Multi-frame Images Fusion with Image Enhancement for Different Focus Points

Abstract

This paper presents a new approach to integrate and enhance registered multi-frame images with varied focuses simultaneously. The objective is to combine the information of multiple images from the same scene and then generate an image that is more suitable for human and machine perception or further image-processing tasks than its original sources. To enhance the integrated image, Smoothing Filter-based Intensity Modulation (SFIM) is first applied to multi-frame images with different focus points. Then the non-orthogonal wavelet decomposition based on contrast measure is used for extracting the details. Finally, the generated image is reconstructed by utilizing the inverse non-orthogonal wavelet transform. Experimental results demonstrate that the proposed approach can create a representation for the obtained scene with more details. Furthermore, this approach can achieve better performance than standard methods of discrete cosine transform (DCT), discrete wavelet transform (DWT), morphological wavelet transform (MWT), and Laplacian Pyramid.