Segmentation And Tracing Of Moving Object In Complex Environment

The recognition and tracing of moving object is a very popular subject inthe area of Computer Vision,image Processing and Pattern Recognition.Althoughsome achievements have been made in the studying of moving object segmentationand tracing ,how to segment and trace moving object in complex background andmeet demand of real-time processing is still hotspot of research in this domainbecause moving object often stands in the diverse surroundings.This present thethesis focues on the the choose of the threshold ,the segmentation and tracingof moving objects in complex background.Drawn on such technology and research of choosing the threshold as maximumentropy method, adaptive threshold ,alternative method , Otsu method,thisthesis proposes, a new method of choosing the threshold based on the improvementOtsu method. This method is simple.Under certain condition ,this method is noteffected on image contrast gradient and brightness change influence.In order to enhance the object segmentation speed and inspecting precision,this paper proposes a method of semiautomatic segmentation of target sector.In view of background change and the background complexity of the image sequence,this paper utilizes this method to choose threshold in object segmentation ofthe image sequence, proposes a new image segmentation approach by usingautomation threshold selection based on the improvement Otsu theory, hasrealized the fast segmentation of moving object under the complex background.The experiment results show that this method of object segmentation reducesthe processings of the image noise reduction and the image enhancement, meanwhilehas the advantages of denoise,simple computation and desirable effect andefficiency.In view of the moving object tracing under the complex background,this paperproposed one kind characteristic point tracing method based on morphologicsubject. In view of target tracking of the continual frame,this method obtainsthe very good tracing effect, and has obtained the object position in the imagesequence ,the path of the characteristic point of the object ,and the contastof the path and the estimate path. But still the target tracking of the gap frameneeds also further research.