CAJ | 학술논문

针对智能交通系统提出的夜间车流检测问题，提出了一种基于视频虚拟线圈的检测方法。该方法主要包括车灯配对和车辆检测两个部分。车灯配对是根据同一车辆的两个车灯大小和相对位置等特性对车灯进行配对；车辆检测是根据车灯配对情况输出车流量。结果表明，该算法取得了良好的检测效果，具备较好的鲁棒性，并且能满足实时性的要求。
침대지능교통계통제출적야간차류검측문제，제출료일충기우시빈허의선권적검측방법。해방법주요포괄차등배대화차량검측량개부분。차등배대시근거동일차량적량개차등대소화상대위치등특성대차등진행배대；차량검측시근거차등배대정황수출차류량。결과표명，해산법취득료량호적검측효과，구비교호적로봉성，병차능만족실시성적요구。
A video detection method based on virtual coil is put forward for traffic flow detection problem at night in intelligent transportation system. This method mainly consists of headlights matching part and vehicle detection part. The headlights matching part worked according to the size and the relative position of the two lights in the same car. Vehicle detection part exports the traffic volume according to the situation of the headlights matching. The result shows that the algorithm achieves good test results and has good robustness. It can satisfy the real-time requirement.