论文部分内容阅读
为了解决激光成像雷达距离像匹配中的尺度和旋转不变性问题,提出了一套完整的距离像匹配方法,主要包括图像比例调整、Krawtchouk矩不变量提取、构造特征空间及搜索匹配4部分内容。以仿真激光成像雷达距离像为实验数据,通过调整实时图与参考图之间的比例缩放、提取两图的Krawtchouk矩不变量并以此构造特征向量进行匹配。实现了仿真激光雷达距离像的景象匹配。实验结果表明,该算法在一定比例的缩放和任意角度旋转的情况下,匹配正确率和匹配精度都较高,算法的抗高斯噪声能力较好。
In order to solve the problem of scale invariance and rotation invariance in distance image matching of laser imaging radar, a complete distance image matching method is proposed, which mainly includes image proportion adjustment, Krawtchouk moment invariant extraction, constructing feature space and search matching 4 parts. The simulated laser imaging radar distance image is used as the experimental data, and the Krawtchouk moment invariants of the two images are extracted by adjusting the scaling between the real-time image and the reference image to construct the eigenvector. Realize the scene matching of distance simulation of laser radar. Experimental results show that the proposed algorithm has high matching accuracy and matching accuracy under a certain scale of scaling and rotation at any angle. The proposed algorithm has better anti-Gaussian noise performance.