查看文章

The advancement of science and technology, with the help of machine learning in agriculture applied to mobile devices, presents an ideal and convenient way for farmers in fruit and crop production. The traditional way of assessing plant diseases needs expert knowledge, and it can be costly, time-consuming, and requires extra work. Hence, the proponent's objective of this study is to develop a mobile application for lanzones leaf disease identification and severity detection using Convolutional Neural Networks MobileNet architecture. The mentioned algorithm is pre-trained using datasets of manually gathered lanzones leaf images classified as Algal Spot, Scale Insect, Healthy, and Unknown leaves and augmented by the researchers. The pre-trained network implements transfer learning for the successful development of the model. A total of 120 samples, 30 for each health condition were tested to evaluate the …