Research On Rice Composition Measurement And Origintraceability Technology Based On FTIR

In order to improve the enforceability of domestic geographical indication management laws,this thesis provides a research basis for subsequent traceability techniques.According to the geographical environment,planting techniques,and seed selection of rice growth areas,the Near-Infrared spectroscopy data of rice will have a theoretical basis for difference.By collecting Near-Infrared spectroscopy data of rice,combined with chemo metric methods,a traceability system for rice origin with rapid detection can be established.In the overall design,the hardware requirements are: easy to operate,can work in harsh environments,and repeatability.The requirements for the software are: to be able to leave the professional,the results are intuitive,the operating instructions are clear,and the computing power is strong.According to the requirements,the Fourier Near-Infrared spectrometer was selected as the main acquisition equipment on the hardware,and the diffuse reflection system was used as the acquisition attachment.The software used MATLAB to write the origin traceability algorithm and LabVIEW to write the operation interface.In the process of writing the operation interface,in order to fully satisfy the requirements of function separation,process display,etc.,the condition structure and event structure in LabVIEW are used extensively,and the operation of the entire origin traceability system can be performed through the interface through console commands..The origination algorithm used in the project is partial least squares and neural network classification.The partial least squares method is used to verify whether the spectral data is separable.The neural network method is used to obtain the best classification result.In addition,by modifying the sample pretreatment method and algorithm parameters,the influencing factors of the traceability accuracy of the origin can be verified experimentally.In the experiment of partial least squares method,the selection of spectral feature bands and the influence of different data processing methods on the traceability of origin are verified.In the experiment of neural network,the effects of different processing methods of samples and different parameters of the algorithm on the traceability of origin were verified.