 2008.
[10]. P.Moallem, A.Memarmoghaddam, M. Ashourian,Robust and Fast Tracking Algorithm in video Sequences by Adaptive Window Sizing Using a Novel Analysis on Spatiotemporal Gradient Powers , Journal of Circuits, Systems, and Computers,Vol. 16, No. 2, pp. 305-317, 2007.
[11]. R. Collins, Mean-shift blob tracking through scale space,IEEE International Conference on Computer Vision and Pattern Recognition, USA, Vol. 2, pp. 234-240, 2003.
