Image Replica Detection based on Binary Support Vector Classifier

In this paper, we present a system for image replica detection. More specifically, the technique is based on the extraction of 162 features corresponding to texture, color and gray-level characteristics. These features are then weighted and statistically normalized. To improve training and performances, the features space dimensionality is reduced. Lastly, a decision function is generated to classify the test image as replica or non-replica of a given reference image. Experimental results show the effectiveness of the proposed system. Target applications include search for copyright infringement (e.g. variations of copyrighted images) and illicit content (e.g. pedophile images).