| 摘要 |
<p>PURPOSE:To make it possible to separate and identify two close targets by separately providing a front-edge detecting sliding window SW and a rear-edge detecting SW, and setting the size of the rear-edge detecting SW so that the size is smaller than the size of the front-edge detecting SW. CONSTITUTION:A quantized video (a) from a quantized video memory 1 is sent into a front-edge counter 2 and a rear edge counter 3. A front-edge detecting SW size WLE is inputted into the counter 2, and a rear-edge detecting SW size WTE is inputted into the counter 3. When the front-edge detecting SW reaches a position P12, the front edge of a first target (b) is detected. When said SW reaches a position P14, the front edge of a second target (c) is detected. Meanwhile, when the rear-edge detecting SW reaches a position P13, a target count becomes equal to a rear-edge detecting reference value KTE in a window, and the rear-edge detected signal (e) of the first target is outputted. At this time, since the size WTE is smaller than the size WLE, the video of the front edge part of the second target does not enter into the window, and the identification of the rear edge of the first target is not hindered. When the rear-edge detecting SW reaches a position P15, the rear edge of the second target is detected, and the two targets can be separated and identified.</p> |
