Authors : Eman I. Abd El-Latif, Ahmed Taha and Hala H. Zayed
Abstract: Recently, the problem of detecting image splicing forgery attracts many researchers. Many algorithms have been presented to deal with this problem. However, most of them suffer from high dimensional feature vector. In this study, an algorithm is offered to reveal the splicing manipulation in the digital image with size feature vector. The proposed algorithm is predicated on Haar Wavelet Transform (HWT) and Uniform Local Binary Pattern (ULBP). The image color space is changed into YCbCr space. This algorithm works on the chrominance components and HWT is employed to get the four sub-bands. For every sub-band, ULBP is applied. The last feature vector is generated by merging features from the four sub-bands. Support Vector Machine (SVM) is used as a classifier. The proposed algorithm is examined on a freely accessible splicing image datasets (CASIA V1.0 and 2.0). The experiments prove that the proposed algorithm is successful for detecting the spliced image.
Eman I. Abd El-Latif, Ahmed Taha and Hala H. Zayed, 2019. Image Splicing Detection using Uniform Local Binary Pattern and Wavelet Transform. Journal of Engineering and Applied Sciences, 14: 7679-7684.