Abstract:The object detection algorithm based on vehicle-mounted lidar is a key component of the perception system on autonomous vehicles. It can provide high-precision and highly robust obstacle information for the safe driving of autonomous vehicles. However, most algorithms are often based on a large amount of point cloud data, which makes real-time detection difficult. To solve this problem, this paper proposes a 3D fast object detection method based on three main steps: First, the ground segmentation by discriminant image (GSDI) method is used to convert point cloud data into discriminant images for ground points segmentation, which avoids the direct computing of the point cloud data and improves the efficiency of ground points segmentation. Second, the image detector is used to generate the region of interest of the three-dimensional object, which effectively narrows the search range. Finally, the dynamic distance threshold clustering (DDTC) method is designed for different density of the point cloud data, which improves the detection effect of long-distance objects and avoids the over-segmentation phenomenon generated by the traditional algorithm. Experiments have showed that this algorithm can meet the real-time requirements of autonomous driving while maintaining high accuracy.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is the real - time, high - precision three - dimensional object detection of vehicle - mounted LiDAR (Light Detection and Ranging) in autonomous vehicles. Specifically, in view of the problem that existing object detection algorithms based on point cloud data are difficult to meet the real - time requirements, the article proposes a fast 3D object detection method. ### Main Problems and Solutions 1. **Real - time and Computational Efficiency** - **Problem**: Most existing point - cloud object detection algorithms directly process a large amount of point - cloud data, which makes real - time detection difficult. - **Solution**: This paper proposes a fast 3D object detection method based on discriminant images and dynamic distance - threshold clustering (DDTC). The detection efficiency is improved through three main steps: 1. **Ground Segmentation**: Use the ground - segmentation discriminant image (GSDI) method to convert point - cloud data into discriminant images, avoiding direct calculation of point - cloud data, thereby improving the efficiency of ground - point segmentation. 2. **Region - of - Interest Generation**: Use an image detector to generate regions of interest for three - dimensional objects, effectively narrowing the search range. 3. **Dynamic Distance - Threshold Clustering**: Design the DDTC method, adjust the distance threshold according to different densities of point - cloud data, improve the detection effect of distant objects, and avoid the over - segmentation phenomenon caused by traditional algorithms. 2. **Accuracy of Distant - Object Detection** - **Problem**: When the object to be detected is far away from the LiDAR, the point cloud on the target surface becomes sparse, and the traditional Euclidean clustering method is prone to cause the over - segmentation problem. - **Solution**: Through the DDTC method, use adjustable parameters to determine the distance thresholds at different distances, effectively solving the over - segmentation problem of distant objects. ### Experimental Verification The experimental results show that this algorithm can maintain a good detection effect in different scenarios and meet the real - time requirements of autonomous driving while ensuring high precision. ### Formula Summary - **Conversion Formula from Point Cloud to 3D Coordinates** \[ x_i = d_i\cos(\alpha_i)\cos(\beta_i)=d(u, v)\cos(v\delta_v)\cos(u\delta_u) \] \[ y_i = d_i\cos(\alpha_i)\sin(\beta_i)=d(u, v)\cos(v\delta_v)\sin(u\delta_u) \] \[ z_i = d_i\sin(\beta_i)=d(u, v)\sin(v\delta_v) \] - **Discriminant Image Generation Formula** \[ \Delta z=\left|d(u, v)\sin(v\delta_v)-d(u - 1, v)\sin(\delta_v(v - 1))\right| \] \[ \Delta d=\cos(v\delta_v)\left|d(u, v)\cos(u\delta_u)-d(u - 1, v)\cos(\delta_u(u - 1))\right| \] \[ R(u, v)=\arctan(\Delta z, \Delta d) \] \[ G(u, v)=d(u, v)\sin(u\delta_u)+H \] \[ B(u, v)=0 \] Through these improvements, this algorithm significantly improves the real - time and accuracy of 3D object detection, and is especially suitable for the safe - driving requirements of autonomous vehicles.

3D Fast Object Detection Based on Discriminant Images and Dynamic Distance Threshold Clustering

Robust vehicle detection using 3D Lidar under complex urban environment

Real-time dynamic obstacle detection and tracking using 3D Lidar

3D Object Detection Based on Extremely Sparse Laser Point Cloud and RGB Images

A Small-Object-Detection Algorithm Based on LiDAR Point-Cloud Clustering for Autonomous Vehicles

Vehicle Detection Based on Point Cloud Intensity and Distance Clustering

3D Vehicle Detection Using Cheap LiDAR and Camera Sensors.

The Research of 3D Point Cloud Data Clustering Based on MEMS Lidar for Autonomous Driving

Obstacle Detection by Autonomous Vehicles: An Adaptive Neighborhood Search Radius Clustering Approach

3D Dynamic Multi-target Detection Algorithm Based on Cross-view Feature Fusion

Robust LiDAR-Based Vehicle Detection for On-Road Autonomous Driving

3D Detection for Occluded Vehicles From Point Clouds

Real-Time 3D Object Detection From Point Cloud Through Foreground Segmentation

Automatic Background Construction and Object Detection Based on Roadside LiDAR

Combined Edge- and Stixel-based Object Detection in 3D Point Cloud

CFPC: The Curbed Fake Point Collector to Pseudo-LiDAR-Based 3D Object Detection for Autonomous Vehicles

Detection and Tracking Dynamic Vehicles for Autonomous Driving Based on 2-D Point Scans.

Roadside LiDAR Vehicle Detection and Tracking Using Range and Intensity Background Subtraction

Research on 3D Point Cloud Object Detection Algorithm for Autonomous Driving

3D Instance Segmentation and Object Detection Framework Based on the Fusion of Lidar Remote Sensing and Optical Image Sensing