Study On Detection Methods And Technology Of Amylose Content

Grain amylose contents influence the cooking quality, appearance quality of grains and food safety. Research of amylose content measurement techniques and methods is significant to improve the market competition capacity of food at home. After Chinese entrance to the WTO, the research becomes more and more important. Standard colorimetric measurement of iodine complex formed with amylose and NIR (Near Infrared Spectroscopy)method are adopted to measure the grain amylose contents in this paper. An apparatus for measuring amylose content based on colorimetric measurement is developed. The contents of the study can be briefly summarized as follows:1. A new apparatus based on the theory of colorimetric measurement for measuring amylose content is developed. Laser lamp-house is adopted to decrease the conflicts of the homochromatism light purity and light intensity of the existing apparatuses based on colorimetric measurement. Because of the good homochromatism characteristic, the departure to Lambert-Beer law produced by non-homochromatism light is greatly reduced. In addition, the high intensity and the pretreatment of degrease and non-degrease can both insure the required measurement accuracy.2. The software system of this amylose measurement apparatus is developed. This system has the serial communication. It can monitor the measure process, display the measurement data, construct the standard curve and predict the unknown samples.3. Performance analysis of the apparatus for measuring amylose content is one of the main contents in this paper. A great deal of performance experiments prove that the apparatus for measuring amylose content has high accuracy and good stability. Experiment analysis shows that temperature has little effect on the measurement results of the apparatus. The comparative measurement experiments for the pretreatment of degrease are done with TU-1800 UV-VIS spectrophotometer and comparative measurement experiments for the pretreatment of non-degrease are done with the chemical automatic analyzer FS-IV of American O-I corporation. Results indicate that the pretreatments of non-degrease can also insure the required measurement accuracy. The performance of the apparatus we developed is equivalent to the alike kind apparatuses abroad.4. Wavelength selection method based on wavelet transform is put forward. This algorithm can not only reduce the effect of density matrix to the wavelength selection, but also filter wavelet detail coefficient entirely. Additionally, the high frequency noise is avoided, the run time of building model is reduced, the correlation coefficient R of PLS-LOO-CV prediction values and standard values is improved, and standard error of cross validation SECV and standard error SEP of prediction is lessened.5. Six wavelength selection methods are studied and compared: correlation coefficient method, deviation analysis, pertinency matrix analysis, uninformative variables elimination, genetic algorithm and wavelet transform. All the methods can extract the characteristic wavelength region of amylose partly. The accuracy and the prediction capacity of the model are improved. Among all these methods, the result of pertinency matrix analysis is best and the wavelength regions number of wavelet transformis least.6. Because the rice amylose content NIR model abroad is not suitable for rice at home, an amylose content NIR model for unpolished rice powder is constructed to measure the actual samples in China. The samples can not be used for breeding then because of the destructive measurement. So amylose content NIR model for millet and unpolished rice are constructed that can be used to the material primary selection for the breeding institution.7. Amylose contents of various grains such as wheat, maize, millet, barley, mung bean and broomcorn are measured by standard colorimetric measurement method. The measurement results compared to that of Academy of Agricultural Science show that our measurements are accurate and reliable. NIR (Near Infrared Spectroscopy) models of wheat and maize amy...