Machine Vision Recognition System of Edible and Poisonous Mushrooms Using a Small Training Set-Based Deep Transfer Learning

A way of classifying if a mushroom is edible or not is presented in this study. As mushrooms are slowly becoming popular, classifying these mushrooms would be crucial as some of the toxic mushrooms that could be found in the mushrooms can kill a person or give them a bad case of stomachache and other effects. Using the YOLOv3 model a model is created that can classify these mushrooms. The model that was chosen has gotten an mAP score of 96.68% and can detect most of the inputs that are used to test the model. The model is also able to achieve a 90% accuracy as it was able to correctly detect 18 photos out of 20 when tested. This model could be used to ensure that there would not be any toxic mushrooms in houses or parks that a child could just pick up and swallow without being noticed by any adults.