應用影像技術於煙霧偵測之研究

A Study of Smoke and Fog Detection Using Image Processing Technique

許家昌(Jia-Chang Shiu)

指導教授 ： 吳明川

國立臺北科技大學/機電學院/機電整合研究所/碩士(2012年)

摘要

煙霧影像偵測技術能用於多種場景，不論是由氣候潮濕所產生的霧氣，亦或是由燃燒的微粒所產生的煙霧，能夠用於監測道路、港口、機場等需要良好的視野與監控。本研究將水氣與燃燒所產生的兩種影像技術整合，完成一套即時煙霧偵測系統。
首先根據燃燒後所產生的煙霧作偵測，由顏色區分出煙霧的區域，再由運動歷史軌跡演算法，區別畫面中的移動物體，煙霧有向四周擴散的特性，檢驗區域中飽合度的變化，最後由於燃燒的煙霧溫度較高，有向上移動的特性，由此特性來避免誤抓，對於潮濕所產生的霧氣，由其影像模糊特性，作快速傅立葉轉換，判別出影像清晰程度，再以顏色區別出煙霧的區域。

關鍵字

煙霧檢測 ； 霧氣檢測

並列摘要

Smoke image detection technology is a widely used application, whether for humid climate, fog, smoke or combustion particles. It's also used for many monitoring scenes, such as highway, air port, port, etc. In this study, the water vapor and burning images are integrated to establish a real-time detection system.
First, detect the smoke generated by combustion, and color-code the areal differentiate of the smoke accordingly. Then use the Motion History Image (MHI) algorithm to differentiate the moving objects on the screen. Smoke spreads to the surrounding properties. Test changes in the regional saturation. Since high temperature of the burning smoke makes the particle move upward. To avoiding mistaken detection is essential. Moist from fog blurs the image and then by using Fast Fourier Transform (FFT) to distinguish image clarity, the color difference between the regions can be accurately established.

並列關鍵字

Smoke Detection ； Fog Detection

參考文獻

[1] Paolo Piccinini, Simone Calderara, Rita Cucchiara, “Reliable Smoke Detection System in the Domains of Image Energy and Color,” ICIP, 2008.

[2] Hidenori Maruta, Yasuharu Katot, Akihiro Nakamural, and Fujio Kurokawall, “Smoke Detection in Open Areas Using Its Texture Features and Time Series Properties,” IEEE International Symposium on Industrial Electronics, 2009.

[4] Guillemant, P., Vicente, J., “Real-Time Identification of Smoke Images by Clustering Motions on a Fractal Curve with a Temporal Embedding Method,” Society of Photo-Optical Instrumentation Engineers, pp.554–563, April 2001.

[5] Thou-Ho (Chao-Ho) Chen, Yen-Hui Yin, Shi-Feng Huang and Yan-Ting Ye, “The Smoke Detection for Early Fire-Alarming System Base on Video Processing,” International Conference on Intelligent Information Hiding and Multimedia Signal Processing, 2006.

[6] Feiniu Yuan, “A Fast Accumulative Motion Orientation Model Based on Integral Image for Video Smoke Detection,” Pattern Recognition Letters 29, 2008, pp.925-932.