Implementasi Deep Leraning Lenet Dengan Augmentasi Data Pada Identifikasi Anggrek

Abstract

: Indonesia is a country with a high level of biodiversity, one of Indonesia's biodiversity with 5,000 species of orchids out of 25,000 species of orchids in the world. Orchids are part of natural life that we must care for and protect to maintain its sustainability. This orchid has a very interesting color and shape and is different for each type of orchid. The diversity of these orchids is quite difficult to recognize if you only look at the colors and shapes. This research utilizes deep learning technology, which is an artificial neural network model that has been widely used and developed in digital image recognition. Deep learning can be used as a technology to solve the problem of identifying orchid species. This research aims to see the accuracy value of deep learning model training on LeNet architecture using augmentation techniques on orchid identification. The dataset used is 1600 images and then augmentation is carried out on the dataset so that the data becomes 3200 images. The tools used in the data training process are Google Colab. The results of the study show that the accuracy value on LeNet using rotate augmentation reaches an accuracy rate of 81.88%.