In the digital world, images are extremely important. The forging of an image is becoming more common in the digital age. Because of the latest image alteration tools and technologies, identifying a forged image has become a difficult undertaking. The suggested method employs a hybrid approach to identify image forgery. The proposed approach is broken down into three stages. In the first phase, the R, G, and B channels of an input image are clustered employing K-Means, and then the Gaussian filter is utilized to blur the image. The entropy feature is then taken from an input image's colour and grey scale and stored in matrix form. The retrieved characteristics are matched in the second phase, and the matched blocks are coloured blue and red. In the third phase, count the continuous locations of matched blocks; if it is larger than 10 pixels, the image is forged; else, it is not forged. According to the experimental results, the hybrid technique detects forgery in an image with great accuracy