Sugarcane Node Identification Based on Structured Learning Model.

Sugarcane node identification is the key techniques for sugarcane cultivation mechanization. The accurate position of nodes that link two consecutive sections should be detected and transferred to microcontroller for cutting. However, current research fails to identify the sugarcane nodes for different kinds of sugarcanes and especially for those under complex background conditions. A novel approach proposed in this work is to recognize nodes of different sugarcanes under complicated background. Firstly, the sugarcane features are extracted, including the target region, target slope and sugarcane node height. Secondly, the edge probability image is generated using the structured learning model, which is trained by a dataset of labeled sugarcane images and dataset BSDS500. Thirdly, the node position is obtained using heuristic line detector. Experiments show the full recognition rate is about 90%, and the location accuracy is less than 36 pixels, which can be further applied to the automation of sugarcane cutting machines.