Machine-Learning Algorithm for Digital Image Forgeries by Illumination Color Classification
G.Reddy Swetha1, Mr.Ravi Kishore2
Citation : G.Reddy Swetha,Mr.Ravi Kishore, Machine-Learning Algorithm for Digital Image Forgeries by Illumination Color Classification International Journal of Innovative Research in Electronics and Communications 2015, 2(6) : 7-14
In this method we propose the method to detect the forensic in the photography. For that here we use the svm classifier for the forensic detection. Initially we identify the illuminant map in the image. We find the face from the photography. For the face detect here we use the violo john method. After face detection After that we identify the GLCM (Gray Level Co-Occurance Matrix). In GLCM is the statistical information of the image such as energy, entropy, correlation sum of energy and sum of correlation are calculated. And also we extract the LBP feature. The extracted feature will pass to the SVM classifier for the training. SVM is stands for Support vector machine. It is a binary classifier. It is a kernel based learning classifier. The trained classifier will predict about the image whether it is original or forensic image.