在監控系統視訊資料庫中對異常物體搜尋之方法

An Abnormal Object Retrieval Method for the Surveillance Video Database

曹佑羽(Yu-Yu Tsao)

指導教授 ： 張厥煒

國立臺北科技大學/電資學院/資訊工程系研究所/碩士(2007年)

https://doi.org/10.6841/NTUT.2007.00190

摘要

傳統的保全監控系統，在災害或犯罪事件發生後想要找出可疑人物，必須經由人工調閱並監看事件發生時所錄下的視訊影像，如此一來將花費過多的人力、時間，而延誤案情。本論文提出一個運用於保全監控系統的視訊檢索系統架構，使用背景相減方法(Background Subtraction Method)偵測異常物體，找出監控畫面中的異常事件段落，並藉由關鍵異常畫面擷取(Key Frame Extraction)演算法，找出可明顯表現出異常物體特徵之關鍵異常畫面，再建立畫面中異常物體的以顏色為基礎之物體模型(Object Model)，並將相關資訊紀錄於資料庫中。使用者可以時間或攝影地點等條件查詢異常事件，並可框選出感興趣的異常物體，與資料庫中的紀錄進行比對，完成監控系統視訊資料庫中異常物體搜尋動作。

關鍵字

監控系統 ； 視訊檢索

並列摘要

In the traditional surveillance system, the search of suspects in the video tapes has to be monitored by human and is always time consuming. In this thesis, we propose a new video indexing and retrieval mechanism for surveillance video database. This mechanism detects abnormal events by using background subtraction method, extracts representative images of abnormal objects by our key frame extraction method, and stores these information into the surveillance video database. When searching for the abnormal suspects in the database, user can select an object by specifying a region of interesting in an image. The Select object can be compared with these abnormal objects in the database by the color-based similarity measure method.

並列關鍵字

Security Surveillance ； Video Indexing

參考文獻

[1]S. Antani, R. Kasturi, and R. Jain, “A survey on the use of pattern recognition methods for abstraction, indexing, and retrieval of images and video,” Pattern Recognit., vol. 35, no. 4, pp. 945-965, 2002.

[2]D. Makris and T. Ellis, “Path Dectection in Video Surveillance,” Image and Vision Computin., vol. 20, pp. 895-903, 2002.

[4]A. Elgammal, D. Harwood, and L. Davis, “Non-Parametric Model for Background Subtraction,” Proc. IEEE Frame Rate Workshop, 1999.

[5]S. Jabri, and et al., “Detection and Location of People in Video Images Using Adaptive Fusion of Color and Edge Information,” Proceedings International Conference on Pattern Recognition, vol. 4, pp. 627-630., 2000.

[7]D. Meyer, J. Denzler, and H. Niemann, “Model based extraction of articulated objects in image sequencesfor gait analysis,” Proc. IEEE Int. Conf. Image Processing, pp. 78-81, 1998.