应用于汽车主动安全的三维目标识别算法研究  

Application to the study of 3D target recognition algorithm of vehicle active safety

作　　者：邓俊文 李正浩[1,2] 韩鹏 王东强 夏静满 

机构地区：[1]重庆大学光电工程学院,重庆400044 [2]重庆市科学技术研究院,重庆401123 [3]重庆长鹏实业集团(有限)公司,重庆400082

出　　处：《自动化与仪器仪表》2015年第3期73-74,77,共3页Automation & Instrumentation

摘　　要：针对车载条件下获取的深度图像场景复杂,部分深度信息缺失的问题,本文引入了鲁棒的方位直方图特征和高效的三维霍夫投票算法,可以在场景复杂,部分深度信息缺失的条件下实现三维目标识别,为车载条件下的三维目标识别提供了解决方案。Range images acquired by sensors on the vehicle generally have a set of nuisances, including clutter, occlusion and holes. We introduced a robust feature descriptor named signature of histograms of orientations and an efficient 3D Hough voting algorithm for recognizing 3D objects in the presence of clutter, occlusion and holes. This work is a potential solution for handling 3D objects recognition using on-board equipment in vehicles.

关 键 词：汽车主动安全 三维目标识别 特征描述符 点云 

分 类 号：TN244[电子电信—物理电子学]