Improved method of multi-focal plane micro-image fusion

In order to implement image fusion with multi-focus sequence images for the 3-D reconstruction of the micro-morphology on machining metal surface, an improved image fusion method based on a variance block of edge sharpening is presented. Firstly, an image capture system including an adjustable measuring microscope was constructed according to the microscopic surface features under different machining methods. The evaluation parameters were selected so as to determine focus parts in some 2-D serial images and to properly evaluate processing results. A multi-focus image fusion algorithm was proposed to obtain a clear focus image and to avoid mosaics, edge blurring and so on. And then some processing images were shown and the proper conditions achieving this algorithm were discussed by image evaluation value under different processing. Data processing results show that the method is feasible, simple and high efficiency, and also provide a basis study for 3-D reconstruction of metal surface and farther roughness measurement.