TITLE:
Development of Real-Time Crop Sorting System Based on Deep Learning and ESP32-CAM Integration
AUTHORS:
Zul Hasrizal Bohari, Hazim Hafizullah Mohd Zaid, Mohamad Na’im Mohd Nasir, Mohamad Fani Sulaima, Emy Zairah Ahmad, Ahmad Zaidi Abdullah, Muzamir Isa
KEYWORDS:
Image Processing, YOLO, Chili Grading, Smart Agriculture
JOURNAL NAME:
Journal of Power and Energy Engineering,
Vol.13 No.8,
August
21,
2025
ABSTRACT: This study presents the development of a real-time chili sorting system that integrates deep learning techniques with ESP32-CAM hardware to automate the grading of chili peppers based on ripeness and color. Manual chili grading, commonly used in agricultural practices, is often inefficient, inconsistent, and labour-intensive. To address these challenges, a YOLO-based object detection model was trained and deployed to classify chilies into red and green categories with high accuracy. The system incorporates an image acquisition setup using ESP32-CAM, a motorized conveyor system, and a microcontroller-based control mechanism for real-time operation. Experimental results demonstrate that the YOLO model can achieve detection accuracies exceeding 80%, effectively identifying and classifying chilies in dynamic environments. The proposed system enhances sorting precision, reduces processing time, and supports scalable implementation in agricultural supply chains. This research contributes to the advancement of smart agriculture by offering a low-cost, efficient, and scalable solution for post-harvest quality control.