基于GM-APD激光雷达数据融合的小目标检测  

Small object detection based on GM-APD lidar data fusion

作　　者：杜达宽 孙剑峰[1] 丁源雪 姜鹏 张海龙[1] DU Dakuan;SUN Jianfeng;DING Yuanxue;JIANG Peng;ZHANG Hailong(National Key Laboratory of Science and Technology on Tunable Laser,Institute of Opto-Electronic,Harbin Institute of Technology,Harbin 150001,China;Science and Technology on Complex System Control and Intelligent Agent Cooperation Laboratory,Beijing 100074,China)

机构地区：[1]哈尔滨工业大学光电子技术研究所可调谐(气体)激光技术重点实验室,黑龙江哈尔滨150001 [2]复杂系统控制与智能协同技术重点实验室,北京100074

出　　处：《光学精密工程》2023年第3期393-403,共11页Optics and Precision Engineering

基　　金：国防重点实验室基金资助项目。

摘　　要：GM-APD激光雷达具有单光子探测灵敏度,大幅降低了系统体积和功耗,但受像元数限制,难以获得远距离小目标清晰轮廓,目标检测率不高。针对该问题,提出了基于强度像和距离像多级处理的小目标深度学习检测算法,充分挖掘强度图像和点云特征信息及相互关联性,提高小目标检测概率。通过改进特征金字塔网络,将感受野模块和注意力机制模块与特征提取网络相结合,增强强度像初筛目标准确性,在候选区域内将强度像与距离像融合成带有强度信息的四维点云。然后,使用动态图卷积网络对候选区内目标进行二次检测,利用点云信息进一步筛选候选框内的目标。经GM-APD激光雷达远距离车辆数据集测试,网络的检测准确率达到98.8%,对于车辆结构不完整,车辆回波弱,背景存在强反射光斑等复杂场景有很好的鲁棒性。相较于SSD,YOLOv5等较为先进的目标检测网络,检测准确率分别提升了3.1%与2.5%,该算法为激光雷达弱小目标检测识别提供了一种可行性解决方案。Geiger mode avanlanche photon diode(GM-APD)lidar has single photon detection sensitivity,which greatly reduces the system volume and power consumption.It makes the system feasible for practical application,and has become a hot topic in recent studies.However,owing to the limitation of the pixel number,the spatial resolution is low,which makes it difficult to obtain the clear contour of the remote target,and the object detection rate is not high.To solve this problem,a detection algorithm based on multilevel processing of the intensity and range images was proposed to find the correlation between the intensity images and point clouds’features to improve the probability of small object detection.First,the improved feature pyramid network(FPN)combines the receptive field block(RFB)and convolutional block attention module(CBAM)with the feature extraction network to enhance the selection accuracy of intensity images.Second,the intensity and range images are combined into point clouds with intensity information in the candidate regions.Finally,a dynamic graph convolution network(DGCNN)is used to perform secondary detection on the target in the candidate regions.Moreover,point cloud information is used to further select the object in the candidate regions.In the GM-APD lidar long-range vehicle dataset,the AP of the network achieves 98.8%,and it has good robustness for complex scenes,such as incomplete vehicle structure,weak echo,and strongly reflected light spot.Compared with the SSD and YOLOv5,the detection accuracy of the network improved by 3.1%and 2.5%,respectively,which is feasible for lidar dim object detection.

关 键 词：激光雷达 目标检测 感受野 注意力机制 动态图卷积神经网络 

分 类 号：TN958.98[电子电信—信号与信息处理]