A Deep Learning Based Smartphone Application for Detecting Mango Diseases and Pesticide Suggestions

Abstract

Mango trees are tropical and subtropical trees that flourish in warm climates. It is a popular, tasty fruit as well as a cash crop. Farmers have a hard time selling their products when their output is reduced owing to diseases that affect mango trees. To improve quality and production, it's vital to address any harmful illnesses as soon as possible. This problem prompted the development of novel technologies for detecting and diagnosing mango plant diseases, as well as expert systems for disease prevention. Three machine learning techniques are employed to detect mango diseases in this paper. A dataset with 20 different classes of infected and healthy mango fruit and leaf photos have been created. Among these machine Learning methods, DenseNet169 obtains the highest accuracy of 97.81%, with precision, recall, and F1-scores of 97%, 96%, and 96%, respectively. An Android app has been developed and coupled with the machine learning model that aids in the identification of mango illness as well as the recommendation of pesticides based on disease detection.