Machine Learning-Based Quality Control for Detecting Coffee Bean Defects: A Bibliometric Analysis

The integration of machine learning and computer vision has advanced automated quality control systems in the coffee processing industry. However, limited industrial-scale validation and real-time implementation remain challenges in intelligent coffee bean defect detection. This study aims to identify research trends, technological developments, and research gaps through a Systematic Literature Review (SLR) and bibliometric analysis. Data were retrieved from the Scopus database, comprising 1,630 publications from 2023 to 2026, and analyzed using VOSviewer, Bibliometrix, and Biblioshiny. The results revealed increasing publication trends from 2023 to 2025, with machine learning, deep learning, artificial intelligence, and computer vision emerging as dominant research themes. Keyword co-occurrence analyses highlighted the prominence of neural networks, support vector machines, and automated inspection technologies. The literature synthesis further identified Convolutional Neural Networks (CNN) and YOLO as relevant approaches for coffee bean classification and defect detection. Despite these advancements, challenges remain in real-time deployment, industrial integration, and model validation. This study contributes to identifying research gaps and proposing a research roadmap toward intelligent quality control systems integrating machine learning, computer vision, and Internet of Things (IoT) technologies to support efficient and automated coffee bean inspection within Industry 4.0.