Automatic Collection And Splicing Of Microscopic Panoramic Images

Automatic microscope system can realize automatic collection of microscopic images,automatic stitching,simultaneous browsing and remote control,obtaining the panoramic image of objects efficiently for further analysis and study.However,most of the products in the market are expensive,difficult for the secondary developing and with a relatively small scanning range.Therefore,in response to realizing the demand of automatic acquisition and splicing of microscopic images at low cost,this thesis mainly does the following work.By analyzing the system components and working principles of common automatic microscopes on the market,the functions of each system module are summarized,and a hardware system and a software development platform are built.A polar coordinate polynomial distortion correction model is selected to achieve microscopic image distortion correction.First of all,with deep insight of the distortion types of microscopic images,a polynomial function are set for the relationship between the coordinates of ideal feature points and accordingly distorted feature points.These parameters of the polynomial function are estimated with corresponding image points.Then,the ideal image pixel is reconstructed by using the backward mapping method and the bilinear interpolation method.A good correction effect could be obtained with this procedure.A new microscopic image panorama stitching method is proposed based on the law of physical displacement in the process of image sequence acquisition.In order to improve the stitching accuracy of adjacent images,a template matching method based on image gray features is used to further optimize the splicing algorithm.