Real-time detection of crop rows in maize fields based on autonomous extraction of ROI

作者：

Highlights：

• A new solution for crop rows detection based on autonomous extraction of ROI is proposed.

• We consider the detection problem of crop rows as the extraction problem of the travelling area of agricultural machinery.

• A robust ROI extraction method is proposed.

• We study the detection results when the image is occluded.

• Experimental results show that the method has a remarkable improvement in real-time performance and accuracy.

摘要

•A new solution for crop rows detection based on autonomous extraction of ROI is proposed.•We consider the detection problem of crop rows as the extraction problem of the travelling area of agricultural machinery.•A robust ROI extraction method is proposed.•We study the detection results when the image is occluded.•Experimental results show that the method has a remarkable improvement in real-time performance and accuracy.

论文关键词：Crop rows detection,ROI,Deep learning,Navigation

论文评审过程：Received 22 December 2021, Revised 2 August 2022, Accepted 10 September 2022, Available online 21 September 2022, Version of Record 25 September 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118826