AUTHOR=Wang Weiwei , Li Cheng , Wang Kui , Tang Lingling , Ndiluau Pedro Final , Cao Yuhe TITLE=Sugarcane stem node detection and localization for cutting using deep learning JOURNAL=Frontiers in Plant Science VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.1089961 DOI=10.3389/fpls.2022.1089961 ISSN=1664-462X ABSTRACT=Sugarcane seed cutting is an extremely important work process in sugarcane pre-cut seed planting mode, but in the actual work process, there are still a series of problems such as poor seed cutting quality, low seed cutting efficiency and high labor intensity. In order to promote sugarcane pre-cut seed planting technology, the development of an intelligent transverse seed cutting machine for sugarcane pre-cut seed is combined with the development of accurate and rapid identification and cutting of sugarcane stem node. In this paper, we propose an algorithm to improve YOLOv4-Tiny for sugarcane stem node recognition. Based on the original YOLOv4-Tiny network, the three maximum pooling layers of the original YOLOv4-tiny network are replaced with SPP modules to fuse the local and global features of the images and enhance the accurate localization capability of the network. And a 1×1 convolution module is added to each feature layer to reduce the parameters of the network and improve the prediction speed of the network. On the sugarcane dataset, compared with the Faster-RCNN algorithm and YOLOv4 algorithm, the improved algorithm has mean average precision (MAP) of 99.11%, a detection accuracy of 97.07%, and a transmission frame per second (fps) of 30, which can quickly and accurately detect and identify sugarcane stem node. The improved algorithm of this paper is deployed in the sugarcane stem node fast identification and dynamic cutting system to achieve accurate and fast sugarcane stem node identification and cutting in real-time.