Detecting Common Coccinellids Found In Sorghum Using Deep Learning Models

This article discusses the development of automation technologies for field scouting in order to meet the demands of a growing human population. Specifically, the article focuses on the development of deep learning software to detect and classify seven coccinellids commonly found in sorghum. The two-stage object detection model, Faster Region-based Convolutional Neural Network (Faster R-CNN) with the Feature Pyramid Network (FPN), and one-stage detection models in the YOLO (You Only Look Once) family (YOLOv5 and YOLOv7) were used to detect and classify the coccinellids.