Dissertation > Excellent graduate degree dissertation topics show

County Level Digital Soil Mapping Using Decision Tree Model

Author: ZhouYin
Tutor: ShiZhou
School: Zhejiang University
Course: Agricultural Remote Sensing and IT
Keywords: Soil digital mapping Soil organic matter Soil type Decision Tree
CLC: P285
Type: Master's thesis
Year: 2011
Downloads: 53
Quote: 1
Read: Download Dissertation

Abstract


Soil is the basis of human agricultural production, plays an important role in the geochemical cycles. Soil information is land management and environment, the basic parameters of the model. A comprehensive understanding of soil characteristics conducive to the rational utilization of environmental protection, agricultural production and land resources. Recent years, with the rapid development of precision agriculture, computer \The decision tree method is focused on reasoning tree classification rules from a set of unordered rules examples can intuitive, clearly express the relationship between variables, suitable for soil digital mapping work. This study selected a number of environmental factors, soil environmental factors and soil organic matter distribution rules of decision tree C4.5 algorithm, distribution of the number of prediction research area with limited samples of soil type and soil organic matter in space. The specific content and the main conclusions are as follows: (1) the distribution of soil type or soil organic matter has a close relationship with a variety of environmental factors. In this study, C4.5, to calculate soil type relationship - between environmental factors and soil organic matter - environmental factors, their decision tree model, respectively. (2) soil theory is based, the choice of land use classification, geological type, elevation, slope, aspect, plan curvature, slope curvature and TM images extracted normalized difference vegetation index (NDVI), normalized of humidity index (NDWI) and Soil color index (SCI) 10 kinds of environmental factors to predict the soil type of soil types in the study area mapping and evaluation and error analysis of the accuracy of prediction results. The precision evaluation results show that the soil type predicted overall accuracy of 70.5%, red soil and paddy soil evaluation accuracy is higher than the other three soil types. (3) Evaluation of Farmland census sampling points in Fuyang City of training samples chosen land use classification, geological type, soil type, topography factor, remote sensing factor for landscape parameters, analysis of individual environmental factors on the distribution of soil organic matter. And the decision tree method, based on the soil organic matter - environmental factors rules of soil organic matter in digital mapping. The precision evaluation results show that soil organic matter in the overall accuracy of 67.0%, in line with the needs of the general digital mapping. (4) using ordinary kriging, inverse distance interpolation and elevation as auxiliary parameters Cokriging three spatial interpolation method of the study area, soil organic matter level spatial distribution to predict the results accuracy were 49.5%, 50.5% and 52.4 %. The prediction accuracy of the elevation of the soil organic matter has a certain role in promoting. With decision tree model to generate the distribution of soil organic matter level results, spatial interpolation method is more intuitive to soil organic matter of the study area, the overall distribution trends, but the lack of details of the distribution of organic matter and the trend show.

Related Dissertations

  1. Fault Diagnosis Method Based on Support Vector Machine,TP18
  2. Research on the Quility of Pear and Soil Physical and Chemical Properties in Crop-Animal Mixed Orchard,S661.2
  3. Study on Dynamic Relationship between Labile Soil Organic C, N and Crop Growth under Straw Incorporation,S158
  4. Design and Development of Teaching Quality Assessment System Based on Data Mining,TP311.13
  5. Study on the Decision Tree Classification Algorithm and Its Application Based on Rough Set Theory,TP18
  6. Decision Tree Based on Generalized Information Entropy Model and Its Performance Evaluation,TP18
  7. Study on the Decision-Making Model of the Hepatitis B Immunization among the People Over 15 Years Old in China,R392.1
  8. Study on the Economic Burden of Hepatitis B-related Diseases in Guangzhou and the Strategies of Hepatitis B Vaccination for Population Aged 1～14 in China,R186
  9. Research on the Application of Data Mining Technology in the Regulating Matriculation for Postgraduate,TP311.13
  10. Research of Web Text Classification Based on Decision Tree Classification Algorithm,TP391.1
  11. Research and Implementation of Decision Tree-based Short-term Load Forecasting System,TM715
  12. Research and Application on Voyage Knowledge Discovery,F552.6
  13. A certain type of active radar -guided missiles Universal Automatic Test Equipment Research,TJ760.6
  14. Study on Extraction of Coniferous Forest Information in Southern China,TP79
  15. The Effects of Different Conservation Tillage Treatmentson Different Soil Physical Properties in the Winter Wheat Field of Oasis Area,S512.11
  16. Chinese hepatitis B vaccine effectiveness tracking Decision,R186
  17. Detection of Soil Information Based on Spectral Analysis and Digital Image Processing Technology,TP274
  18. Multi-level Selective Ensemble Learning Based on Multi-task Learning,TP181
  19. The Application of SVM to Decision Tree Induction,TP18
  20. Fuzzy Rough Set Attribute Reduction and Multi-fuzzy Decision Information Fusion,TP18
  21. Study on Data Annotation of Deep Web Data Integration System,TP393.09

CLC: > Astronomy,Earth Sciences > Surveying and Mapping > Cartography ( Cartology ) > Specialized map production
© 2012 www.DissertationTopic.Net  Mobile