Authors : Kwang-Hyoung Lee , Jong-Hee Lee and Keun-Wang Lee
Abstract: We propose a method for extracting and tracking objects from real-time images using adaptive background imaging. Adaptive background images are generated by removing background noise from images and updating the background area, not the object area, to extract objects strong against lighting. The objects are then extracted using the difference between the background image and the input image from the camera. A minimal rectangular area is set using the internal points of the extracted object and utilized for tracing objects. The implemented system was developed to monitor objects moving in real-time through Web and mobile devices. The results of performance tests on the proposed method are compared with those of existing tracking algorithms and an assessment is given.
Kwang-Hyoung Lee , Jong-Hee Lee and Keun-Wang Lee , 2005. A Real-time Digital Video Recorder System in Mobile Environments . Asian Journal of Information Technology, 4: 809-815.