多重物件追蹤技術在運動上的運用

Multiple Object Tracking Techniques in Sports

譚家棟(Chia-Tong Tang)

指導教授 ： 葛煥昭

淡江大學/工學院/資訊工程學系碩士班/碩士(2005年)

https://doi.org/10.6846/TKU.2005.00512

摘要

近年來，由於各式工具取得的便利，數位化資料的儲存方式已漸漸的由靜態影像轉為以動態連續影像紀錄。隨著各項學術研究的發展，可取得的資料也越來越多，但如何快速的從大量資料中選取我們所需要的資訊以及如何從現有的數位化資料分析出我們所需的結果亦是目前研究發展的重點。
目前在物件追蹤的領域裡，有許多學者提出了各種用來預測與判斷比對物件的方法來加強物件追蹤技術的正確性其中以KalmanFilter 和 Condensation演算法較為被廣泛使用，前者為利用物件移動量上的資訊來預測物件可能會出現的位置，而後者則為以物件色彩分佈的資訊來比對物件。
在本篇論文中，我們提出了物件追蹤的技術，其中整合了許多相關的影像處理技術，例如：影像灰階化(Gray Level)、Closing運算等，來協助系統能分析出較為正確的前景物件資訊，以及使用物件
取(Object Extraction)、物件分析(Object analysis)的技術來擷取出每一個物件在目前的時間點上所擁有的特徵值與各項資訊，接著並利用以物件的移動向量(motion vector)與色彩分佈上的資訊來正確的追蹤物件。

關鍵字

物件追蹤

並列摘要

In the field of object tracking, there have many scholars proposed methods which is to predict and match object in each frame for enhanced the correctness of the object tracking technique. Kalman filter and Condensation algorithm were used generally. The former is used object’s motion information to predict object’s position in next frame and the latter is used object’s information of color distribution to match object in next frame.
In this project, we proposed a technique which combine many image processing methods like gray-level, closing operator, median filter, etc. to assist system can analyze more correct information of foreground object. Then we used object segmentation and object analysis technique to extract every object’s information and feature in present time and use object’s motion vector and the information of color distribution to tracking object correctly.

並列關鍵字

object tracking ； object segmentation

參考文獻

[3] Cheng-Yuan Tang, Yi-Ping Hung, Zen Chen, “Automatic Detection and Tracking of Human Heads Using Active Stereo Vision System,” International Journal of Pattern Recognition and Artificial Intelligence, 14(2): 137-166, March 2000

[4] Wen-Nung Lie and Ruey-Lung Chen, (Aug. 2001) “Tracking moving objects in MPEG-compressed videos,” Proc. Of IEEE Int’l Conf. On Multimedia and Expo, ICME-2001, Tokyo, Japan.

[5] L.Marchesotti, S. Piva, C. Regazzoni, "A dynamic model integrating colour and shape information for objects tracking in conditions of occlusion", Taipei The Grand Hotel, Taipei, Taiwan, June 27th - 30th, 2004

[6] Zhongxiang Luo, Yueting Zhuang, Feng Liu, Yunhe Pan, “Multiple animated characters motion fusion”. Volume 13, The Journal Of Visualization and Computer Animation,275-285, Number 5, December 2002

[7] Chih-Yi Chiu, Shih-Pin Chao, Ming-Yang Wu, Shi-Nine Yang, and Hsin-Chih Lin "Content-Based Retrieval for Human Motion Data", to appear on Journal of Visual Communication and Image Representation. pp. 605-612, Kinmen, Taiwan, R.O.C., Aug. 17-19, 2003.