METHOD AND APPARATUS FOR TRACKING AND DIVIDING OBJECTS ON AN IMAGE USING MARKOV CHAIN

摘要

PROBLEM TO BE SOLVED: To provide a tracking algorithm and framework by a simple and new segmentation which applies an absorption Markov chain to a visual tracking for an atypical and deformable object.;SOLUTION: An object of the present invention is to trace the object of a moving image data frame by using the absorption Markov chain by acquiring a bounding box including a target object on a current frame and a segmentation result for the target object in a previous frame, expanding a bounding box in order to include a part of the information on a background surrounding the target object, acquiring the ROI on the current frame, acquiring the information on a local area within the ROI on the current frame, constituting an absorption Markov chain graph by using the local area within both ROIs of the current frame and the previous frame, and acquiring the segmentation result for the target object on the current frame by using the number of times of absorption and determining the threshold value of each node in the absorption Markov chain graph.;SELECTED DRAWING: Figure 5A;COPYRIGHT: (C)2019,JPO&INPIT