Recognition on porn image based on weight strategy
Zhou Liqia · Journal of Central South University(Science and Technology) · 2013
For the problem that the porn images on the network spread unchecked, a new method based on weight strategy was proposed for the recognition of porn image based on the face detection, skin-color detection, image texture analysis and other methods. The appropriate color space was firstly selected. Then the texture information from skin region was extracted. Finally the corresponding characteristics of porn image and normal picture collected from network were trained and classified by the feature weight strategy and the method of support vector machine(SVM).The results show that the accuracy rate of this method is 88.85% for the recognition of porn image and 86.70% for normal picture,and the method can be applied to filter porn image in actual software system.