Analysis on telepointer tracking methods for endoscopic images
Rohana Abdul Karim, Mohd Marzuki Mustafa, Mohd Asyraf Zulkifley · ARPN Journal of Engineering and Applied Sciences · 2015
Recently, the growth of endoscopic surgery for accelerating the operation performance with minimum pain regardless of the distance limitations becomes the crucial agenda for enhancing the health care qualities. Therefore, telemedicine is one of the best alternatives to break the limitation. On top of that, telepointer is a well-known supported tool for remote guided used in telemedicine as well as to facilitate the inexperienced physician at the remote place during surgery operation. The function of telepointer is to label the tissues and organs of interest (TOB). However, the non-rigid shape and movement of tissues and organs lead to lose the pointer track. Unfortunately, there are only a few studies has been done on telepointer tracking, especially for endoscopic application. In this paper, we analyzed three methods to keep track the TOB and maintenance the pointer towards the TOB by using feature matching approached as an observation data. Therefore, our aim is to evaluate the most effective method of telepointer tracking. Three techniques have been compared, which are least square method (LMS), Delaunay triangulation with line intersection (DT) and maximum probability keypoints with scale circles intersection (PMAX). Simulation results show that the most effective technique is PMAX with average mean error distance MED lower than six pixels followed by DT and LMS. Hence, the tracking system will allow the physician to track the pointer better especially in noisy environment.