Title :
Detection of leaked oil using image processing
Author :
Aoki, Nobuya ; Inari, Takahiko
Author_Institution :
Fac. of Biol.-Oriented Sci. & Technol., Kinki Univ., Osaka, Japan
Abstract :
This paper describes a new detection method for leaked oil in plants using fluorescence of the oil under ultraviolet illumination and optical image processing techniques. In order to extract the oil image from the background image, we used image subtraction. This principle was proved by the fundamental experiment in this paper. The practical system used for such as mobile robots requires short time illumination because of elimination of the image fluctuation. We propose in this paper the detection system including the control function of the illumination and the exposure of the CCD-camera. The possibility of this function was examined by experiments
Keywords :
CCD image sensors; feature extraction; fluorescence; leak detection; optical signal detection; CCD-camera; background image; fluorescence; illumination control function; image processing; image subtraction; leaked oil detection; mobile robots; oil image extraction; optical image processing techniques; short time illumination; ultraviolet illumination; Biology; Brightness; Fluorescence; Image processing; Inspection; Leak detection; Light sources; Lighting; Mobile robots; Petroleum;
Conference_Titel :
Industrial Electronics, Control and Instrumentation, 1997. IECON 97. 23rd International Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-3932-0
DOI :
10.1109/IECON.1997.668536