Determination of threshold for IR image by using PICA with RBF neural network

Rappy Saha, Jai-Eun Kim, Ki-Doo Kim

Abstract: This paper is about determining the threshold for a gray scale IR image. At first, Pixel Intensity Clustering Algorithm (PICA) is used to have the center for the Radial Basis Function (RBF) neural network. PICA clusters the image into a defined number. Every group has a centroid which has been used as RBF neural network center. The number of hidden neurons will also be equal to the number of clusters. We give the weight to the output of the hidden neurons on the base of the desired output and the neural network output will be a threshold value for the corresponding gray scale image. The simulation result gives us a clear idea about the desired result.