The Research Of Leucorrhea Image Capture And Mosaic Technique In The Microscope Environment

The image splicing technology is a technique for synthesizing a large-field wide-format image by multiple image with overlapping areas through image processing.In recent years,image mosaic technology as a key technology for synthesizing panoramic images has attracted the attention of researchers and has been rapidly developed at home and abroad.At present,panoramic images have become hotspots and key technologies in computer simulation,computer vision simulation,image processing,computer special effects,and virtual reality research.This thesis mainly designs and develops an image acquisition and splicing system based on leucorrhea imaging.This system is an integrated system that integrates optics,motion control,software design,and image processing.It is an exploration of the clinical testing system for gynecological diseases.It is mainly to achieve the full automation of leucorrhea microscopy,while avoiding multiple shots to separate the components of interest in the sample,resulting in inaccurate judgments,and thus affecting the doctor's diagnosis and treatment of patients.The design of image acquisition and splicing system mainly includes the following modules: 1.Hardware module: including light source,microscope,camera,motor,development board,etc.2.Software module: communication module,control module and interface module;3.Algorithm module.The system is based on the windows7 operating system,using C + +,Qt language developed in the Visual Studio 2012 environment.The main research work and achievements of this thesis are as follows:First of all,in this thesis,the overall design of the system from hardware to software and algorithm is proposed for the functional requirements and performance requirements of the image acquisition and splicing system.Then,the software design of this thesis is mainly two aspects of communication module and control system.In the communication module,asynchronous I/O technology based on multi-threads is used.At the same time,the reliability of communication is ensured through the mechanism of timeout retransmission.The control system is the basis of the entire image acquisition and splicing system,so its effect will directly affect the quality of the subsequent image acquisition and splicing.This topic uses a second-level buffer to improve the quality of image acquisition.At the same time,it uses the design mode of the responsibility chain to separate the request and execution of the command and improve the flexibility and extensibility of the software.Then,this thesis designs a fully automatic large-format stitching algorithm based on SURF partial feature extraction and brightness balance,and uses Matlab to do algorithm rehearsal,and finally uses Opencv to implement a complete image stitching algorithm.The algorithm design solves the problem that the OpenCV function may cause information loss due to image transformation.At the same time,aiming at the problem of unequal brightness of stitched images,a method with smaller amount of calculation and higher accuracy is proposed,and the entire algorithm is optimized.At last,OpenMP was used to optimize the algorithm and improve the running speed of the algorithm.The image acquisition and splicing system based on leucomicroscopy imaging designed by this project can effectively complete the image collection and splicing work.