Measurement of image similarity is important for a number of image processing applications. Image similarity assessment is closely related to image quality assessment and is based on the apparent differences between a degraded image and the original, unmodified image. Automated evaluation of image retrieval systems relies on accurate quality measurement of similarity among the input image and the database images. In this paper, we have treated the image under pixel level where we used the mean squared error (MSE) algorithms for measuring similarity between the input image and the training data images, The mean squared error (MSE) simulations have demonstrated its promise through a set of examples by showing its accuracy and low computation cost, though it didn't show good results with precision on similarity of images and for rotated, translated or flipped images hence we proposed the use of minimum circumscribed circle (concentric circles) with local binary pattern and compare the results to get the best image similarity method. Hence comparing different results some improved performance were observed.