動態背景之移動物偵測及追蹤

Moving Object Detection and Tracking under Dynamic Background

潘宗聖(Zong-Sheng Pan)

指導教授 ： 吳明芳

崑山科技大學/工程學院/電機工程研究所/碩士(2009年)

摘要

本論文主要利用固定式攝影機來實現物件偵測及移動物追蹤。為達此目的，最簡單的方法是運用背景相減後所產生的差異量，但此方法只適合背景無動態物出現的情況。若背景中含有重複性動態物，則在偵測物件時，將導致無法正確判斷移動物件，所以本研究提出重複性背景動態物濾除技術來達成正確判斷移動物件的目標。
本文在重複性背景動態物濾除技術上，首先藉由統計的方式建立靜態背景影像，接著利用背景動態物中擁有重複出現的特性來建立動態背景，最後將動態背景區塊覆蓋於背景相減影像，達成背景動態物濾除的目的。
在物件偵測及移動物追蹤方面，利用背景相減法來得知前景資訊，接著運用重複性背景動態物濾除技術，便可將因重複性背景動態物所產生的差值區塊濾除，其它剩下來的點或區塊即為真正移動物件資訊。

關鍵字

物件偵測 ； 移動物件追蹤 ； 背景相減法 ； 動態背景

並列摘要

In this thesis, the fixed digital camera is applied to the moving object detection and tracking. In order to achieve this purpose, the easiest way is to utilize the variation between the current and the background images. But, it only suits the image with non-stationary background. If there is any repeated-dynamic object existing in the background, then the moving object is unable to judge correctly when we are doing the object detection. So, an eliminating technique of repeated-dynamic background is proposed in this thesis to achieve the goal that the moving object can be judged correctly.
In the eliminating technique of repeated-dynamic background, the statistical method is firstly used to build the static background image, and then the dynamic background is built by the characteristic of repeated appearance of dynamic object in background. Finally the dynamic background region is covered in background subtracting image to achieve the purpose of dynamic object eliminating of background.
In moving object detection and tracking aspect, after the foreground information is obtained by using the background subtracting method, the eliminating technique of repeated-dynamic background is applied to eliminate the differential image region which is produced by repeated-dynamic object of background, and the remaining points or regions are the true information of moving objects.

並列關鍵字

Object detection ； Moving object tracking ； Background subtraction ； Dynamic background

參考文獻

[1] E. Durucan and T. Ebrahimi, “Change Detection and Background Extraction by Linear Algebra,” Proceedings of the IEEE, Vol. 89, No. 10, pp.1368-1381, 2001.

[2] S. Dockstader and A. M. Tekalp, “Multiple Camera Tracking of Interacting and Occluded Human Motion,” Proceedings of the IEEE, Vol. 89, No. 10, pp.1441-1455, 2001.

[5] C. Kim and J. N. Hwang, “Fast and automatic video object segmentation and tracking for content-based applications, ”IEEE Transactions Circuits and Systems for Video Technology, vol. 12, pp. 122-129, Feb. 2002.

[6] A. Naohiro and F. Akihiro, “Detecting Obstructions and Tracking Moving Objects by Image Processing Technique,” Electronics and Communication, vol. 82, no. 11, pp. 28-37, 1999.

[7] A. J. Lipton, H. Fujiyoshi, and R. S. Patil, “Moving target classification and tracking from real-time video,” in Proc. IEEE Workshop on Applications of Computer Vision, pp. 8-14, 1998.