抄録
Recently, Robot Technology (RT) is widely expanded from the industrial to other areas. Home service is one of the popular trends of the applying RT into our daily life. In Intelligent space environment, human posture is one of the useful information for interpretation of our intentions and needs. In this paper, we propose a novel feature for human posture recognition which is DFT (Discrete Fourier Transform) Coefficients of log-polar histogram. DFT Coefficients of log-polar histogram virtue is providing rotation and scaling. Moreover, our system implemented based on Top-view Vision. The top-view vision enables our system to observe the whole area with a single camera, and also avoid obstacle problems in horizontal view which are the serious issues of front-view vision in the real environment like intelligent space. In the initial, we have experimented the system to recognized four different basic human behaviors (lying, standing, sitting and moving).
