IoT Enabled Mushroom Farm Automation Based on CNN-BiGRU-CRF Method

Mushroom farming's huge contributions to the global market are echoed in the latest data on agriculture. As a result, mushroom farms and cultivation are gaining in popularity each year. The impact of temperature and humidity on mushroom growth is significant, making weather monitoring and management an important aspect of mushroom cultivation. Most mushroom farmers throughout the world still employ time-tested methods from their grandparents' generations. Due to insufficient monitoring of the weather and the cultivating process, deadly mushrooms frequently occur while using the old method. In this method, the proposed system provide the framework for intelligent mushroom farming that makes use of the Internet of Things and machine learning. Incorporating RMM, farm automation, and mushroom categorization, the suggested system is a first of its kind. Microcontrollers, “the Internet of Things,” Smart monitoring and farm automation are made possible by ESP32 and several agriculture-related sensors. The water supply is adjusted based on the mushroom's specific humidity and temperature needs at each stage of growing. The suggested method includes preprocessing, edge detection, feature extraction, and model training following data collection. The suggested method employs reflectance calibration, normalization, and spectral smoothing for preprocessing. It employs GLCM for usage in feature extraction. CNN-BiGRU-CRF is employed to assess the model's efficiency. When compared to conventional CNN and CNN-BiGRU, the proposed method fares quite well.