TY - JOUR ID - 88085 TI - Automated Detection of Region of Interest using Non-Parametric Distribution Based on Bayesian Risk JO - Journal of Machine Vision and Image Processing JA - JMVIP LA - en SN - AU - Razavi, Mahnaz AU - Taherinia, Amir Hossein AU - Sadoghi Yazdi, Hadi AD - Faculty of Engineering, Ferdowsi University of Mashhad, Mashhad, Iran AD - Department of Computer Engineering Faculty of Engineering,Ferdowsi University of Mashhad AD - Faculty of Engineering, Ferdowsi University of Mashhad Y1 - 2020 PY - 2020 VL - 6 IS - 2 SP - 159 EP - 174 KW - Region of interest KW - Event KW - non-parametric distribution function KW - Bayesian risk KW - live surveillance cameras DO - N2 - In this paper, a new method for automated detection of a human region of interest is provided that makes use of camera surveillance in department stores. In this work, a region of interest is an area in the image where more people commute. For this purpose, first humans are distinguished from other objects in the image utilizing a histogram of oriented gradients (HOG) descriptors. Every detected individual is considered as an event in the image. Then, a non-parametric distribution based on Bayesian risk is applied to obtain the most interested regions from the position of detected humans. In the proposed distribution, a new high-efficiency kernel is provided. In Bayesian risk, a novel loss function is proposed that has a higher accuracy in compared with square loss function and performs better in finding peaks of a distribution function. For the evaluation, data from live surveillance cameras located in different parts of some stores are used. For the proposed kernel, on average, an accuracy of 85% and for the loss function, an accuracy of 93.5% on artificial data and 90% on real data are acquired which are better results in compared with other similar works. UR - https://jmvip.sinaweb.net/article_88085.html L1 - https://jmvip.sinaweb.net/article_88085_ed8a9c81adea479137e47b39f5dd18bc.pdf ER -