Source Camera Identification Using Wavelet FeatureFusion Technique

Source camera identification is one of the significant phases in digital image forensics, which identifies the device used to capture the given image. The state-of-the-art methods for source camera identification are based on the differences in image formation process, storage techniques and the wavelet based frequency domain features. This paper proposes a feature fusion technique for source camera identification, which combines higher order wavelet statistics (HOWS) and wavelet coefficient co-occurrence features to achieve better identification accuracy. Experiment is conducted on a set of images and the evaluation section proves the robustness of the proposed method.