Analysis of Copy Move Forgery Detection Process by Applying Fuzzy C Means Algorithm Based on Deep Learning in Digital Image Processing

Abstract

- The popularity of digital photos has developed due to technological advancements in the digital environment. Image alteration has become more manageable thanks to powerful and user-friendly photo editing software programs. Therefore, there was a prerequisite to detect the forged part of the image efficiently. Hence, this work emphasizes passive forgery recognition on images tampered by the copy move method, better called Copy Move Forgery Detection (CMFD). Copy Move Forgery (CMF) was fundamentally concerned with covering or repeating one area in a picture by pasting certain regions of a similar picture. Initially, the input digital images were preprocessed through a Gaussian filter to blur the picture to decrease noise. After preprocessing, Multi-Kernel Fuzzy C-Means clustering (MKFCM) was performed to divide the images into numerous clusters to extract the features based on distinctive attributes using the SIFT method. Lastly, with the deep learning technique, the forged parts of the images were detected. The experimental analysis demonstrates that the method was efficient and robust in identifying the forged part of the digital picture, and the performance of the proposed strategy was established on numerous forged pictures.