基于空间频域成像技术的水果组织光学特性参数测量及光传输机理解析

It is difficult to accurately detect the structure-associated fruit quality indices by near-infrared spectroscopy. The key scientific issues for deep expansion of agricultural biotechnology research for fruits and other agro-products optical properties was raised. It is important to get absorption and scattering parameters of tissue and to analyze the interaction mechanism between light and tissues. Based on the results of preliminary research, this project intends to study the optical properties of apples and other staple fruits. For the advantages of spatial frequency domain imaging technique in biomedical photonics researches, this project intends to build a spatial frequency domain imaging system that is suitable for fruits. The system will be optimized for the characteristics of detection accuracy, detection speed, and maneuverability. To obtain the absorption and scattering properties of different fruit tissues, the acquired images need to be processed by image demodulation, information extraction and parameters reconstruction. Based on absorption and scattering parameters, the light transmission process in the fruit tissues can be simulated by Monte Carlo method. The simulation models can be used for analyzing the transmission rule and mechanism. Absorption and scattering parameters will also be used to correct the near infrared spectra. Models based raw spectra and corrected spectra will be compared to assess the effect of optical properties correction for fruit quality prediction models. The results of this research project are expected to provide a new method for fruit quality detection and provide a theoretical basis for the development of technical reserves fruit quality detection technology and equipment. It has significant scientific and practical values.

针对目前广泛应用的近红外光谱分析技术难以准确检测与水果组织结构相关的品质指标这一实际，提出了水果等农业生物物料光学特性研究领域向纵深拓展的关键科学问题——获取组织的吸收和散射特性参数，阐明光在组织中的传输机理。本项目拟以苹果等大宗水果为研究对象，结合前期预研结果，利用空间频域成像技术在生物组织光学研究领域的优势，构建适用于水果的空间频域成像检测系统，对检测精度、检测速度、可操作性等进行优化，经图像解调、信息提取和参数重构等分析过程，获取不同水果组织的吸收和散射特性参数，进而对光在水果组织中的传输过程进行蒙特卡罗仿真并解析其传输规律，评估光学特性参数对水果品质光谱检测模型的修正效果，最终为推动水果光学特性研究的纵深拓展提供理论基础。本项目研究成果有望为水果品质检测提供新方法，为我国水果品质检测技术与装备的研发提供技术储备，具有重要的学术价值和现实意义。

空间频域成像（spatial frequency domain imaging, SFDI）可同时获取组织深度信息和反映光在组织中吸收和散射的量化光学特性参数，是近年来生物组织光学领域的研究热点，在获取生物组织光学特性、阐明光与组织相互作用机理以及基于此的实际检测应用方面具有广阔前景，在推动生物物料光学特性研究向纵深拓展方面显现出极大潜力。本项目将空间频域成像技术应用于水果组织光学特性参数的测量，探索光与水果组织相互作用，研究基于光学特性参数的水果品质分析方法，主要研究内容和成果有：1）构建了一套适用于水果组织光学特性检测的空间频域成像系统，开发了基于LabVIEW的系统操作软件，完成了系统的改进优化和精度验证；2）设计了适应不同检测对象的固体仿体制作模具并制作了平板形和半球形固体仿体，实现了基于相位测量轮廓术的样本高度校正，提高了非平板样本光学特性参数的空间频域成像检测精度，约化散射系数检测相对误差<4.94%、吸收系数检测相对误差<8.74%；3）获取了水果组织的吸收和散射特性参数，实现了基于光学特性参数的水果品质预测分析，以梨的损伤判别为例，正常梨的判别正确的率为98.33%、损伤梨的判别正确率为100%，不同损伤程度的判别正确率>82.5%；4）探索了光与组织的相互作用，通过有限元模拟分析了结构光在组织中的传输情况，通过蒙特卡罗模拟并优化了光学特性参数的反演，优化的空间频率提高了吸收系数和约化散射系数的平均反演精度；5）申请了国家发明专利7件（其中已授权3件），发表了论文11篇（其中SCI收录8篇、EI收录会议论文3篇）。本项目研究为水果品质检测提供了新方法，推动了农业生物物料光学特性研究向纵深拓展，为促进我国水果品质检测技术与装备的研发提供了技术储备。