Abstract: A novel two stage multiple parameter statistical algorithm to identify pornographic images is proposed in this study. In the current study we present an analysis on various color spaces to identify an optimal color space for human skin pixel identification. A new algorithm is proposed to identify and avoid the explicit image by considering high skin pixel rate. The proposed algorithm was tested in terms of accuracy, true negatives and false positives and the experimental results presented in the current study show that the algorithm worked well and fast in detecting pornographic images.
R. Balamurali and A. Chandrasekar, 2016. A Statistical Algorithm Approach for Explicit Image Discretion. Asian Journal of Information Technology, 15: 5093-5101.