|Home||<< 1 >>|
Allik, T., Ramboyong, L., Roberts, M., Walters, M., Soyka, T., Dixon, R., et al. (2016). Enhanced oil spill detection sensors in low-light environments. In Proc. SPIE 9827, Ocean Sensing and Monitoring VIII, 98270B (May 17, 2016). SPIE.
Abstract: Although advances have been made in oil spill remote detection, many electro-optic sensors do not provide real-time images, do not work well under degraded visual environments, nor provide a measure of extreme oil thickness in marine environments. A joint program now exists between BSEE and NVESD that addresses these capability gaps in remote sensing of oil spills. Laboratory experiments, calibration techniques, and field tests were performed at Fort Belvoir, Virginia; Santa Barbara, California; and the Ohmsett Test Facility in Leonardo, New Jersey. Weathered crude oils were studied spectroscopically and characterized with LWIR, and low-light-level visible/NIR, and SWIR cameras. We designed and fabricated an oil emulsion thickness calibration cell for spectroscopic analysis and ground truth, field measurements. Digital night vision cameras provided real-time, wide-dynamic-range imagery, and were able to detect and recognize oil from full sun to partial moon light. The LWIR camera provided quantitative oil analysis (identification) for >1 mm thick crude oils both day and night. Two filtered, co-registered, SWIR cameras were used to determine whether oil thickness could be measured in real time. Spectroscopic results revealed that oil emulsions vary with location and weathered state and some oils (e.g., ANS and Santa Barbara seeps) do not show the spectral rich features from archived Deep Water Horizon hyperspectral data. Multi-sensor imagery collected during the 2015 USCG Airborne Oil Spill Remote Sensing and Reporting Exercise and the design of a compact, multiband imager are discussed.