Vol. 7, Issue 8, Part J (2025)

The agri-food sector is undergoing a technological revolution, with computer vision emerging as a key enabler in the automated grading and sorting of fruits and vegetables. Traditionally reliant on labour-intensive and subjective manual grading, the industry now leverages image processing, artificial intelligence, and machine learning to assess produce quality with enhanced precision, consistency, and efficiency. Computer vision systems evaluate features like size, shape, color, texture, and defects-enabling real-time classification and reducing post-harvest waste. Advanced deep learning models such as CNNs, YOLO, and MobileNetV₂ have demonstrated classification accuracies exceeding 95%, significantly outperforming traditional methods. These innovations not only cut operational costs but also improve market value, food safety, and sustainability. However, challenges like lighting sensitivity, high setup costs, and integration complexities persist. Despite these barriers, the technology’s benefits-faster throughput, reduced human error, better quality control, and enhanced decision-making-underscore its transformative potential. The paper explores current applications, underlying technologies, and prospects of computer vision in agriculture. As the sector evolves, the convergence of AI and imaging technologies promises to reshape food production and supply chains, ensuring higher standards and promoting sustainable agricultural practices worldwide.