目的探究白酒感官品评与白酒风味成分之间的关系,实现通过感官品评对风味成分进行预测。方法采用变分自编码器(variational auto encoder,VAE)对原始数据进行增强,以多核支持向量回归(multi-kernel support vector regression,MKSVR)结...目的探究白酒感官品评与白酒风味成分之间的关系,实现通过感官品评对风味成分进行预测。方法采用变分自编码器(variational auto encoder,VAE)对原始数据进行增强,以多核支持向量回归(multi-kernel support vector regression,MKSVR)结合遗传算法(genetic algorithm,GA)建立单预测模型,再采取逐步预测的方式按照酸、酯、醇、醛类物质的顺序进行预测,从而构建最终模型。结果在经过VAE对数据进行增强的条件下,多元线性回归(mixed logistic regression,MLR)对酸、酯、醇、醛类物质预测的拟合优度分别为0.9660、0.9106、0.8767、0.8686,随机森林(random forests,RF)对酸、酯、醇、醛类物质预测的拟合优度分别为0.9663、0.9186、0.8805、0.8708,GA-MKSVR对酸、酯、醇、醛类物质预测的拟合优度分别为0.9715、0.9423、0.9072、0.8809,GA-MKSVR逐步预测对酸、酯、醇、醛类物质预测的拟合优度分别为0.9715、0.9447、0.9102、0.8851,GA-MKSVR逐步预测的效果均为最优。结论GA-MKSVR逐步预测方法相较于传统的机器学习方法,具有更好的性能,对数据具有更高的适应性,能更好地构建白酒感官与风味成分之间的关系模型。展开更多
To match human perception, extracting perceptual features effectively plays an important role in image quality assessment. In contrast to most existing methods that use linear transformations or models to represent im...To match human perception, extracting perceptual features effectively plays an important role in image quality assessment. In contrast to most existing methods that use linear transformations or models to represent images, we employ a complex mathematical expression of high dimensionality to reveal the statistical characteristics of the images. Furthermore, by introducing kernel methods to transform the linear problem into a nonlinear one, a full-reference image quality assessment method is proposed based on high-dimensional nonlinear feature extraction. Experiments on the LIVE, TID2008, and CSIQ databases demonstrate that nonlinear features offer competitive performance for image inherent quality representation and the proposed method achieves a promising performance that is consistent with human subjective evaluation.展开更多
文摘目的探究白酒感官品评与白酒风味成分之间的关系,实现通过感官品评对风味成分进行预测。方法采用变分自编码器(variational auto encoder,VAE)对原始数据进行增强,以多核支持向量回归(multi-kernel support vector regression,MKSVR)结合遗传算法(genetic algorithm,GA)建立单预测模型,再采取逐步预测的方式按照酸、酯、醇、醛类物质的顺序进行预测,从而构建最终模型。结果在经过VAE对数据进行增强的条件下,多元线性回归(mixed logistic regression,MLR)对酸、酯、醇、醛类物质预测的拟合优度分别为0.9660、0.9106、0.8767、0.8686,随机森林(random forests,RF)对酸、酯、醇、醛类物质预测的拟合优度分别为0.9663、0.9186、0.8805、0.8708,GA-MKSVR对酸、酯、醇、醛类物质预测的拟合优度分别为0.9715、0.9423、0.9072、0.8809,GA-MKSVR逐步预测对酸、酯、醇、醛类物质预测的拟合优度分别为0.9715、0.9447、0.9102、0.8851,GA-MKSVR逐步预测的效果均为最优。结论GA-MKSVR逐步预测方法相较于传统的机器学习方法,具有更好的性能,对数据具有更高的适应性,能更好地构建白酒感官与风味成分之间的关系模型。
基金Project supported by the National High-Tech R＆D Program （863） of China （No. 2015AA016704c）, the National Science Technology Support Program of China （No. 2013BAH03B01）, and the Zhejiang Provincial Natural Science Foundation of China （No. LY14F020028）
文摘To match human perception, extracting perceptual features effectively plays an important role in image quality assessment. In contrast to most existing methods that use linear transformations or models to represent images, we employ a complex mathematical expression of high dimensionality to reveal the statistical characteristics of the images. Furthermore, by introducing kernel methods to transform the linear problem into a nonlinear one, a full-reference image quality assessment method is proposed based on high-dimensional nonlinear feature extraction. Experiments on the LIVE, TID2008, and CSIQ databases demonstrate that nonlinear features offer competitive performance for image inherent quality representation and the proposed method achieves a promising performance that is consistent with human subjective evaluation.
