Unified Comprehensive Freshness Classification Dataset (UC-FCD) for Diverse Food Categories
收藏资源简介：
The proper evaluation of food freshness is critical to ensure safety, quality along with customer satisfaction in the food industry. While numerous datasets exists for individual food items,a unified and comprehensive dataset which encompass diversified food categories remained as a significant gap in research. This research presented UC-FCD, a novel dataset designed to address this gap. The dataset comprised of meticulously curated images of multiple food categories, which included fish (Labeo rohita, Mola mola, Pampus argenteus, Dendrobranchiata), grain (Oryza sativa L.), meat product (Gallus gallus domesticus liver), dairy product (Withania coagulans), baked goods (bread, cake, samosa, laddu, dhokla), pickles (Berry pickle, Mango pickle), and egg. Each food category is annotated and divided into two classes representing its freshness status. The validation of the dataset for utility was done using state-of-the-art deep learning models for freshness classification. The dataset enabled comprehensive experiments on cross-category generalization, transfer learning and multi-modal classification approaches, which provided a robust foundation for researchers and industry practitioners. The results underscored the potential of advanced neural networks to achieve high accuracy in freshness classification which challenges posed by inter-category variability. The UC-FCD dataset is publicly available and aimed for further advancements in food quality assessment, ultimately paving the way for more intelligent and automated food safety solutions.
在食品工业领域，对食品新鲜度开展精准评估，对于保障食品安全、食品品质与消费者满意度至关重要。尽管当前已有诸多针对单一食品品类的数据集，但能够涵盖多样化食品类别的统一化综合数据集，仍是研究领域的一大显著空白。本研究提出UC-FCD这一新型数据集，旨在填补这一研究空白。 该数据集包含经过精心筛选的多品类食品图像，涵盖鱼类（露斯塔野鲮(Labeo rohita)、翻车鲀(Mola mola)、银鲳(Pampus argenteus)、枝鳃亚目(Dendrobranchiata)）、谷物（水稻(Oryza sativa L.)）、肉制品（家鸡肝脏(Gallus gallus domesticus liver)）、乳制品（凝乳睡茄(Withania coagulans)）、烘焙食品（面包、蛋糕、萨摩萨、拉杜甜饼、多克拉蒸糕）、腌菜（浆果腌菜、芒果腌菜）以及蛋类。每一类食品均已完成标注，并被划分为代表其新鲜度状态的两个类别。 本研究采用当前最先进的深度学习模型开展新鲜度分类任务，以此验证该数据集的实用价值。该数据集支持开展跨品类泛化、迁移学习以及多模态分类方法等全面实验，为研究人员与行业从业者提供了坚实的研究基础。 实验结果凸显了先进神经网络在应对品类间差异带来的挑战时，实现新鲜度分类高准确率的潜力。UC-FCD数据集已公开上线，旨在推动食品品质评估领域的进一步发展，最终为构建更智能、自动化的食品安全解决方案铺平道路。
