Research On Metallographic Microscope Image Fusion

Metallographic microscopy images are used for scientists to improve the accuracy of metal material microstructure services, such as analyzing the composition of new metal materials and testing the quality of metal products, etc. With the demand for in-depth scientific research, researchers hope to obtain high-resolution, large field optical microscopy images. Currently, the microscopes use high numerical aperture and large magnification objectives to obtain high-resolution metallographic microstructure images. However, due to the small field depth of such objectives, it is prone to capture a picture out of focus. At the same time, view field of these objectives are relatively smaller. Ordinary microscope is difficult to meet this requirement and the actual needs of the scientific research. In view of these problems, the subject of the research on image fusion and image stitching technology has important theoretical significance and application value. By improving the image fusion and image stitching technology to improve microscopic imaging techniques, the global clear pictures can be obtained finally.The main contents of this thesis are as follows:1.. By research on the principle of high-resolution optical microscope imaging, pulse neural network and Wavelet-Contourlet principle, the image fusion algorithm based on pulse neural network and wavelet-Contourlet is brought out.2.. After the further research on microscopic picture feature, surf feature point detection method and the phase correlation method, the combination of surf and phase feature point detection method of image stitching technology is brought out3.. Through build the experimental platform to acquire images, innovation and accuracy of the proposed algorithm about the image fusion and image mosaic are verified.4.. The results of this study is initially applied to the type of microscope JX-4 of the high-resolution flat field with a semi-apochromatic lens which is self-developed by our laboratory which improve the resolution of optical microscopy images, expand its imaging view field , improving the overall imaging performance of the JX-4.After two-year research, the principles of optical microscopy image are fully understood, and the image fusion and image stitching technology theory are mastered. Based on the above research, image fusion algorithm and the image stitching method can be improved eventually. The experiments show that, the proposed image fusion algorithm and image stitching algorithm are advanced and accuracy, which can improve the image technology of the microscope, and lay a solid foundation for the further development of prototype products.