(3.238.96.184) 您好!臺灣時間:2021/05/18 16:17
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:賴東賢
研究生(外文):TUNG-HSIEN LAI
論文名稱:運用邊緣流量及蟻群最佳化於視訊中移動物件之主動式輪廓追蹤研究
論文名稱(外文):Active Contour Tracking of Moving Objects UsingEdge Flows and Ant Colony Optimization in VideoSequences
指導教授:張元翔張元翔引用關係
學位類別:碩士
校院名稱:中原大學
系所名稱:資訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2008
畢業學年度:96
語文別:英文
論文頁數:43
中文關鍵詞:物件追蹤邊緣流量蟻群最佳化主動式輪廓模型
外文關鍵詞:edge flowobject trackingant colony optimizationactive contour model
相關次數:
  • 被引用被引用:0
  • 點閱點閱:95
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
物件的分割和追蹤在視訊應用中是相當重要的技術。本論文提出一個視
訊中移動物件的主動式輪廓追蹤系統,其方法包含前處理及物件輪廓分割兩
部分。前處理的目的在於擷取物件初始輪廓,物件輪廓分割的目的則在於使
移動物件的輪廓更貼近物件真實的邊界。本系統結合邊緣流量及蟻群最佳化
演算法以增進系統輪廓能量的收斂效率。經過實驗結果驗證,系統自動產生
的分割結果與手動分割產生的結果,其平均誤差已達到小於一個像素距離之
精確度。總結而言,本系統特別適用於動態物件分割與追蹤,且不需在場景
中建立背景模型。預計本系統將可以應用在物件導向的視訊編碼,或是其他
研究如視訊監控系統的行為分析等。
Object segmentation and tracking are important techniques in video
applications. In this paper, we present a novel system for active contour tracking
of moving objects in video sequences. Our method includes preprocessing to
identify an initial object contour, and object contour segmentation to refine the
contour of the moving object. The edge flows and ant colony optimization are
incorporated to improve the efficiency during system convergence. Experimental
results demonstrated that our system has achieved the automatic segmentation
accuracy of < 1 pixel on average as compared with manual segmentation results.
In summary, our system is particularly useful in segmenting and tracking a
moving object without constructing a background model for a video scene.
Ultimately, our system could be used in object-based video coding or other
analysis such as behavior analysis in video surveillance systems.
Abstract in Chinese ………………………….……………………………...I
Abstract ……………………..…………………………………………….. II
Acknowledgement ………….……………………………………………. III
Contents …………………….……………………………………………. IV
List of Figures ……………….……………………………………………VI
List of Tables ………………………………………………………..…… IX
1. INTRODUCTION ……………………………………………………...1
2. METHOD ….……..…………………………………………………….4
2.1 Preprocessing ………………………………………………………….6
2.1.1 Region Growing ….…………………………………………………6
2.1.2 Motion Image Detection ………...………………………………….7
2.1.3 Background Edge Removal …......………………………………….8
2.1.4 Motion Boundary Extraction ……………………………………….9
2.2 Object Contour Segmentation ……………………………………….10
2.2.1 Energy Calculation ………………………………………………...11
2.2.2 Search Space Construction ………………………………………...12
2.2.3 Pheromone Definition ……………………………………………..14
2.2.4 Transition Probability Construction ……………………………….14
2.2.5 Concave and Convex Processing ……………………………………15
2.2.6 Pheromone Update ………………………………………………….16
2.2.7 Energy Convergence ………………………………………………..17
2.3 System Evaluation …………………………………………………….19
3. Results …………………………………………………………...........20
4. Conclusion …………………………………………………………….29
References ……………………………………………………………........31


List of Figures
Fig.1. A simplified flow chart in our system framework for active contour
tracking of moving objects in video sequences……………………….5
Fig.2. A simplified flow chart of the preprocessing in our system…………..5
Fig.3. A simplified flow chart of the object contour segmentation in our
system ………………………………………………………………...5
Fig.4. An example of the region growing, where (a) contains the contour and
the geometric center in the 28th frame from Frank.avi shown in ‘red’;
and (b) is the resulting image after region growing……….…………..7
Fig.5. An example of the motion image detection, where (a) is the original
28th frame from Frank.avi; (b) is the original 29th frame from Frank.avi;
(c) is frame difference between 28th and 29th frame after thresholding;
and (d) is the resulting image containing the moving object.................8
Fig.6. An example of the background edge removal, where (a) is the resulting
image edge after Canny edge detection in the 29th frame in Frank.avi
and (b) is the same image with background edges removed………… 9
Fig.7 An example of motion boundary extraction, where (a) is the motion
boundary of 29th frame in Frank.avi; and (b) shows part of the motion
boundary in (a)…………………………………………………….. 10
Fig.8. (a) concave case; (b) convex case. The ideal object contour is obtained
by edge detection, and vi-1, vi, and vi+1 are represent three consecutive
pixels in the current object contour ………………………………...16
Fig.9. An example of the object contour segmentation for the 29th frame of
Frank.avi, after preprocessing; where (a) is initial object contour after
preprocessing; (b) is the object contour after the 1st iteration; (c) is the
object contour at the 2nd iteration; and (d) is the resulting object
contour after convergence…………………………………..……….18
Fig.10. The corresponding energy as computed during system convergence
for the object contour segmentation shown in Fig. 9……………..18
Fig.11. Segmentation results of object contour in the (a) 29th, (b) 39th, (c) 49th,
and (d) 59th frame from Frank.avi, where the segmented object
contour is highlighted in ‘red’……………………………………..21
Fig.12. Segmentation results of object contour in the (a) 2nd, (b) 42nd, (c) 82nd
and (d) 122nd frame from Akiyo.avi, where the segmented object
contour is highlighted in ‘red’……………………………………..22
Fig.13. Segmentation results of object contour in the (a) 2nd, (b) 62nd, (c) 82nd
and (d) 102nd frame from vid01.avi, where the segmented object
contour is highlighted in ‘red’……………………………………. 22
Fig. 14. Segmentation results of object contour in the (a) 5th, (b) 15th, (c) 35nd
and (d) 45nd frame from vid02.avi, where the segmented object
contour is highlighted in ‘red’……………………………………..23
Fig. 15. Results of object contour segmentation using the GVF snake and the
proposed approach, where (a) is original image sequences; (b) is the
object contour using the GVF snake; and (c) is the object contour
using the proposed approach. The images from left to right are the
29th, 39th, 49th, 59th, 69th, 79th and 89th frames in Frank.avi…………26



List of Tables
Table 1. System parameters used in the active contour tracking of moving
object..……………………………………………………………20
Table 2. The RMS error of selected frames between the manual and utomatic
segmentation results in Frank.avi. The mean and standard deviation
of the RMS error are also shown (Unit: pixels)…………………..23
Table 3. The RMS error of selected frames between the manual and
automatic segmentation results in Akiyo.avi……………………...24
Table 4. The RMS error of selected frames between the manual and
automatic segmentation results in vid01.avi………………………24
Table 5. The RMS error of selected frames between the manual and
automatic segmentation results in vid02.avi………………………25
Table 6. A Comparison of the execution time of the GVF snake and the
proposed approach using the image sequences in Frank.avi………28
1. Xiang, T., Gong, S.: Video behavior profiling for anomaly detection. IEEE
Trans. Pattern Anal. Mach. Intell. 30, 893--908 (2008)
2. Gupta, A., Mittal, A., Davis, L.S.: Constraint integration for efficient
mutiview pose estimation with self-occlusions. IEEE Trans. Pattern Anal.
Mach. Intell. 30, 493--506 (2008)
3. Sundaramoorthi, G., Yezzi, A., Mennucci, A.C.: Coarse-to-fine
segmentation and tracking using sobolev active contours. IEEE Trans.
Pattern Anal. Mach. Intell. 30, 851--864 (2008)
4. Zhao, T., Nevatia, R., Wu, B.: Segmentation and tracking of multiple
humans in crowded environments. IEEE Trans. Pattern Anal. Mach. Intell.
30, 1198--1211 (2008)
5. Han, B., Comaniciu, D., Zhu, Y., Davis, L.S.: Sequential kernel density
approximation and its application to real-time visual tracking. IEEE Trans.
Pattern Anal. Mach. Intell. 30, 1186--1197 (2008)
6. Briassouli, A., Ahuja, N.: Extraction and Analysis of multiple periodic
motions in video sequence. IEEE Trans. Pattern Anal. Mach. Intell. 29,
1244--1261 (2007)
7. Comaniciu D., Meer, P.: Mean shift: A robust approach toward feature
space analysis. IEEE Trans. Pattern Anal. Mach. Intell. 24, 603--619
(2002)
8. Shi, J., Malik, J.: Normalized cuts and image segmentation. IEEE Trans.
Pattern Anal. Mach. Intell. 22, 888--905 (2000)
9. Xu, N., Ahuja, N.: Object contour tracking using graph cuts based active
contours. IEEE Proceedings International Conference on Image Processing,
vol. 3, pp. III-277--III-280 (2002)
10. Kass, M., Witkin, A., Terzopoulos, D.: Snakes: Active contour models.
International Journal of Computer Vision. 1, 321--331 (1988)
11. Stauffer, C., Grimson, W.: Adaptive background mixture models for
real-time tracking. IEEE Computer Society Conference on Computer
Vision and Pattern Recognition, vol. 2, pp. -252 (1999)
12. Elgammal, A., Duraiswami, R., Hardwood, D., Davis, L.S.: Background
and foreground modeling using nonparametric kernel density estimation
for visual surveillance. IEEE Proceeding., vol. 90, no. 7, pp. 1151-1163
(2002)
13. Canny edge detection tutorial,
http://www.pages.drexel.edu/~weg22/can_tut.html (2008)
14. Gonzalez, R.C., Wood, R.E.: Digital Image Processing 2nd Edition.
Prentice Hall, New Jersey (2002)
15. Dorigo, M., Maniezzo, V., Colorni, A.: The ant system: Optimazation by
a colony of cooperating agents. IEEE Trans. Systems, Man, and
Cybernetics. 26, 29--41 (1996)
16. Wang, X.N., Feng, Y.J., Feng, Z.R.: Ant colony optimization for image
segmentation. IEEE Proceeding International conference on Machine
Learning and Cybernetics, vol. 9, pp. 5355-5360 (2005)
17. Ma, W.Y., Manjunath, B.S.: Edge flow: A framework of boundary
detection and image segmentation. IEEE Proceedings. Computer Society
Conference on Computer Vision and Pattern Recognition, pp. 744-749
(1997)
18. Chenyang, X., Prince, J.L.: Gradient vector flow: A new external force
for snake. IEEE Proceedings. Computer Society Conference on Computer
Vision and Pattern Recognition, 66--71 (1997)
19. Litvin, A., Konrad, J., Karl, W.C.: Probabilistic video stabilization using
kalman filtering and mosaicking. Proceedings of SPIE-IS&T Electronic
Imaging, SPIE., vol. 5002, pp. 663-674 (2003)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top