Study On The Inversion Model Of Cultivated Land Quality Based On NDVI

As an important carrier of grain production,cultivated land is the material basis for human survival and development.Studying the quality and spatial distribution of cultivated land has important practical significance for the effective use and protection of cultivated land resources.The text uses Dongxiang District of Jiangxi Province as the research area,and uses SPSS data processing software,Envi and Arcgis mapping software to cover the landsat8 OLI multispectral images of the study area,the land use status data at the end of 2014,the yield of cultivated land,and external factors affecting cultivated land yield.Data,to discuss the evaluation method of cultivated land quality based on NDVI inversion model;based on this,using Geoda and Arcgis software to analyze the spatial distribution characteristics of cultivated land quality evaluation results;and reusing the annual update results of farmland quality in Dongxiang District in 2014.The data was used to verify the accuracy of the farmland quality evaluation results based on the NDVI inversion model,and to investigate the accuracy and feasibility of the farmland quality assessment based on the NDVI inversion model.The main findings and conclusions are as follows:(1)The quality assessment of cultivated land based on NDVI inversion model.The NDVI weighted average of cultivated land in a certain area has a certain correlation with the grain yield in the area,and the quadratic equation model is a function model with the highest degree of fit between the NDVI weighted average and grain yield;The correction of external factors that affect cultivated land yield can be analyzed to obtain the cultivated land grain output that can be achieved under the local average climatic conditions and utilization levels.According to the classification of grain yield per unit of grain yield,the available cultivated land in the first-class land,second-class land,third-class land and fourth-class land in Dongxiang District is 11341.50 hm2,13310.65 hm2,2236.93 hm2,and 1033.47 hm2 respectively.The proportions of cultivated land of various quality and other types were 40.62%,47.67%,8.01% and 3.70%,respectively.Research shows that the overall quality of cultivated land in Dongxiang District is good,and the proportion of cultivated land with first-class land and second-class land with high quality is relatively large.In the future use and management of cultivated land,appropriate improvement in quality,etc.,will be the level of production input in the third and fourth places,and there will be full use of the productive potential of cultivated land.(2)Analysis of Characteristics of Cultivated Land Quality Evaluation.The spatial Gini coefficients of the first-class lands,second-class lands,third-class lands,and fourth-class lands in Dongxiang District were 0.68,0.11,0.36,and 0.18,respectively,indicating that the cultivated land quality in Dongxiang District was equal to the first-class land concentration.The degree is the best,and the degree of agglomeration is the worst in the second class;at the 95% confidence level,the clusters of 236 village-level units show a correlation of 54.The proportion is 22.88%;there is no correlation.There are 182 of them,accounting for 77.12%,indicating that a certain number of administrative villages within the study area are clustered in space,which is closely related to the complex terrain and rapid urban development of Dongxiang District.(3)Accuracy analysis of cultivated land quality evaluation results.In the two accuracy verification methods,the accuracy of the accuracy analysis based on the quality of cultivated land is 93.60%;the accuracy of the first place is 94.79%,that of the second place is 97.43%,and that of the third place is The evaluation accuracy is 81.27%,and the evaluation accuracy of the fourth class is 58.21%.According to the accuracy analysis conducted by townships,the area of cultivated land with consistent results in the two evaluation methods was 23609.35 hm2,and the comprehensive evaluation accuracy reached 84.55%.The research shows that the method of farmland quality evaluation based on NDVI inversion model has preliminary feasibility and certain accuracy.