Abstract

Avocados (Persea americana Mill.) are a climacteric fruit which ripen until after harvesting, and their ripeness is an important quality attribute that determines consumer liking. In this study, the ripening degree of ‘Hass’ avocados was evaluated non-destructively by measuring the skin and flesh using the fluorescence fingerprint (FF). FF, also known as the excitation-emission matrix (EEM), is a set of fluorescence spectra obtained at consecutive excitation wavelengths. It was found that as ripening progressed, the fluorescence signal of chlorophyll A in the skin and flesh decreased significantly as the hardness of the avocado flesh decreased. The hardness value was estimated from the FFs of the skin and flesh using partial least-squares regression, and minimum prediction errors of 2.02 N cm⁻² and 2.05 N cm⁻² were obtained for the prediction models using FFs of the flesh and skin, respectively. Furthermore, ripeness levels (unripe, ripe, and overripe) were discriminated non-destructively from the FFs of the skin with an accuracy of 90% for the validation dataset. The measurement and analysis technique demonstrated in this study is rapid and accurate, and can contribute to supplying uniform agricultural products to consumers.

中文翻译：

---

通过荧光指纹测量估算“ Hass”鳄梨（Persea americana Mill。）的成熟度

摘要

鳄梨（Persea americana Mill。）是一种更年期的水果，成熟到收获后才成熟，其成熟度是决定消费者喜好的重要品质属性。在这项研究中，通过使用荧光指纹（FF）测量皮肤和果肉，无损评估了“哈斯”鳄梨的成熟程度。FF，也称为激发发射矩阵（EEM），是在连续激发波长下获得的一组荧光光谱。发现随着成熟的进行，鳄梨果肉的硬度降低，皮肤和果肉中叶绿素A的荧光信号显着下降。使用部分最小二乘回归从皮肤和果肉的FF估计硬度值，最小预测误差为2.02 N cm ^-2对于预测模型，分别使用肉和皮肤的FF获得了2.05 N cm ^-2和2.05 N cm ^-2的预测值。此外，成熟度水平（未成熟，成熟和过度成熟）与皮肤的果皮无损地进行了区分，对于验证数据集，准确性为90％。这项研究中证明的测量和分析技术是快速而准确的，并且可以为消费者提供统一的农产品。