Klasifikasi Jenis Tanaman Sayuran Selada Menggunakan Metode Transfer Learning

Abstract

One of the causes of malnutrition in Indonesia is the lack of vegetable consumption among the population. One of the tasks of the Health and Nutrition sector is to conduct research related to addressing malnutrition in Indonesia. One of the vegetables that is frequently studied is lettuce. To perform more accurate and informative nutritional analysis, lettuce can be used as an example of a readily available vegetable in Indonesia. In this research, Transfer Learning method is used to classify different types of lettuce, including Butterhead Lettuce, Green Romaine Lettuce, Green New Grand Rapid Lettuce, Oakleaf Green Lettuce, Red Rapid Lettuce, and Red Romaine Lettuce. The dataset comprises 1,242 data points, with 993 for training and 249 for validation. Additionally, 72 testing data points were captured using a smartphone. The utilization of Transfer Learning in classifying lettuce types resulted in an accuracy rate of 93%. These results demonstrate that the system developed using Transfer Learning has effectively classified various types of lettuce plants.