Ayu Sarah Annisa
Department of Physics, Faculty of Mathematics and Natural Sciences, Andalas University, Padang, Indonesia, 251613

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Classification of Box Sizes in Automated Systems Using YOLOv8 and Computer Vision Ayu Sarah Annisa; Meqorry Yusfi
Journal of Physics and Its Applications Vol 8, No 2 (2026): May 2026
Publisher : Diponegoro University Semarang Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.14710/jpa.v8i2.29716

Abstract

The logistics and distribution industry requires a fast and accurate automated sorting system to improve operational efficiency. This research develops a computer vision-based automated sorting system using YOLO (You Only Look Once) to detect and classify box sizes in real time. The system consists of an ESP32-CAM as a visual sensor, an ESP8266 NodeMCU as a microcontroller, and a servo motor as an actuator using the MQTT communication protocol. The detection results are sent through a local MQTT broker for low latency processing without the internet. The YOLOv8 model used successfully achieved a detection and classification accuracy of 98.15%. The top camera showed more stable performance (89-96%) than the front camera (83-96%) due to the influence of distance and angle of image capture. The tests were conducted under fixed lighting conditions and only distinguished between small (< 5×5×5 cm) and large (≥ 5×5×5 cm) boxes, with a maximum load limit of 700 grams. The system is still limited in classifying objects close to the size limit, and is not optimal for variable lighting.