2024-03-29T11:05:25Zhttps://eprints.lib.hokudai.ac.jp/dspace-oai/requestoai:eprints.lib.hokudai.ac.jp:2115/576642022-11-17T02:08:08Zhdl_2115_20053hdl_2115_145Robust Object Detection in Severe Imaging Conditions using Co-Occurrence Background ModelLiang, Dong1000050134789Kaneko, Shun'ichiHashimoto, ManabuIwata, KenjiZhao, XinyueSatoh, Yutakaopen accessThis is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Optomechatronics on vol. 8, no. 1, 2014, available online: http://www.tandfonline.com/10.1080/15599612.2014.890686.background modelinglow-illuminationnarrow dynamic rangeobject detectionunderexposureIn this study, a spatial-dependent background model for detecting objects is used in severe imaging conditions. It is robust in the cases of sudden illumination fluctuation and burst motion background. More importantly, it is quite sensitive under the cases of underexposure, low-illumination, and narrow dynamic range, all of which are very common phenomenon using a surveillance camera. The background model maintains statistical models in the form of multiple pixel pairs with few parameters. Experiments using several challenging datasets (Heavy Fog, PETS-2001, AIST-INDOOR, and a real surveillance application) confirm the robust performance in various imaging conditions.Taylor & Francis2014-01engjournal articleAMhttp://hdl.handle.net/2115/57664https://doi.org/10.1080/15599612.2014.8906861559-96121559-9620International Journal of Optomechatronics811429https://eprints.lib.hokudai.ac.jp/dspace/bitstream/2115/57664/1/forIJO1.pdfapplication/pdf2.05 MB2014-01