Study Of Pathological Images Stitching Based On Feature Points

As one of the important basis of pathology tissue for physician diagnosis,its accuracy directly determines the accuracy of the judgment of physician.Optical microscopy imaging techniques have been found widespread applications in medicine because of its capability of identifying anatomic features of pathological tissue.For the purpose of diagnostic,the image should be obtained in a form of wide-view with ultra-high resolution which provide quantitative information about the tissue.At present,the latest digital pathology scanning system can provide high resolution and the pixels of the image is more than 400 M.But because of the conventional image acquisition sensor sizes,we can not use it to realize the one-time acquisition of such high pixel image.Even if using the ultra high pixel image acquisition array,not only the sensor is expensive,but also the requirements of the making of the optical system is high,which we can not afford to.A practical and feasible solution is to use the current mainstream technology of image acquisition system,using high magnification,small field of vision and 2D mobile precision bearing units with stepping motor driven within the effective area of a pathological tissue slides of tissue boundary of overlapping collection for many times,and then use the fast accurate image stitching algorithm to join the each sub-graph collected many times into a complete image with ultrahigh resolution.In this contribution,the main work of this paper involves three section.First,an automated imaging system of wide-view of optical microscopy of pathological tissue was presented.Second,four algorithms of extracting feature points are studied.Finally,an improved algorithm of fast image stitching for the device described in the paper.Digital pathological images were obtained using our new microscope system and these images were scanned in blocks.During image acquisition,overlapping regions were appeared between neighboring block images to avoid missing border details.The whole images acquired were then reconstructed as a panorama of the whole sample with our improved fast stitching methods.This system is suitable for the field of clinical.The feasibility and performance of our proposed method was compared with present methods and modern key-point detectors to validated in processing clinical pathological images and proved to be high efficient and accurate,which made the doctor give a fast and accurate diagnose.