CLASSIFICATION OF CHILI PLANT MATURITY USING TEACHABLE MACHINE: IMAGE-BASED APPROACH

Abstract

Object recognition is an important aspect in the process of retrieving information in humans. Rapid advances in artificial intelligence technology have brought the technology closer to, even beyond, the capabilities of human senses. This research aims to develop a maturity classification model for chili plants using Teachable Machine with an imagebased approach. Chili ripeness is an important factor that influences quality and selling value, so accurate identification is very necessary. In this study, images of chilies at various stages of ripeness (green (unripe), yellow (half-ripe), and red (ripe)) were taken using a smartphone and uploaded to Teachable Machine to train the model. The data collected consists of 300 images divided into three classes. The model is then trained and tested to measure classification accuracy. The research results show that the model is able to achieve classification accuracy of up to 92%, which indicates the effectiveness of this technology in identifying chili ripeness. These findings contribute to the application of artificial intelligence-based technology in agriculture, especially to help farmers determine optimal harvest times. In addition, this research opens up opportunities for the development of a more automated and efficient chili ripeness monitoring system in the future