於視覺基礎上之室內吸菸行為識別與偵測

The Vision-Based Recognition and Detection for Indoor Smoking Behavior

應念廷(Nien-Ting Ying)

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(2009年)

https://doi.org/10.6841/NTUT.2009.00137

摘要

在面對國內龐大的吸菸人口，於公共場所吸菸的限制，卻越趨嚴格的菸害防制法令下，本研究提出一個利用一般固定式攝影機，基於影像視覺基礎的方法上，使用即時系統來監控室內環境，對於室內吸菸行為加以識別與偵測，以期減少監控人力上的負擔，並且保障及維護其他更多非吸菸者的權益。在系統上採用影像背景相減法來偵測監控畫面，找出畫面中發生變動的前景部份，判斷在前景之中是否出現可能為煙霧之像素。在前景之中篩選出煙霧像素後，搭配人臉偵測模組，找出畫面中出現之人臉，並且利用煙霧發生位置與人臉出現位置，定義畫面空間邏輯關係，進而判斷畫面之中是否出現吸菸之行為。
在煙霧偵測方法上，採用以色彩為基礎的作法，將原本監控視訊影像由RGB色彩空間轉換到HSI色彩空間上，利用煙霧半透明特性，在HSI色彩空間中找出具有煙霧特徵的像素並加以紀錄。而在人臉偵測演算法上，則採用HAAR Object Detection之作法，使用HAAR-like feature找出五官輪廓特徵作為分類器的輸入，再利用類似AdaBoost方法的Rejection Cascade來加以分類輸入分類器之區塊是否為人臉。最後在定義吸菸行為的模組上，將畫面切割成數個邏輯區塊，用以表達煙霧像素的叢聚性，並且進一步篩選可能發生之煙霧。配合上人臉座標位置，定義人臉與煙霧之間的關係，作為最後發出吸菸警告的決策依據。

關鍵字

煙霧偵測 ； 人臉偵測 ； HSI 色彩空間

並列摘要

This paper introduces a method based on vision image processing to detect smoke in the indoor environment. This smoking surveillance system captures stable video with a fixed camera and detect smoke in real time. One feature of smoke which generated from cigarettes is translucent. According to this feature, system finds smoke in images by translating RGB color space into HSI color space. Face detection in the system discovers the smokers and locates their faces. Finally, this paper define the relationship between smoke and faces according to the results of smoke detection and face detection. Also, display the warning messages on the system interface.

並列關鍵字

Smoke Detection ； Face Detection ； HSI Color Space

參考文獻

[1] I. Kopilovic, B. Vagvolgyi and T. Sziranyi, “Application of Panoramic Annular Lens for Motion Analysis Tasks: Surveillance and Smoke Detection,” Proceedings of 15th International Conference on Pattern Recognition, Vol. 4, 3-7 , 2000, pp. 714 – 717.

[2] T. Chen, P. Wu and Y. Chiou, “An Early Fire-Detection Method Based on Image Processing,” in Proc. of IEEE ICIP ’04, Vol. 3, 2004, pp. 1707–1710.

[3] T. Chen, Y. Yin, S. Huang and Y. Ye, “The Smoke Detection for Early Fire-Alarming System Base on Video Processing,” Proceedings of the 2006 International Conference on Intelligent Information Hiding and Multimedia Signal Processing, 2006, pp. 427-430.

[4] B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin, “Wavelet based real-time smoke detection in video,” EUSIPCO ’05, 2005.

[5] Zhengguang Xu and Jialin Xu, “Automatic Fire Smoke Detection Based on Image Visual Features,” Computational Intelligence and Security Workshops, 2007, pp. 316-319.