Kevin Arya Rizki Mahesa
Politeknik Negeri Banyuwangi

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

Found 1 Documents
Search
Journal : journal of deep learning computer vision and digital image processing

Design and Development of an Internet of Things-Based Automatic Chicken Egg Sorting System for Real-Time Sorting and Monitoring Kevin Arya Rizki Mahesa; Endi Sailul Haq; Indra Kurniawan; Galih Hendra Wibowo; Arif Fahmi
Journal of Deep Learning, Computer Vision, and Digital Image Processing Volume 4 Issue 3 September 2026
Publisher : CV. Sakura Digital Nusantara

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.61255/decoding.v4i3.1569

Abstract

Purpose – This study develops and evaluates an IoT-based egg sorter integrating weight grading, gas-response screening, automatic diversion, and monitoring. Methods – The prototype combined an ESP32, a load cell with HX711, an infrared sensor, an MQ-135 sensor, and a servo actuator. Records were transmitted to a PHP REST API and MySQL database on a local server. Evaluation covered ten calibration masses, ten egg-weight measurements, separate gas-screening sample sets, infrared detection, ten end-to-end routing trials, server communication, and manual-versus-automatic sorting time. Findings – Mean relative error was 0.33% for calibration masses and approximately 0.24% for egg measurements. The MQ-135 separated the selected good and spoiled sample groups using a threshold of 250 ppm-equivalent sensor-library units. All ten end-to-end trials routed eggs according to the assigned category. Server response averaged 1.17 s (SD 0.09 s). Sorting time decreased from 7.35 to 3.18 s per egg, a reduction of 56.73%. Research implications – The small samples support preliminary functional feasibility. Routing agreement does not establish independently verified classification accuracy. Gas response was not traceable to a calibrated concentration standard, and several environmental and sensor-preparation conditions were unrecorded. Larger controlled studies are needed to validate quality screening and boundary-case grading. Originality – The design integrates weight sensing, gas-response screening, mechanical diversion, and local-server records within one low-cost egg-sorting workflow.