Open Access Journal of
eISSN: 2575-9086
Submit manuscript...
Literature Review Special Issue Palm Oil in Health and Technology
AI-driven quality control and ripeness classification in the palm oil industry: a systematic review of computer vision and deep learning methods
Loso Judijanto
Received: March 02, 2026 | Published: March 24, 2026
Citation: Judijanto L. AI-driven quality control and ripeness classification in the palm oil industry: a systematic review of computer vision and deep learning methods. Open Access J Sci. 2026;9(1):107-118. DOI: 10.15406/oajs.2026.09.00294
Abstract
The growing need for consistent and technology-enhanced quality assurance in the palm oil sector has stimulated increasing adoption of artificial intelligence (AI), particularly for automated ripeness classification of fresh fruit bunches (FFB). Variability in manual grading and environmental conditions has encouraged the development of computer vision and deep learning approaches to improve objectivity and operational reliability. This study aims to systematically identify, evaluate, and synthesize peer-reviewed evidence on AI-driven quality control and ripeness classification in the palm oil industry, with emphasis on algorithmic trends, dataset characteristics, performance metrics, and deployment orientations. A Systematic Literature Review (SLR) design was applied using the Scopus database. A structured Boolean query and predefined inclusion–exclusion criteria were implemented. From an initial broad search yielding 245 records, progressive refinement, time filtering (2020–2026), language screening, and open-access eligibility assessment resulted in 37 studies for final synthesis. Data were analyzed through thematic categorization, comparative performance mapping, and methodological trend evaluation. The findings reveal a clear transition from classical machine learning to deep convolutional neural networks, with transfer learning adopted in 59% of deep learning studies. Reported accuracy commonly exceeded 90%, with a weighted mean of 94.2% for deep models. Increasing attention to deployment architectures, including conveyor-integrated and mobile-based systems, indicates growing industrial applicability. Overall, AI-driven computer vision demonstrates robust capability for ripeness classification. Future research is encouraged to strengthen dataset standardization, cross-location validation, and openscience practices to enhance reproducibility and scalability.
Keywords: artificial intelligence; computer vision; deep learning; palm oil industry; ripeness classification
©2026 Judijanto. This is an open access article distributed under the terms of the, which permits unrestricted use, distribution, and build upon your work non-commercially.