In ancient times, images were used very rarely & there was a possibility of less amount of forgery or no forgery in images. It requires much knowledge to create a forged image earlier. Nowadays, Images have gained a very vital importance in our daily life and it is not very difficult to make forged images because of the availability of powerful digital image editing software’s that does not require any expert knowledge. So, it becomes very easy to create a tampered image. As a result we have to prove the authenticity of an image. In this paper we have discussed about one of the most common forms of image forgery which is image splicing and other forms. We discussed about various existing forgery detection methods and techniques based on color illumination & machine learning approach which results in automatic decision making. We have also discussed about the existing work drawbacks and the possibility of future improvements. Index Terms— Forgery, Tampered Image, Image Splicing, color Illumination