| 摘要 |
<P>PROBLEM TO BE SOLVED: To precisely predict the positional relationship of an object and its own vehicle without increasing any calculation load. <P>SOLUTION: A camera 2 picks up a two-dimensional image in the periphery of its own vehicle, and an on-image featured point following part 3 extracts a featured point from the two-dimensional image picked up by the camera 2, and acquires time series data Pi of the featured point by following the extracted featured point in a prescribed time, and an on-image moving position predicting part 4 calculates the movement prediction value of the featured point by using the time series data Pi of the featured point, and predicts positions(x<SB>i</SB>, y<SB>i</SB>) of the featured point in the two-dimensional image according to the calculated movement prediction value. <P>COPYRIGHT: (C)2006,JPO&NCIPI |
