Recognition of Food Material and Measurement of Quality using YOLO and WLD-SVM
Bima Sena Bayu Dewantara, Azifah Zusrina Devy, Mochamad Mobed Bachtiar, Setiawardhana Setiawardhana
Abstract
This paper presents the development of a system for recognizing types of food materials and measuring their quality visually using a camera. The food material can be in the form of meat, vegetables, fruits and other processed food materials such as tempeh and tofu. We build a food material recognition system using YOLOv3-tiny, and measure the quality of food ingredients using the Weber Local Descriptor (WLD) feature which is then classified using the Support Vector Machine (SVM). Based on the results of experiments that have been carried out, the results of the recognition of food materials using YOLOv3-tiny reached 82.02% and the performance of food quality measurement results using WLD-SVM reached 83.33%.