Research on Edge Detection of LiDAR Images Based on Artificial Intelligence Technology

LiDAR works by emitting laser pulses and measuring their reflection times to accurately obtain threedimensional spatial information, thus generating high-resolution point cloud data and images. However, the application of LiDAR images faces numerous challenges, particularly in edge detection, where traditional methods often fail to meet practical needs due to insufficient detection accuracy and high computational complexity.Edge detection, as a crucial step in image processing, directly impacts subsequent tasks such as image segmentation, object recognition, and scene understanding[1]. Accurate edge detection can improve target recognition accuracy, optimize navigation path planning, and enhance environmental perception reliability. Therefore, studying an efficient and accurate LiDAR image edge detection method has significant theoretical value and application prospects.Existing edge detection methods, such as the Canny and Sobel algorithms, perform well on conventional images but often struggle with the unique noise characteristics and data structure of LiDAR images. With the rapid advancement of artificial intelligence technology, deep learning has achieved remarkable results in image processing. However, applying deep learning to LiDAR image edge detection still faces challenges such as complex data preprocessing, high difficulty in model training, and significant computational resource demands. Hence, there is an urgent need for an innovative AI-based edge detection method to address these challenges. This study aims to explore and develop an AI-based edge detection method for LiDAR images. The main research contents include: 1. Reviewing the current state of LiDAR technology and its application in edge detection.