A new Solar-IoT Based Method for Mushroom Cultivation

Abstract

Although, Mushroom cultivation is a growing industry for recent years, but maintaining proper conditions in mushroom farms may be challenging, especially for small-scale farmers without accessibility to modern controller systems. In this paper, a new IoT-based approach based on Blynk platform for mushroom cultivation is presented. The Blynk dashboard includes gauges for monitoring humidity and temperature in real time, as well as sliders for controlling the environment situations. The sliders are connected to relays which are powered by a solar power system. This mechanism making it a proper option for using in a remote locations without access to main grid, also the low cost of the system makes it an affordable option for small-scale farmers. For improving the identification characteristic, a hybrid method consists of differential evolution algorithm and wavelet transform is used. The model is a classification model trained on a dataset of healthy and diseased mushrooms, which can determine the real-time conditions of mushrooms. For practical applications, reducing the number of extracted features and minimizing redundancy features have great importance. In the proposed method, features are extracted from the original data by applying differential evolution algorithm. Then these features were evaluated by methods such as logistic regression, k-nearest neighborhood and decision tree. It was found that among the specified features, the first two features covered more than 89% of the variance of the entire set. Then the wavelet transform method is applied to get final identifications. Finally, by reducing the number of features to 2, a considerable amount of calculations is reduced and only about 4% of the accuracy provided for the estimated results is reduced.