开集分类识别是近10多年来模式识别领域研究的热点,它能够识别训练集中已知类别的测试样本,同时还能够有效“拒识”未知类别的测试样本;这些未知类别样本不包含在训练集中。现有的开集分类识别算法主要是基于Support Vector Machine(SVM...开集分类识别是近10多年来模式识别领域研究的热点,它能够识别训练集中已知类别的测试样本,同时还能够有效“拒识”未知类别的测试样本;这些未知类别样本不包含在训练集中。现有的开集分类识别算法主要是基于Support Vector Machine(SVM)和深度学习网络框架进行改进,并且主要应用在自然景物图像领域中;在光谱分析领域中还鲜有报道。将传统的闭集框架下的模糊推理分类器进行模型改进,提出了开集框架下的改进模糊推理分类器,并将其应用到木材树种近红外光谱分类识别中。首先,使用Flame-NIR近红外微型光谱仪采集木材样本横切面的近红外光谱曲线,采用Metric Learning算法进行光谱向量维度约简降维至4维(4D)。其次,改进闭集框架下的模糊推理分类器,根据模糊规则置信度和各维度隶属度概率的乘积构建Generalized Basic Probability Assignment(GBPA),再根据GBPA进行分类处理。在20个树种的具有不同的Openness指标下的近红外光谱数据集的分类识别对比实验表明,改进的开集模糊推理分类器(fuzzy reasoning classifier in an open set,FRCOS)优于现有的基于机器学习和深度学习的开集分类识别主流算法,具有较好的评价指标F-Score,Kappa系数及总体识别率。展开更多
Near-infrared (NIR) spectroscopy combined with chemometrics methods was applied to the rapid and reagent-free analysis of serum urea nitrogen (SUN). The mul-partitions modeling was performed to achieve parameter stabi...Near-infrared (NIR) spectroscopy combined with chemometrics methods was applied to the rapid and reagent-free analysis of serum urea nitrogen (SUN). The mul-partitions modeling was performed to achieve parameter stability. A large-scale parameter cyclic and global optimization platform for Norris derivative filter (NDF) of three parameters (the derivative order: d, the number of smoothing points: s and the number of differential gaps: g) was developed with PLS regression. Meantime, the parameters’ adaptive analysis of NDF algorithm was also given, and achieved a significantly better modeling effect than one without spectral pre-processing. After eliminating the interference wavebands of saturated absorption, the modeling performance was further improved. In validation, the root mean square error (SEP), correlation coefficient (RP) for prediction and the ratio of performance to deviation (RPD) were 1.66 mmol?L-1, 0.966 and 4.7, respectively. The results showed that the high-precision analysis of SUN was feasibility based on NIR spectroscopy and Norris-PLS. The global optimization method of NDF is also expected to be applied to other analysis objects.展开更多
花生球蛋白、伴花生球蛋白及亚基含量显著影响蛋白质的凝胶性和溶解性等功能特性,进而影响其在肉制品、植物蛋白饮料中的应用效果。目前常采用提取蛋白质后再用电泳及光密度法测定球蛋白、伴球蛋白及亚基含量的方法,操作步骤繁琐,样品...花生球蛋白、伴花生球蛋白及亚基含量显著影响蛋白质的凝胶性和溶解性等功能特性,进而影响其在肉制品、植物蛋白饮料中的应用效果。目前常采用提取蛋白质后再用电泳及光密度法测定球蛋白、伴球蛋白及亚基含量的方法,操作步骤繁琐,样品损失量大。为此收集了178个花生品种,分别提取蛋白,采用电泳法测定球蛋白、伴球蛋白、23.5和37.5 kDa亚基含量并获得大量数据的基础上,利用近红外光谱技术进行整粒花生样品的光谱扫描,将其与传统方法测定的化学值进行拟合,采用偏最小二乘回归(PLSR)化学计量法构建数学模型。通过比较单一和复合光谱预处理方式,对比模型相关系数和误差评估预测模型性能。确定球蛋白模型最佳预处理方法为2^(nd)-der with Detrend,校正集相关系数为0.92,标准差为1.41;伴球蛋白模型最佳预处理方法为Detrend with 1^(st)-der,校正集相关系数为0.85,标准差为1.46;23.5 kDa亚基含量模型最佳预处理方法为Normalization with 2^(nd)-der,校正集相关系数为0.91,标准差为0.53;37.5 kDa模型最佳预处理方法为Detrend with Baseline,校正集相关系数为0.91,标准差为0.89。外部验证结果表明,球蛋白预测均方根误差(square errors of predi ction,SEP)为1.25,伴球蛋白SEP为0.73,23.5 kDa模型SEP为0.47,37.5 kDa模型SEP为0.75。本研究基于近红外光谱技术实现了对整粒花生进行球蛋白、伴球蛋白、23.5 kDa和37.5 kDa亚基含量的同步、快速和无损检测,为育种专家加工专用品种选育和蛋白加工企业原料选用提供了根据。展开更多
文摘开集分类识别是近10多年来模式识别领域研究的热点,它能够识别训练集中已知类别的测试样本,同时还能够有效“拒识”未知类别的测试样本;这些未知类别样本不包含在训练集中。现有的开集分类识别算法主要是基于Support Vector Machine(SVM)和深度学习网络框架进行改进,并且主要应用在自然景物图像领域中;在光谱分析领域中还鲜有报道。将传统的闭集框架下的模糊推理分类器进行模型改进,提出了开集框架下的改进模糊推理分类器,并将其应用到木材树种近红外光谱分类识别中。首先,使用Flame-NIR近红外微型光谱仪采集木材样本横切面的近红外光谱曲线,采用Metric Learning算法进行光谱向量维度约简降维至4维(4D)。其次,改进闭集框架下的模糊推理分类器,根据模糊规则置信度和各维度隶属度概率的乘积构建Generalized Basic Probability Assignment(GBPA),再根据GBPA进行分类处理。在20个树种的具有不同的Openness指标下的近红外光谱数据集的分类识别对比实验表明,改进的开集模糊推理分类器(fuzzy reasoning classifier in an open set,FRCOS)优于现有的基于机器学习和深度学习的开集分类识别主流算法,具有较好的评价指标F-Score,Kappa系数及总体识别率。
文摘Near-infrared (NIR) spectroscopy combined with chemometrics methods was applied to the rapid and reagent-free analysis of serum urea nitrogen (SUN). The mul-partitions modeling was performed to achieve parameter stability. A large-scale parameter cyclic and global optimization platform for Norris derivative filter (NDF) of three parameters (the derivative order: d, the number of smoothing points: s and the number of differential gaps: g) was developed with PLS regression. Meantime, the parameters’ adaptive analysis of NDF algorithm was also given, and achieved a significantly better modeling effect than one without spectral pre-processing. After eliminating the interference wavebands of saturated absorption, the modeling performance was further improved. In validation, the root mean square error (SEP), correlation coefficient (RP) for prediction and the ratio of performance to deviation (RPD) were 1.66 mmol?L-1, 0.966 and 4.7, respectively. The results showed that the high-precision analysis of SUN was feasibility based on NIR spectroscopy and Norris-PLS. The global optimization method of NDF is also expected to be applied to other analysis objects.
文摘花生球蛋白、伴花生球蛋白及亚基含量显著影响蛋白质的凝胶性和溶解性等功能特性,进而影响其在肉制品、植物蛋白饮料中的应用效果。目前常采用提取蛋白质后再用电泳及光密度法测定球蛋白、伴球蛋白及亚基含量的方法,操作步骤繁琐,样品损失量大。为此收集了178个花生品种,分别提取蛋白,采用电泳法测定球蛋白、伴球蛋白、23.5和37.5 kDa亚基含量并获得大量数据的基础上,利用近红外光谱技术进行整粒花生样品的光谱扫描,将其与传统方法测定的化学值进行拟合,采用偏最小二乘回归(PLSR)化学计量法构建数学模型。通过比较单一和复合光谱预处理方式,对比模型相关系数和误差评估预测模型性能。确定球蛋白模型最佳预处理方法为2^(nd)-der with Detrend,校正集相关系数为0.92,标准差为1.41;伴球蛋白模型最佳预处理方法为Detrend with 1^(st)-der,校正集相关系数为0.85,标准差为1.46;23.5 kDa亚基含量模型最佳预处理方法为Normalization with 2^(nd)-der,校正集相关系数为0.91,标准差为0.53;37.5 kDa模型最佳预处理方法为Detrend with Baseline,校正集相关系数为0.91,标准差为0.89。外部验证结果表明,球蛋白预测均方根误差(square errors of predi ction,SEP)为1.25,伴球蛋白SEP为0.73,23.5 kDa模型SEP为0.47,37.5 kDa模型SEP为0.75。本研究基于近红外光谱技术实现了对整粒花生进行球蛋白、伴球蛋白、23.5 kDa和37.5 kDa亚基含量的同步、快速和无损检测,为育种专家加工专用品种选育和蛋白加工企业原料选用提供了根据。
