Mo Changyeun, Kim Giyoung, Lim Jongguk, Kim Moon S, Cho Hyunjeong, Cho Byoung-Kwan
National Institute of Agricultural Science, Rural Development Administration, 310 Nonsaengmyeong-ro, Wansan-gu, Jeonju-si, Jeollabuk-do 54875, Korea.
Environmental Microbiology and Food Safety Laboratory, BARC-East, Agricultural Research Service, US Department of Agriculture, 10300 Baltimore Avenue, Beltsville, MD 20705, USA.
Sensors (Basel). 2015 Nov 20;15(11):29511-34. doi: 10.3390/s151129511.
Rapid visible/near-infrared (VNIR) hyperspectral imaging methods, employing both a single waveband algorithm and multi-spectral algorithms, were developed in order to discrimination between sound and discolored lettuce. Reflectance spectra for sound and discolored lettuce surfaces were extracted from hyperspectral reflectance images obtained in the 400-1000 nm wavelength range. The optimal wavebands for discriminating between discolored and sound lettuce surfaces were determined using one-way analysis of variance. Multi-spectral imaging algorithms developed using ratio and subtraction functions resulted in enhanced classification accuracy of above 99.9% for discolored and sound areas on both adaxial and abaxial lettuce surfaces. Ratio imaging (RI) and subtraction imaging (SI) algorithms at wavelengths of 552/701 nm and 557-701 nm, respectively, exhibited better classification performances compared to results obtained for all possible two-waveband combinations. These results suggest that hyperspectral reflectance imaging techniques can potentially be used to discriminate between discolored and sound fresh-cut lettuce.
为了区分完好生菜和变色生菜，开发了快速可见/近红外（VNIR）高光谱成像方法，该方法采用了单波段算法和多光谱算法。从在400-1000nm波长范围内获得的高光谱反射图像中提取完好生菜和变色生菜表面的反射光谱。使用单因素方差分析确定区分变色生菜和完好生菜表面的最佳波段。利用比率和减法函数开发的多光谱成像算法，使生菜叶片正反两面变色区域和完好区域的分类准确率提高到99.9%以上。与所有可能的双波段组合的结果相比，分别在552/701nm和557-701nm波长下的比率成像（RI）和减法成像（SI）算法表现出更好的分类性能。这些结果表明，高光谱反射成像技术有可能用于区分变色和完好的鲜切生菜。
