http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Estimation of Head Position by Multiple Cameras for 3D Position Control of Sound Spot
Kazuki TADA,Hiroshi TAKEMURA,Hiroshi MIZOGUCHI 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
This paper introduces how to form the sound spot in a target person's head by a speaker array. The spherical area, where sound pressure is high on a partial target made by the speaker array, is called sound spot. It was generable in the fixed arbitrary positions until now by using 128ch plane speaker array two or more sets. However, it also has to keep moving with the target person’s motion. In this research, the information on the 3D position of a target person's head is acquired by the image group obtained from multiple cameras, and the sound spot is generated in the head position. By this method, we achieved that the sound spot can move according to a target person.
Path Planning in Consideration of Relative Velocity for Mobile Robot
Ryo OKUMURA,Hiroshi TAKEMURA,Hiroshi MIZOGUCHI 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
This paper presents an avoidance efficiency improvement of path planning for mobile robot in consideration of the relative velocity of an obstacle. The potential method was used as the method of path planning. The potential method is one of the methods of path planning for mobile robot in unknown area. It creates a contour line figure, such as taking the target position low an obstacle high, based on the information from a LRF sensor. In this potential method, the area which takes the potential of an obstacle into consideration of is called “Interference Area”, and it is very important to select a threshold value. This paper, we focus on the form of Interference Area as a threshold value to select. By selecting the form of this Interference Area aims at efficiency improvement of path planning. We confirmed that the proposed Interference Area shape improves efficiency in path planning.
Yoshitaka YAMAUCHI,Hiroshi TAKEMURA,Satoshi KAGAMI,Hiroshi MIZOGUCHI 대한전자공학회 2009 ITC-CSCC :International Technical Conference on Ci Vol.2009 No.7
This paper describes the calculation method of the optimal BBF threshold by performing a simulation which using the background image, a real object image and a 3D model image. The background image is captured by a mobile robot camera, and the object image is also captured by the same camera. As a result of the simulation, we confirmed that more inliers could be passed than the conventional threshold, in case SIFT keypoints are the poses of the model extracted.