Navigation line extraction based on root and stalk composite locating points. (June 2021)

• Record Type: Journal Article
• Authors: Gong, Jinliang; Wang, Xiangxiang; Zhang, Yanfei; Lan, Yubin; Mostafa, Kazi
• Is Part Of: Computers & electrical engineering. Volume 92(2021)
• Publication Date: 2021-06
• Subjects: Machine vision -- Composite locating point -- Robust regression -- Navigation line; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2021.107115
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Navigation line extraction was difficult for 6–14 leaf corn crop environment. By white pixels proportion of total pixels index to get stalk location points. By maximum value of column pixel adding curve to get root locating points. Navigation line was extracted effectively using stalk and root locating points. Abstract: The presence of weeds undermines the accurate algorithmic detection of navigation lines in cornfields, which plays a pivotal navigational role in the operation of intelligent agricultural machinery for tasks such as weeding and spraying. A method was proposed for extracting navigation lines by using the composite locations of the root and stalk of corn in the 6th- to 14th-leaf stages growing in a complex environment. The positional and area features were first used to eliminate the interference regions and obtain the root locations. The white pixel percentage index was used to filter the line segments obtained through the Hough transform and thereby determine the stalk locations. Experiments showed that the navigation lines were detected at an accuracy of 93.8%, an improvement of 10.2% over that achieved using conventional methods. Graphical abstract: Image, graphical abstract