Abstract

Leaf plant diseases are difficult to recognize because the symptoms of the disease appear very similar causing farmers to solve problems that don’t match the symptoms of the disease. This research presents the creation of a smartphone application for the identification of lime leaf diseases with deep learning technique. This uses 2 sets image data classified by plant disease specialist, 400 images for algorithm training and 100 images for algorithm testing for each disease. Algorithm development using Convolutional Neural Network that imitates the working of human brain neurons. This uses the supervised learning network type. The algorithm was developed based on the TensorFlow Framwork, Google’s open source library, and was used by Android Studio software to build smartphone applications in the Android operating system. For use, the algorithm that has been learned through the process of learning is embedded into the developed smartphone application. Then it was tested the application accuracy. The results of a smartphone application in the identification of lime leaf diseases found more than 90% accurate.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call