UDC 004.932.2
OBJECT LOCATION DETECTION IN DIFFICULT VISION CONDITIONS BASED ON TEMPLATE MATCHING METHOD
S. M. Larionov, master student, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
A. A. Selyaev, PhD (technical sciences), associate professor, RSREU, Ryazan; This email address is being protected from spambots. You need JavaScript enabled to view it.
The problem of object location detection in image sequence in difficult vision conditions is studied. The aim of the work is modification of template matching method in the case of possible partial distortion of viewing object and in the case of presence of background areas, similar by brightness to the viewing object, in region of interest. As a result of study appropriate methods of relevant problem solution are received. Method of the object points sorting by the distorted and undistorted is offered. Maximum permissible degree of distortion/occlusion of interest object is measured. Modified criterion function, that considers brightness variations, is studied. The results of experimental studies of proposed algorithms are given. Practical recommendations on their use are given.
Key words: object localization, template matching method, difficult vision conditions, object, nonuniform background, criterion function, partial object occlusion, brightness gradient.
Скачать статью