Time-Efficient and Low-Cost Neural Network to detect plant disease on leaves and reduce food loss and waste

Abstract

About 25% of the food grown never reaches consumers due to spoilage, and 11.5 billion pounds of produce from gardens are wasted every year. Diseases are a major cause of food loss as Asia loses 14.2% of its crops due to disease, which is equivalent to 43.8 billion dollars as of 1988-1990. Current solutions involve farmers manually looking for and treating diseased crops. These methods of tending crops are neither time-efficient nor feasible. I used a convolutional neural network to identify signs of plant disease on leaves for garden owners and farmers. The overall accuracy goal of 90% or higher was set to develop the most accurate models. The learning rates of 0.1, 0.01, and 0.001 were tested for the network. I hypothesized that the step size at each iteration or the learning rate of 0.001 would result in the highest accuracy, which was supported through testing. By using machine learning Python libraries, the solution met the standard accuracy goal and was evaluated by several performance metrics, including precision, recall, f-score, specificity, and overall accuracy rates. The model had an accuracy of 95%. By uploading a picture of the crops into the highly-accurate neural network, farmers and gardeners can receive results in seconds on whether or not their crops have a disease, and if they do, which ailment specifically.