Aim: Image Forgery detection is real-time photos with enhanced accuracy utilizing support vector machines over polynomial regression. Materials and Methods: The G-power setting parameters were used to accomplish image forgery using Support Vector Machine (N=10) and Random Forest (N=10) with the partition length of testing and training datasets being 60% and 40%, accordingly. Results: The Support Vector Machine is 93.1% which is more accurate than Random Forest of 79% in classifying Satellite Image Segmentation attained the significance value 0.071 (Two tailed, p>0.05). Conclusion: When attempting to identify picture counterfeiting in real-time photographs, the novel Support Vector Machine model performs noticeably better than Random Forest (RF).