Content based image retrieval through feed forward neural networks

We present you a process-method that has no previous knowledge of image appearance sample within the database, but retrieves basing on content information of image called content based image retrieval. Here we are trying to enhance the work nature of image retrieving with more efficient accuracy in Image similarity through multilayer feed forward network perceptron named Radial basis Function neural networks such that the input parameters given will able to provide us with most accurate relevant image to that of given query image, promoting through applying k means clustering to form clusters which are parameterized with Gaussian function application. Finally compare the actual output with observed output, the weighted neurons adjustment for getting the actual accurate output.