RANCANG BANGUN APLIKASI IDENTIFIKASI PENYAKIT TANAMAN PEPAYA CALIFORNIA BERBASIS ANDROID MENGGUNAKAN METODE CNN MODEL ARSITEKTUR SQUEEZENET

Abstract

The problem that often occurs in the Agricultural Experimental Garden of UdayanaUniversity, especially in the field of agricultural crops in holticulture is about plant diseases, thiscauses a decrease in production results, so the need for early diagnosis of diseases of plants.The study focused on California papaya plants. Diseases of this plant often appear on theleaves and fruits. With advances in technology in the field of image processing can helpproblems that occur in the field of agriculture. In this study build an Android application withCNN (Convolutional Neural Network) method using SqueezeNet architecture. Classifyingdiseases in this plant are Anthracnose, and Ringspot Viruse, as well as classifying healthypapaya. Based on the validation results, the application built using CNN method andSqueezeNet Architecture, can recognize Anthracnose disease, Ringspot Viruse and Papayahealthy through leaves with accuracy of 97% while through fruit accuracy reaches 70%.