Authentication of food quality is a global concern. In recent years, multiple-level information fusion technologies have been widely studied and applied to make further improvement for food pattern recognition and property parameters estimation. Information from different sources can help the decision system better understand the internal and external states of food samples. This review surveys the applications of information fusion in food quality authentication, especially concerning that for meat, fruits, vegetables, medicinal foods, mushrooms, beverages, and edible vegetal oils. We introduce briefly the sensing technologies for the information acquisition and further summarize in detail of the research objective, observation tools, data processing algorithms, fusion strategies as well as the comparison between fusion and non-fusion methods. This survey provides clear evidence of the superiority of fusion-based methods for food authentications. Furthermore, we bring some trends of information fusion approaches into attention and discuss the challenges and future perspectives with recommendations.

食品质量认证是全球关注的问题。近年来，多层次信息融合技术得到了广泛的研究和应用，以进一步改善食品模式识别和属性参数估计。来自不同来源的信息可以帮助决策系统更好地了解食品样品的内部和外部状态。这篇综述调查了信息融合在食品质量认证中的应用，特别是在肉类，水果，蔬菜，药用食品，蘑菇，饮料和食用植物油方面。我们简要介绍了用于信息获取的传感技术，并进一步详细概述了研究目的，观察工具，数据处理算法，融合策略以及融合与非融合方法之间的比较。这项调查提供了基于融合的食品认证方法优越性的明确证据。此外，我们将引起人们对信息融合方法的一些趋势的关注，并讨论有关挑战和未来观点的建议。