以多目標優(yōu)化的輪廓定位分割視頻對象

宋立鋒; 韋崗; 王群生

以多目標優(yōu)化的輪廓定位分割視頻對象

計量
- 文章訪問數: 2468
- HTML全文瀏覽量: 136
- PDF下載量: 702
- 被引次數: 0
出版歷程
- 收稿日期: 2000-11-24
- 修回日期: 2002-03-25
- 刊出日期: 2002-11-19

Segment video objects by contour locating of multiobjective optimization

摘要

摘要: 該文提出一種用于半自動分割視頻對象的對象跟蹤方法,它以方塊均方色差和圖像像素邊緣強度構造目標函數,把對象輪廓點的運動估值和補償表達為用帶容錯的分層排序法實現多目標優(yōu)化,從而確定與參考幀對象輪廓點對應的當前幀像素為輪廓點;以邊緣強度最大為目標將這些輪廓點連接成閉合曲線,從而得到當前幀的對象分割掩膜,用這種方法處理多個視頻測試序列得到滿意結果。
- 視頻對象分割; 半自動視頻對象分割; 對象跟蹤; 多目標優(yōu)化; 運動估值和補償
Abstract: An object tracking method for semiautomatic video object segmentation is proposed in this paper. An objective function is constructed with mean square color error of blocks and edge strength of pixels. Then a multilevel ordering method with permissive error is used to accomplish the programming of this objective function so as to realize motion estimation and compensation of contour pixels. Thus the pixels in the current frame, which correspond to the contour pixels in the reference frame, are found out. As the contour pixels in the current frame, these pixels are connected to form a close curve by an objective of maximal edge strength. Thus the segmentation mask of the current frame is generated. The experimental results of this method on several test sequences are satisfactory.

HTML全文

參考文獻(1)

R. Mesh, M. Wollborn, A noise robust method for 2D shape estimation of moving objects in video sequences considering a moving camera, Signal Processing, 1998, 66(2), 203-217.[2]A. Neri, et al., Automatic moving object and background separation, Signal Processing, 1998,66(2), 219-232.[3]M. Kim, et al., A VOP generation tool: automatic segmentation of moving objects in image sequences based on spatio-temporal information, IEEE Trans. on CAS VT, 1998, 9(8), 1216-1226.[4]C. Toklu, et al., Semiautomatic video object segmentation in the presence of occlusion, IEEE Trans. on CAS VT, 2000, 10(4), 624-629.[5]D. Zhong, S. F. Chang, An integrated approach for content-based video object segmentation and retrieval, IEEE Trans. on CAS VT, 1998, 9(8), 1259-1268.[6]C. Gu, M. C. Lee, Semiautomatic segmentation and tracking of semantic video objects, IEEE Trans. on CSVT, 1998, 8(5), 572-584.[7]ISO/IEC JTC1/SC29/WG11 14496-2 FDAM 1. Information Technology-Coding of Audio-Visual Objects-Part 2: Visual/Amendment 1. N3056, Maui, Dec. 1999, Annex F: 452-462.[8]F. Leymarie, M. D. Levine, Tracking deformable objects in the plane using an active contour model, IEEE Trans. on PAMI, 1993, 15(6), 617-634.[9]趙雪春,戚飛虎,用可變形模板進行基于內容的圖像分割算法,電子學報,2000,28(4),69-72.[10]A.M.Tekalp,Digital video processing,Prentice Hall,Inc.,1998,Chapter 7:118.[11]林銼云,董加禮,多目標優(yōu)化的方法與理論,長春,吉林教育出版社,1992,第四章,83-87. ［12］J.Canny,A computational approach to edge detection,IEEE Trans.on PAMI,1986,8(6),679-698.[12]呂鳳軍,數字圖像處理編程入門-做一個自己的Photoshop,北京,清華大學出版社,1999,第6章,第7章.[13]J. Liu, Y. H. Yang, Multiresolution color image segmentation, IEEE Trans. on PAMI, 1994,16(7), 689-700.[14]ISO/IEC JTC1/SC29/WG11, ETRIs user-assisted video object segmentation tool. M4479,Seoul, Mar. 1999.

施引文獻

資源附件(0)

訪問統(tǒng)計