Images are a natural carrier of information. Images are used in an immense range of applications nowadays, including military purposes, surveillance systems, insurance processing, the internet, television, advertising media, forensic investigation, and so on. However, because powerful, low-cost image editing tools are readily available, these images can be easily tampered with. Therefore, the authenticity of images has become questionable. In this age of advanced PC innovation, digital picture and video have high significance in our everyday life. For editing or modifying the original multimedia contents, a range of low-cost multimedia content handling tools, techniques, and applications with various advanced features are available on the Internet, such as Adobe Photoshop. To handle this issue, numerous investigations have been centered around how to identify, this kind of controlled media. Existing computerized fraud identification techniques are grouped into two significant classes: Active and passive. This paper proposes a technique of copy-move forgery detection in which feature extraction is done by Local Binary Patterns (LBP) and Harlick features. For the verification of the authenticity of the image various supervised machine learning classifiers like Support Vector Machine (SVM), Random Forest (RF) Gradient Boost classifiers are used and the performance of forgery detection is based on these classifiers are analyzed.