Development of Artificial Intelligence Algorithm based on Digital Image Processing for Calculating Growth Rate of Mushrooms

Abstract

Mushroom growth depends on the microclimate in greenhouses. The environmental control system of greenhouses cannot monitor mushroom growth. Thus, the control of microclimate is not for mushroom growth but for farmers’ feelings or experiences. To develop an intelligent system for monitoring mushroom growth, an artificial intelligence algorithm based on digital image processing was proposed in this study to automatically locate mushrooms and calculate the pileus circle. Compared to the method in the literature, the low-cost image analysis algorithm was used to calculate the pileus circle in the method. The advantage of this method was using low-cost computers or embedded systems which greatly reduces the deployment cost of intelligent image systems and the utilization rate. In the proposed method, the Bayes classifier was used to separate the target from the background to improve the accuracy of the mushroom location. Then, the image preprocessing, Hough transform for circle and self-developed circle-based region matching algorithm were used to locate the mushroom and then determine the mushroom size based on the pileus circle found. In order to verify the effectiveness of the proposed method in terms of the localization accuracy of the mushroom pileus circle, the average accuracy of the proposed method was 87.0%, which was higher than that of the traditional Circle Hough Transform method by 60.7%. Moreover, its localization stability was superior to that of Circle Hough Transform and the average running time of a single image is 2.3 s. Based on the result, the effectiveness of the proposed method meets the practical requirements of mushroom cultivation.