Milk Composition Detection Methods Based on Infrared Spectral Features and Fourier Transforms
Currently proposed milk composition detection methods have poor detection ability for protein,fat and lactose in milk,which leads to high relative errors in detection.In order to solve the above problems,a new milk composition detection method based on infrared spectral characterization and Fourier transform was investigated.Through the liquid milk composition characterization and various morphological changes,infrared spectral tracking and monitoring,according to the results of milk infrared spectral analysis to achieve data preprocessing,identification and processing of infrared spectral bands characteristic wavelengths,considering the concentration of the milk components and the estimated distribution probability of the variables,the realization of parameter estimation,according to the method of least squares to calculate the estimated value of the parameters of the components,can be calculated according to the predicted reference value of the milk composition,according to the characteristics of the ingredients According to the difference of spectral absorption theory,the observation sample type is divided,and according to the milk quality grade,the different components content concentration is classified,and the feature extraction is completed.The experimental results show that the relative error of the milk composition detection method based on infrared spectral features and Fourier transform is less than 0.2%,and the detection accuracy is high.
万方数据
维普
