波长比和近红外光谱的番茄品质检测方法

番茄的可溶性固形物(SSC)、pH值和坚实度(Firmness)是决定番茄的口感及收获后品质的主要因素。提出一种基于波长比和近红外光谱的番茄SSC，pH和坚实度检测方法。首先用Vis/SWNIR便携式光谱仪 (波长：400～1 100 nm)和中波近红外便携式光谱仪 (波长：900～1 683 nm)在相互作用模式下，对6个不同成熟度的600个番茄样本进行光谱采集，对所采集的光谱用波长比法和波长比+自动缩放法进行预处理后，分别建立番茄SSC，pH和坚实度的预测模型，比较单一自动缩放、单一波长比、波长比+自动缩放及不做预处理四种方法的预测结果。结果显示，波长比法结合自动缩放预处理可有效提高可见/短波近红外光谱对SSC，pH和坚实度的预测精度(r_p=0.779，0.796和0.917)；波长比法+自动缩放法可提高中波近红外光谱对SSC的预测效果(r_p=0.818)。这说明波长比法在优化和处理番茄光谱信息方面具有一定的潜力。

Measurement of Tomato Quality Attributes Based on Wavelength Ratio and Near-Infrared Spectroscopy

The soluble solids content (SSC), pH and firmness (Firmness) of tomato are the key factors that determine the taste and post harvest quality of tomato. A new method for detecting tomato SSC, pH and firmness based on wavelength ratio and near infrared spectroscopy is proposed in this paper. Thespectra of six hundreds tomato samples with different maturity were collected with the portable Vis/NIR spectrometer (wavelength: 400～1 100 nm) and the portablenear infrared spectrometer (wavelength: 900～1 683 nm) in the interaction mode, respectively. After these spectra were pretreated with the wavelength ratio method are as follows：automatic scaling one and the wavelength ratio+automatic scaling one, the prediction models for SSC, pH and firmness of tomatowere developed, respectively, and then the prediction results of the four methods: are as follows automatic scaling, wavelength ratio, wavelength ratio + automatic scaling and no preprocessing were compared and analyzed. The experimental results showed that the prediction accuracy of the visible/near infrared spectra for SSC, pH and firmnesscouldbe visibly improvedby the wavelength ratio combined with the automatic scaling pretreatment, with r_p=0.779, 0.796 and 0.917, respectively. The wavelength ratio combined with the automatic scaling also could enhance the prediction ability of the Near infrared spectroscopy for SSC with r_p=0.818, which suggests that the proposed wavelength ratio method in this paper had considerable potential in optimizing and processing the spectral information of tomato.