摘要
针对概率霍夫直线检测算法检测速率不高、检测精度易受其他特征和连续间断线干扰的缺点,提出了一种改进的概率霍夫直线检测算法。改进后的算法将注意力集中在直线可能存在的若干个区间,提升了检测速率与精度,在直线检测全过程监控并剔除连续间断线,提升了算法的抗干扰能力。经过100组铝锭边缘线检测的实验验证,改进后算法相较原算法平均检测速率提升至1.77倍,查准率提升至1.45倍,且改进后算法在其他图片上的检测速率和检测效果相较原算法也有明显提升。
Aiming at the shortcomings of probabilistic Hough line detection algorithm that the detection rate needs improving and the detection accuracy is easily interfered by other features and conti-nuous discontinuous lines,an improved probabilistic Hough line detection algorithm was proposed.The improved algorithm focused on several possible intervals of the straight line,improved the detection speed and accuracy,monitors and eliminates continuous discontinuous lines in the whole process of straight line detection,and improved the anti-interference ability of the algorithm.In 100 sets of experimental verification of the edge line detection of aluminum ingots,the average detection rate of the improved algorithm is 1.77 times of the original algorithm,while the precision rate is 1.45 times.The detection rate and effect of the improved algorithm on other pictures is significantly improved in comparison with that of the original algorithm.
作者
李建华
宋刘毅
樊高峰
雷春丽
LI Jianhua;SONG Liuyi;FAN Gaofeng;LEI Chunli(College of Mechanical and Electrical Engineering,Lanzhou Univ.of Technology,Lanzhou 730050,China;Yunnan Wenshan Aluminum Co.LTD,Wenshan 663000,China)
出处
《海军工程大学学报》
CAS
北大核心
2024年第1期62-68,共7页
Journal of Naval University of Engineering
基金
国家重点研发计划基金资助项目(2020YFB1713603)。
关键词
直线检测算法
霍夫直线检测算法
注意力
检测速率
line detection algorithm
Hough line detection algorithm
attention
detection rate
