高光谱图像技术在水果品质无损检测中的应用  

Review of hyperspectral image technology for non-destructive inspection of fruit quality

作　　者：洪添胜[1] 李震[1,3] 吴春胤[1] 刘敏娟[1] 乔军[2] Wang Ning 

机构地区：[1]华南农业大学工程学院“南方农业机械与装备关键技术”省部共建教育部重点实验室,广州510642 [2]中国农业大学网络中心,北京100083 [3]Department of Biosystems and Agricultural Engineering, Oklahoma State University, 111 Ag Hall, Stillwater, OK, 74078

出　　处：《农业工程学报》2007年第11期280-285,共6页Transactions of the Chinese Society of Agricultural Engineering

基　　金：广东省科技计划(国际合作项目)(2006B50106002)资助

摘　　要：传统的近红外光谱分析法和可见光图像技术应用于水果品质无损检测中存在的检测区域小、检测时间长、仅能检测表面情况等局限性。高光谱图像技术结合光谱技术与计算机图像技术两者的优点,可获得大量包含连续波长光谱信息的图像块,其图像信息可检测水果的外部品质,光谱信息则可用于水果内部品质的检测,达到根据水果内、外部综合品质进行分类的目的。根据不同的采集设备,简介了两种获得高光谱图像的方法。综述了国内外将该技术应用于水果品质检测方面的研究进展,检测内容包括外观品质、损伤与缺陷,成熟度和坚实度,含糖量、含水率等内部品质,着重介绍了各高光谱图像的成像波段范围、分辨率、成像源,实验数据处理的方法以及实验结果等。根据综述所得提出了高光谱图像技术应用中需要解决的光谱降维、降低样品差异影响和实时检测平台搭建等问题。Small detecting zone, long detecting period and limitation to external inspection are included in the deficiencies of singly using computer vision or spectroscopy for non-destructive inspection of fruit quality. Image cubes.containing continuous spectral waveband information, in which the image information could be used for external attribute inspection while the spectral information could be applied to the internal attribute inspection, could be obtained from implementing a hyperspectral image technology which combines the advantages of computer vision and spectroscopy. As a result, fruit classification based on both internal and external quality attributes could be achieved. Two different methods for acquiring hyperspectral images and the corresponding hardware of hyperspectral imaging system were introduced in this paper. Applications of byperspectral images to the inspection of bruises, feces or earth contamination, maturity, firmness, SSC（Soluble Solid Content） and other parameters of fruits were reviewed. It mainly focused on the wave band, resolution, image source, data analysis methods and the experimental results. The problems needed to be solved in applying this technique, such as spectral dimensionality reduction, real-time platform building and sample diversity impact, were put forward.

关 键 词：高光谱图像技术 水果品质 无损检测 机器视觉 水果分级 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN911.73[自动化与计算机技术—计算机科学与技术]