A novel method for determination of the oil slick area based on visible and thermal infrared image fusion

Abstract

It is significant to prevent and supervise marine pollution through estimating the area of oil slick, when oil spill occurs. The existing determination method of the oil slick area has been developed based on visible image, which is influenced by illumination condition changes more easily. Since the thermal infrared image is almost not affected by illumination changes, it is introduced to determinate the area of oil slick. But it is hard to discriminate between oil slicks and look-alikes (the same thermal characteristics as oil film) without background and prior information. So, based on the visible and thermal infrared image fusion, a novel determination method of the oil slick area is proposed in this paper. The oil slick regions of interest (ROI) are extracted with help of the thermal infrared image processing. Then, based on Principal Component Analysis (PCA) fusion between visible and thermal infrared image, the background and priori information are obtained to discriminate the oil slick. The area of the real oil slick is calculated by the pixel area calculation method. The experimental results show that the proposed method can accurately determinate the oil slick area under different illumination, and the mean error is 2.78%. Moreover, compared with other method, the proposed method can achieve better results in both subjective evaluation and objective indicator. It seems to provide novel insights and supports for marine oil spill pollution control.