摘要
基于近红外高光谱成像技术对马铃薯干物质含量进行无损检测研究。运用偏最小二乘回归系数法对多元散射校正(MSC)预处理后的光谱优选出8个特征波长,采用粒子群算法优化支持向量机(PSO-SVM)和偏最小二乘回归算法(PLSR)分别建立特征波长预测模型并对比分析。结果表明,采用粒子群算法优化支持向量机(PSO-SVM)建立的特征波长预测模型优于偏最小二乘回归算法(PLSR)预测模型,校正和验证模型的相关系数和均方根误差分别为0.944 37、0.919 77和0.155 01、0.156 90,高光谱成像技术对马铃薯干物质含量无损检测是可行的。
Dry matter content of potatoes was determined using nearinfrared hyperspectral imaging technique. The original spectra were pretreated by multiplicative scatter correction, 8 optimal wavelengths were selected by regression coefficients of partial least-squares mod- els in the spectral region between 990 nm and 1 630 nm. Prediction models were built using Particle Swarm Optimization Algorithm optimizing Support Vector Machine method （PSO-SVM） and Partial Least Squares Regression method （PLSR） based on the optimal wavelengths. The results showed that prediction models based on PSO-SVM method in the optimal wavelengths are better than PLSR method for predicting the dry matter content in potatoes, its correla tion coefficient and root mean square error of calibration and validation models are 0. 944 37, 0.919 77 and 0. 155 01, 0. 156 90, respectively. Therefore, I＇s feasible to determinate the dry matter content in potatoes using hyperspectral imaging technique.
出处
《食品与机械》
CSCD
北大核心
2014年第4期133-136,150,共5页
Food and Machinery
基金
宁夏自然科学基金资助项目(编号:NZ13005)
关键词
高光谱成像技术
马铃薯
干物质
无损检测
hyperspectral imaging technique
potato
dry matter
non-destructive detection
