Rice-ResNet: Rice classification and quality detection by transferred ResNet deep model

Efficient classification and quality assessment of rice varieties are essential for market pricing, food safety, and consumer satisfaction in the global rice sector. Leveraging pre-trained ResNet architectures, Rice-ResNet significantly enhances feature extraction, ensuring accurate classification and quality detection of rice cultivars. This system, accessible in Python repositories, promises improved crop management and yield. Despite requiring real-world implementation, Rice-ResNet marks a significant advancement in rice classification, fostering enriched digital experiences. •Rice-ResNet utilizes deep ResNet models to accurately classify rice varieties, crucial for global agriculture.•Rice-ResNet employs transfer learning, enhancing ResNet50’s performance, and addressing data imbalance for precise quality evaluation.•Rice-ResNet, available on Python repositories, scales for purity detection, ensuring accurate classification across rice types.•Accessible via Python repositories, Rice-ResNet enables practical use in agriculture, and food tech, promising accurate classification.