Abstract The objective of this research was to measure the changes of optical properties and quality or microstructural properties of peaches during fungal infection, and classify the fungal infected peaches based on the optical parameters. Spectra of the absorption (μa) and reduced scattering coefficients (μs') over 600-1000 nm for healthy and fungal infected peaches over a period of four days were measured by using a spatially-resolved spectroscopic technique. The color and microstructural features of fruit pulp and peel were measured and evaluated, using colorimetry and scanning electron microscopy (SEM), as indicators of the changes in tissue appearance and internal quality in infected peaches. The μa and μs' spectra exhibited a pattern of decrease during the fungal infection, and their values at wavelengths of 670 nm and 970 nm were correlated with the microstructural parameters of fruit peel and pulp (i.e., mycelial area, intrusion rate, and the energy, entropy and contrast extracted from the SEM images). Significant differences in the quality parameters between healthy and infected peaches were found after 3 d of inoculation for the peel tissues and after 2 d for pulp tissues. Significant differences between the healthy and infected peaches after 1 d of inoculation were also observed for both μa and μs'. The optical parameters were more sensitive to disease infection than some of the quality parameters. Partial least squares discriminant analysis (PLSDA) models were developed, based on the two optical parameters and their combinations, for classifying diseased and healthy peaches. The PLSDA model for the optical parameter of μa × μs' achieved better overall classification accuracies of 70–88 %, when the peaches were classified into four (based on infection days) and two (i.e., healthy and diseased) classes, respectively. This research demonstrated that optical properties can be used to assess quality or structural changes and detecting disease infection in peach fruit.

中文翻译：

---

基于光学和微观结构特性的桃子真菌感染评估

摘要 本研究的目的是测量受真菌感染期间桃子的光学特性和品质或微观结构特性的变化，并根据光学参数对受真菌感染的桃子进行分类。通过使用空间分辨光谱技术测量了健康和受真菌感染的桃子在四天内的 600-1000 nm 范围内的吸收光谱 (μa) 和减少的散射系数 (μs')。使用比色法和扫描电子显微镜 (SEM) 测量和评估果肉和果皮的颜色和微观结构特征，作为感染桃子组织外观和内部质量变化的指标。μa 和 μs' 光谱在真菌感染期间表现出降低的模式，它们在 670 nm 和 970 nm 波长处的值与果皮和果肉的微观结构参数（即菌丝面积、侵入率以及从 SEM 图像中提取的能量、熵和对比度）相关。果皮组织接种 3 天后和果肉组织接种 2 天后，发现健康桃和感染桃的质量参数存在显着差异。对于 μa 和 μs'，还观察到接种 1 天后健康桃和受感染桃之间的显着差异。与某些质量参数相比，光学参数对疾病感染更敏感。基于两个光学参数及其组合开发了偏最小二乘判别分析 (PLSDA) 模型，用于对患病桃子和健康桃子进行分类。当将桃子分别分为四个（基于感染天数）和两个（即健康和患病）类别时，μa × μs' 光学参数的 PLSDA 模型实现了 70-88% 的更好的总体分类精度。这项研究表明，光学特性可用于评估桃果实的质量或结构变化以及检测疾病感染。