基于光学成像的不同产地宁夏枸杞外观识别研究

枸杞产地的快速、准确鉴别，对规范枸杞交易市场、推动不同产地枸杞差异化、品牌化战略发展具有重要意义。本文以2018-2019年宁夏、新疆两地四产区的宁杞7号夏果干果为研究对象，利用可见/近红外高光谱成像系统，对图像进行Hue Saturation Value(HSV)色彩空间变换和纹理特征提取。配合人工测定的百粒重、果形指数（L/D）等枸杞果形参数指标，采用最小显著差异法（LSD）比较不同产地的差异性。在R 3.6.2环境支持下，对数据进行决策树(Decision Tree,DT)、随机森林(Random Forest,RF)、支持向量机(Support Vector Machine,SVM)、多元逻辑回归(Multinomial Logistic Regression, MLR)等分类器模型训练并建立产地识别模型，开展基于“高光谱成像+计算机视觉”的枸杞产地识别技术研究。结果表明：枸杞百粒重新疆明显高于宁夏，果形指数宁夏高于新疆，宁夏枸杞纹理更深更复杂但色泽较暗；4种枸杞产地识别模型中DT模型表现最稳定，且果形指数参与建模后的识别精度更高。

Study on Appearance Recognition of Lycium barbarum from Different Geographical Origins in Ningxia Based on Hyperspectral Imaging

The rapid and accurate identification of the origin of Lycium barbarum is of great significance for regulating the trading market of Lycium barbarum and promoting the differentiation and branding strategy development of Lycium barbarum in different origins. In this paper, the dried fruit of Ningqi No.7 from four producing areas in Ningxia and Xinjiang from 2018 to 2019 was used as the research object. The research object was scanned and imaged using the HyperSpec VNIR N-type visible-near-infrared hyperspectral imaging system. Perform the HSV color space transformation and texture feature extraction on scanned images. The least significant difference method was used to compare the differences between different origins in combination with the manually measured 100-grain weight, fruit shape index and other Lycium barbarum fruit shape parameters. The data are trained with classifier models such as Decision Tree, Random Forest, Support Vector Machine, and Multinomials Logistic Regression, an origin identification model is established, and the research on the origin identification technology of Lycium barbarum. The results showed that the 100-grain weight of Lycium barbarum in Xinjiang was significantly higher than that in Ningxia, the fruit shape index in Ningxia was better than that in Xinjiang, and the texture of Lycium barbarum in Ningxia was deeper and more complex, but the color was darker. Among the four types of Lycium barbarum origin identification models, the Decision Tree model has the most stable performance, and the identification accuracy was higher after the fruit shape index was involved in the modeling.