The application of intelligent system to digital image forensics
Cheng-Liang Lai, Yi-Shiang Chen · 2009
Digital image capture devices, such as digital cameras or camera-equipped mobile phones, have become very common. Digital images also have the problem of being easy to edit and to tamper. As a result, digital forensics is now an important field in image processing. In this study, the features of images taken different cameras were used as the basis for determining the source of the digital images. The Genetic Algorithm was used to automatically select the most suitable and minimum number of features for the image content then the Support Vector Machine used for training and classification in order to identify the source cameras. This study also used image editing software for the post-processing of images, including resizing, blurring and tamper in order to determine if the Genetic Algorithm selected features were still effective for identification after the images were tampered. The results showed that the features selected automatically using the Genetic Algorithm could not only use less features, but also achieved better identification rates for the source camera of the digital images, and save images to extract the time of features after Genetic Algorithm select optimal feature.