论文部分内容阅读
根据智能车辆主动驾驶辅助系统中的重要性,提出了一种融合毫米波雷达数据和视觉多特征的车辆检测算法。车辆检测算法通过三个步骤实现,首先,提出一种空间对准算法实现毫米波雷达和视觉的空间对准;其次,根据空间对准结果和搜索策略提取目标车辆的感兴趣区域;最后,融合车底阴影、对称轴、左右边缘等车辆特征实现车辆检测,其中,为了准确得到目标车辆的车底阴影,提出一种改进的车底阴影分割算法。算法的性能在不同的场景下得到证实,实验结果表明该车辆检测算法是有效和可靠的。
According to the importance of active vehicle driving assistant system, a vehicle detection algorithm based on millimeter wave radar data and visual multi-feature is proposed. Firstly, a spatial alignment algorithm is proposed to realize spatial alignment between millimeter-wave radar and vision. Secondly, the region of interest of the target vehicle is extracted based on the spatial alignment result and the search strategy. Finally, the fusion Vehicle bottom shadow, symmetry axis, left and right edges and other vehicle features to achieve vehicle detection, in order to accurately get the target vehicle shadow at the bottom of the car, an improved vehicle shadow segmentation algorithm is proposed. The performance of the algorithm is verified under different scenarios. The experimental results show that the algorithm is effective and reliable.