一种3D点云数据转换2D图像的皮带质量检测算法  

A Belt Quality Detection Algorithm for Converting 3D Point Cloud Data Into 2D Images

作　　者：吉日格勒 刘畅 Jirigele;Liu Chang(Zhunneng Coal Preparation Plant,Inner Mongolia Ordos 017000)

机构地区：[1]准能选煤厂,内蒙古鄂尔多斯017000

出　　处：《山东煤炭科技》2023年第3期205-207,共3页Shandong Coal Science and Technology

摘　　要：皮带质量的好坏关系到煤料运输的健康状态,提出了一种监控皮带质量的检测算法。针对皮带质量形成的缺陷检测精度以及皮带宽度,采集3D点云数据,将获取的3D点云数据的深度数据转化为2D图,使用YOLOV5目标检测算法检测出2D图的皮带质量缺陷。实验证明,提出的3D点云数据转化为2D的检测方法,大幅度提高了皮带质量检测速度以及检出率。The quality of belt is related to the health condition of coal material transportation.A detection algorithm for monitoring the quality of belt is proposed.For the defect detection accuracy and belt width caused by belt quality,3D point cloud data is collected,convert the acquired depth data of 3D point cloud data into 2D image,and the belt quality defect in 2D image is detected using YOLOV5 target detection algorithm.The experiment shows that the proposed 3D point cloud data transformation into 2D detection method greatly improves the detection speed and detection rate of the belt quality.

关 键 词：皮带质量 点云数据 2D图 目标检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]