Quality Classification and Segmentation of Sugarcane Billets Using Machine Vision

Machine learning is widely used in agriculture to optimize practices such as planting, crop detection, and harvesting. The sugar industry is a major contributor to the global economy, valuable both as a food source and as a sustainable crop with useful byproducts. This paper presents three machine vision algorithms capable of performing quality classification and segmentation of raw sugarcane billets, developing a proof-of-concept for implementation at our industry partner's mill in NSW. Such a system has the potential to improve quality and reduce costs associated with an essential yet labor-intensive, inefficient, and unreliable process. Two recent iterations of the popular You Only Look Once (YOLO) algorithm, YOLOR and YOLOX, are trained for classification, with the state-of-the-art Mask R-CNN network used for segmentation. The best performing classification model, YOLOX, achieves a classification mAP
50:95
of 90.1% across 7 classes in real time, with an average inference speed of 19.36 ms per image. Segmentation accuracy of AP
50
of 70.8% and AR
50-95
of 83.5% was achieved using the Mask CNN-R network. These results conclusively show that machine learning techniques are capable of accurate classification of sugar cane billet quality.