Research On Depixelation And Image Mosaicking Algorithm Of Medical Confocal Endoscopy

With the improvement of living standards,people are increasingly concerned about their own health,the cancer has not been conquered in medicine.The survival time and crue rate of cancer patients are normally increased the sooner cancer is discovered.The early diagnosis of cancer are particularly important.At present,the diagnosis of cancer is mainly completed by the tissue biopsy in the medicine.The efficiency of biopsy depends on the accuracy of selection of suspicious region for biopsy.The medical confocal endoscopy can observe the tissue with sub-cellular solution so as to finish early diagnosis of cancer,while it also remains compatibility with the conventional endoscopy.The imaging results have problems of honeycomb pattern and small field of view,because of medical confocal endoscopy's physical structure.This article will use the depixelation and image mosaicking algorithm of medical confocal endoscopy to solve these problems.First,this article introduces the principle and the development history of confocal imaging technology.Its conjugated structure leads to high signal-to-noise ratio,good axial resolution.Medical confocal endoscopy is the combination of confocal imaging technology and endoscopy technology.Medical confocal endoscopy is based on a fiber bundle.The imaging results are affected by the honeycomb structure of the fiber bundle.While traditional image algorithms can't achieved ideal effects,we further study the pre-calibration algorithm for confocal endo-microscopy imaging.Pre-calibration algorithm can remove effectively honeycomb pattern without the degrade of image quality.On the basis of pre-calibration algorithm,we further study blind calibration algorithm which is more suit for clinical requirement.Medical.confocal endoscopy have the problem of small field of view.Aiming at this problem,this article studies image mosaicking algorithm for medical confocal endoscopy.The algorithm forms panoramas from dozens of continuous moving images.It solves the imperfection of small field of view.The field of view is increased from micron-scale to centimeter-scale.