Answers are grounded in this paper and cite their sources.
This study explores a Computer Vision approach to identify inter-row planting in soybean areas. Related work already explores the same problem, but our work differs by focusing on inter-row identification to support the alignment of weeding machines (commonly used by small farmers who produce organic products). We created an experimental database with images collected with a camera attached to a weeder. The planting lines and inter-rows were manually labeled. To detect planting lines and inter-rows, we use two segmentation algorithms based on Convolutional Neural Networks (Mask R-CNN and YOLACT), achieving an accuracy of up to 0.656 with the interpolation of the obtained segmentation results. The segmentation results obtained made it possible to estimate the inter-rows satisfactorily. We provide a database of collected images, with the planting lines and inter-rows noted. With these results, we intend to create a solution in future work that allows automatic alignment of the weeder. We also plan to develop similar solutions for other crops (in addition to the soybeans explored in the experiments).
No takes yet. Share an insight, caveat, or question.
Panizzon et al. (2024) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: