Multi Focus Image Fusion Based on Spatial Frequency and Contrast Based Analysis under Stationary Wavelet Transform Domain

Due to the limited depth of focus of cameras a scene can sometimes not be described accurately on the basis of a single image. Multiple Images have to be fused together to get a clear description. In this paper we use stationary wavelet transform with contrast analysis and spatial frequency to perform multi focus image fusion. The existing Discrete Cosine Transform (DCT) method and other wavelet transforms suffer from problems like blocking or ringing artifacts or computational complexity. The performance of the system is evaluated on the basis of the various parameters like Peak signal to noise ratio, minimum mean square error, entropy and correlation coefficient. The results over a number of Joint Photographic Experts Group (JPEG) images show improvement over the existing method of spatial frequency and average based DCT image fusion and Non-subsampled Contourlet Transform (NSCT).