This paper analyses and implements a novel technique to filter objectionable adult images displayed in websites, as this problem remains an interesting issue to be addressed in the current scenario. The algorithm is aimed to achieve precise performance, even in distinctive circumstances, where face detection becomes impossible. This is realized by using an ancillary technique in which the Human Body region is analyzed using shape, color and image centered pixel scanning analysis. Further a series of MATLAB simulation results are shown. The results of pixel scanning approach are analyzed and an enhanced integrated filter is designed to improve the performance. Finally the concept of pixel scanning approach is tested and demonstrated in real time using a web portal.
展开▼
