Font Size: a A A

Study On Confirmation Of Meihe Rice Producing Area By Hyperspectral Imaging Technology

Posted on:2020-07-10Degree:MasterType:Thesis
Country:ChinaCandidate:J Z YangFull Text:PDF
GTID:2370330599962809Subject:Food, grease and vegetable protein engineering
Abstract/Summary:PDF Full Text Request
As one of the well-known rice “Geographical Marking Products”,Meihe Rice has been delivering high-quality rice to people all over the country for many years.However,some lawless elements mixed the inferior or non-marked source of rice into the Meihe rice,and obtained a high amount of illegal benefits with sub-filling,and had a very bad influence on the reputation and reputation of Meihe rice products.It has exposed the lack of research on the confirmation of the relevant origin of the domestic rice industry.At present,most of the domestic research on the confirmation of rice production uses the traditional physical and chemical indicators detection method.Although it is feasible,it has many defects such as strong destructiveness and long experimental period.In foreign countries,the concept of “fast non-destructive testing” has been introduced in a targeted manner,using a new generation of hyperspectral imaging technology,and making the technology shine in food production confirmation and food testing.Therefore,in view of the advantages of rapid non-destructive testing of hyperspectral imaging technology,combined with the taste index of Meihe rice samples and its growth environment factors,this paper conducts a stable and reliable study on the confirmation of excavation factors and confirmation of origin of Meihe rice.In order to overcome the other factors,the rice samples from the purchase of commercially available rice,the source instability,further confirm the correlation between the Meihe rice sample and its surrounding growth environment factors,and maximize the accurate analysis of the regional characteristics of the Meihe rice space,this paper uses space The grid point method was used to collect 300 samples of Meihe rice and its corresponding root soil samples and environmental factors.The variance analysis and correlation analysis of the sample taste quality index and soil and environmental information were carried out.The results showed that the total protein,amylose,palmitic acid,oleic acid,linoleic acid and fatty acids in the rice quality index were the five mains in Wanlong Town,Hailong Town,Heishantou Town,Shancheng Town and Jile Township.There are significant differences between the production areas.The six taste quality indicators are significantly correlated with the longitude,precipitation,pH value and organic matter content of the environmental factors,and the six taste quality indicators are in Meihe rice and others.The rice samples collected in the approximate area have obvious characteristics of origin,and can be combined with hyperspectral imaging technology as an important reference index for screening the confirmatory factors of the stable expression characteristics of Meihe rice.After the hyperspectral imaging test of the Meihe rice sample,and the steps of region selection,spectral noise elimination,spectral data optimization processing,dimensionality reduction,etc.,the remaining 14 wavelengths of the system replace the complicated and complex full-band spectral information,but the above wavelengths It may also reflect other indicators other than the taste quality indicators,so it is further screened in combination with the taste quality indicators.Finally,the hyperspectral nine characteristic wavelengths TB1~TB9(455 nm,515 nm,583 nm,615 nm,621 nm,776 nm,874 nm,923 nm,942 nm)of Meihe rice are obtained and compared with the above six food quality indicators.Shows significant correlation.In order to verify the ability to confirm the origin,the SPSS discriminant analysis of Meihe rice and Liuhe county rice was carried out by using nine characteristic wavelengths of hyperspectral.The overall correct rate of the training set was 95%,and the correct rate of cross validation was 93%.The overall correct rate of the verification set is 94%,and the correct rate of cross validation is 91%.The model established by SIMCA partial least squares method can distinguish the Meihe rice samples from the Liuhe County rice samples.It is indicated that the nine characteristic wavelengths of Meihe rice hyperspectral can be used as the confirmation index of its origin.This experiment provides spatial data and characteristic band data for the Meihe rice production confirmation database,and also provides a data foundation and a good foundation for the establishment of the Meihe rice production confirmation system.In the future,if the corresponding rice confirmation system in the region is established,it will bring a benefit advantage to the growers to some extent.
Keywords/Search Tags:provenance of origin, hyperspectral imaging technique, taste quality index, correlation analysis, discriminant analysis
PDF Full Text Request
Related items