广西科技大学学报2018，Vol.29Issue(3)：71-76,6.DOI:10.16375/j.cnki.cn45-1395/t.2018.03.011
一种改进的田间导航特征点提取算法
An improved feature point extraction algorithm for field navigation
摘要
Abstract
An improved feature point extraction algorithm is proposed to solve the problem of the low precision of the navigation line,which takes the midpoint of the field image background pixel coordinates as the navigation fea-ture point. This algorithm uses the normalized 2G-R-B ultra green feature to grayscale the image. And it uses the OSTU method and the morphological closed operation to separate and filter the gray image respectively. Then,the midpoint of the background pixel coordinates in each row is extracted as the initial feature point. The feature points are corrected according to the number of pixels between the initial feature points and the center line of the image. Fi-nally,the navigation line is obtained by fitting the modified feature points. The simulation results show that the im-proved algorithm is less affected by environmental noise and has good real-time performance. The fitted navigation line has obvious advantages in accuracy.关键词
特征点/导航线提取/农机导航Key words
feature point/navigation line extraction/agricultural machinery navigation分类
农业科技引用本文复制引用
陈艳,李春贵,胡波..一种改进的田间导航特征点提取算法[J].广西科技大学学报,2018,29(3):71-76,6.基金项目
国家自然科学基金项目(61302178)资助. （61302178）
