Recently, there is a continuing need to verify the quality of being authentic of digital images and large-scale research has been conducted to discover methods for detecting fake images. Being a member of the famous varieties of digital manipulation, copy-based forgery (CMF) is mostly studied. This editing technique is translating a fraction of an image and then inserting it into another image or back onto the original. The paper is made for the objective of recognition the copy-move forgeries in images and provides an analysis of four machine learning algorithms such as KNN, logistic regression, naïve Bayes, and convolutional neural networks which are commonly used. The work is completely utilizing the CoMoFoD dataset; CoMoFoD is a standard for copy-move forgeries, and our study performs an exhaustive evaluation of various techniques. We analyze the positive aspects and drawbacks of each approach for resolving all the difficulties posed by copy-move forgery identifications. The outcomes are in the research demonstrated that the CNN and KNN models are particularly strong competitors in forgery detection especially deep learning approach like CNN providing good precision (95.46), recall (90.62), and accuracy (90.62) scores.