Shahrood University of Technology Journal of AI and Data Mining 2322-5211 11 3 2023 07 01 An Optimal Hybrid Method to Detect Copy-move Forgery 429 442 2903 10.22044/jadm.2023.13166.2453 EN Fatemeh Zare Mehrjardi Computer Engineering Department, Yazd University, Yazd, Iran. Alimohammad Latif Computer Engineering Department, Yazd University, Yazd, Iran. Mohsen Sardari Zarchi Computer Engineering Department, Meybod University, Meybod, Yazd, Iran. Journal Article 2023 05 25 Image is a powerful communication tool that is widely used in various applications, such as forensic medicine and court, where the validity of the image is crucial. However, with the development and availability of image editing tools, image manipulation can be easily performed for a specific purpose. Copy-move forgery is one of the simplest and most common methods of image manipulation. There are two traditional methods to detect this type of forgery: block-based and key point-based. In this study, we present a hybrid approach of block-based and key point-based methods using meta-heuristic algorithms to find the optimal configuration. For this purpose, we first search for pair blocks suspected of forgery using the genetic algorithm with the maximum number of matched key points as the fitness function. Then, we find the accurate forgery blocks using simulating annealing algorithm and producing neighboring solutions around suspicious blocks. We evaluate the proposed method on CoMoFod and COVERAGE datasets, and obtain the results of accuracy, precision, recall and IoU with values of 96.87, 92.15, 95.34 and 93.45 respectively. The evaluation results show the satisfactory performance of the proposed method. https://jad.shahroodut.ac.ir/article_2903_1eefe9ff33dc6b4c3f3dc2454e429c12.pdf