A Novel Approach to Cashew Nut Detection in Packaging and Quality Inspection Lines

YOLO standing for You Only Look Once is one of the most famous algorithms in computer vision used detecting objects a real-time environment. The newest version this algorithm, namely with seventh or YOLOv7, proposed present study cashew nut detection (good, broken and not peeled) packaging quality inspection lines. Furthermore, algorithm using an efficient convolutional neural network (CNN) to be able successfully detect identify unsatisfactory nuts, such as chipped burnt cashews. In order deal process, new dataset called CASHEW has been built at first by collecting images environments different brightness camera angles ensure model's effectiveness. nuts tested huge number YOLOv7 models their effectiveness will also evaluated. experimental results show that all are obtain high accuracy. Among them, YOLOv7-tiny model employs least parameters, i.e. 6.2M but many output parameters higher accuracy than some other models. As result, approach should clearly feasible solutions cashew’s inspection.