应用纹理分析方法在线监测苹果冻干含水率

在果蔬真空冷冻干燥过程中对果蔬物料含水率进行实时监测,可为冻干过程监控和优化提供依据。该文研究以冻干果蔬物料含水分图像纹理分析为技术手段,实现连续表达干燥过程中物料的含水率。试材选用苹果,采用环形光源和高速CCD组件,在冻干仓窗外分别采集原始苹果样本MA和经CuSO4溶液染色的苹果样本MB在一个完整冻干周期中表面含水纹理变化的动态图像,运用主成分分析法对图像均值等6个含水纹理特征指标进行统计分析,并对原始样本纹理特征的第一主分量与其含水率W1、染色样本含水纹理特征的第一主分量与其含水率W2、W1和W2分别进行非线性回归分析。结果表明,模型的决定系数达到0.9376、0.9289和0.9930,且显著性检验概率均<0.0001,模型检验极显著。同时,模型相对误差基本<3%。因此,由果蔬物料含水图像的纹理特征可实现含水率的在线监测。该方法不但为真空冷冻干燥加工过程控制探索一种利用水分图像处理方法进行水分在线监测的新方法,而且也可在其他干燥加工水分监测及过程控制中应用。

On-line monitoring of moisture ratio for apple during vacuum freeze-drying based on image texture analysis

Based on image texture analysis, this paper provides a new experimental method on real-time monitoring of moisture content. The images of original apple slice sample and sample dyed by CuSO4 solution were captured by high-speed CCD and annular light source with polariscope during vacuum freeze-drying process. Mean value, standard deviation, smoothness, the third moment, uniformity and entropy were analyzed statistically by principal component analysis method, and the relationship between PC1 and moisture ratio W1 of original apple slice, PC1 and moisture ratio W2 of dyed apple slice were analyzed using non-linear regression method. In addition, the relationship between W1 and moisture ratio W2 of dyed apple slice was carried out the same computation with non-linear regression method. The results indicate that the p values of models are both <0.0001, whose coefficient of determination are 0.9376, 0.9289 and 0.9930, respectively. In addition, most relative errors to the moisture ratio are limited 3%. Therefore, the mentioned models not only provide a new monitoring method of moisture ratio, but also give foundation of monitoring moisture ratio for other drying processes.