Using deep learning algorithms and a flatbed scanner to assess rice quality

This article presents the results of research on the application of deep learning techniques in the automatic assessment of rice grain quality. A measurement methodology and a computer program were developed, which uses a deep learning model to identify individual grains in an image and detect impurities for further qualitative analysis. The program was implemented in Python using the OpenCV 4, Numpy, and Ultralytics YOLO libraries. The study used a flatbed scanner, enabling the identification of approximately 3,000 objects in a single scanner ruler pass.