Detection of Wild Mushrooms Using Machine Learning and Computer Vision

Over the past several centuries, as the global population has experienced a significant increase, there has been a growing need to expand agricultural production and focus on improving the quality of agricultural goods. Contemporary society places emphasis on environmentally friendly practices, sustainable production, and minimally fertilized biological products. With the rapid advancement of machine learning algorithms, precision agriculture has the potential to utilize a wide range of innovative solutions. One such algorithm, YOLOv5 (You Only Look Once), is capable of recognizing objects with high precision in real-time. The identification of wild mushrooms is of significant practical and scientific importance, as certain species are edible and can serve as a viable food source. This research presents a novel architecture utilizing multispectral images and experimental findings from the Yolov5 algorithm on a unique dataset consisting of wild mushroom biomass, including Macrolepiota Procera, with the goal of enhancing the resilience of precision agriculture.