Research on Asparagus Recognition Based on Deep Learning

In recent years, there have been several outbreaks of extensive flooding in northern provinces. Asparagus stem blight disease and brown spot diseases have become increasingly serious, significantly reducing the yield of asparagus grown in large fields, and asparagus production has developed toward facility agriculture. To solve the problems of high labor costs, labor shortages, and low production efficiency faced by facility agriculture, the adoption of mechanized harvesting for asparagus is an inevitable trend. A prerequisite for mechanized harvesting is target detection. Compared with algorithms such as Faster R-CNN, which require the use features of candidate regions for classification and recognition and cannot meet the real-time requirements of mechanized harvesting, this paper proposes a YOLO-based asparagus recognition scheme that can quickly perform target detection with a detection accuracy of 85.45% and significantly enhanced interference resistance, which can greatly improve the production efficiency of facility agriculture and accelerate the mechanization process of facility agriculture.