An All Approach for Multi-Focus Image Fusion Using Neural Network
'Image Fusion' is the Information from multiple images which are combined to generate the new image, the generated image is more suitable for humans and machines for further image-processing tasks like image segmentation, edge detection, stereo matching, enhancement, extraction and recognition. Novel feature-level multi-focus image fusion technique is proposed in this paper, which fuses multi-focus images using classification. In this technique, Multi-focus images of ten pairs are divided into blocks and the most favorable block size for each image was found in an adaptive manner. The resultant block feature vectors are fed to feed towards neural network.