Application of Convolution Neural Network Algorithm for Rice Type Detection Using Yolo v3: Penerapan Algoritma Convolution Neural Network untuk Deteksi Jenis Padi Menggunakan Yolo v3

Kiki Ahmad Baihaqi (Universitas Buana Perjuangan Karawang)
Yana Cahyana (Universitas Buana Perjuangan Karawang)

Systematics
Vol 3 No 2 (2021): August 2021

Rice is a staple food that contains a lot of energy for human life. There are several types of rice that are often sold in rice shops in general, namely IR42 rice, Pera rice, glutinous rice and Pandan fragrant rice. For now, there are still many people who do not recognize the 4 types of rice, especially millennials, for this reason, research is carried out on the introduction of rice types. The purpose of this study is to make it easier for buyers to identify the type of rice that is in the rice shop so as to minimize fraud by rice traders. The method used in this study is the YOLO (You Only Look Once) v3 method for detecting rice types. The implementation of the image detection process using YOLO (You Only Look Once) v3 has been tested for 12 samples. Based on the results of testing 12 detection experiments on digital image objects, it was obtained 100% where in the picture there were 4 types of rice, 4 grains of rice and 3 types of rice shapes.