Based On Multi-source Data Of City Regional Income Level Estimation Method

Application of remote sensing in urban areas, gradually from the environment,ecology, resource utilization, etc., to social and economic fields, one of the more successful research is population estimation by remote sensing means, but in terms of social and economic indicators such as income level of the research is still in exploring stage. In this paper, we study the high resolution remote sensing image classification method and other methods of data, and the income level of residents as main index,social and economic analysis of high resolution remote sensing image texture feature and other natural and social factors and the relationship between the income levels of community, set up different types of different scales of urban residents’ income level estimation model. Research by means of a certain income level of the typical cities in China and the United States are calculated and compared, institute set up to estimate the precision of the model, robustness, availability, and the suitability of the practical application. Research based on the high-resolution remote sensing images, vector map and income data such as multi-source data of urban residents’ income level modeling, to reveal the urban space characteristics and social economic factors, the relationship between the further use of the means of remote sensing in the field of social economy,has important scientific significance and application value, the research content mainly includes:(1) Urban areas of high-resolution remote sensing image classification method research, study realizes the classification of the urban community, wavelet transform,texture feature extraction and classification of the training sample texture feature research. To classify the whole research data area of economic income, for the back of the characteristics of the economic income estimate model provides data, and compare the different wavelet feature extraction under different difference and the precision of model.(2) Based on multi-source data of urban regional income estimate model, including statistical analysis model, artificial neural network model accuracy and the precision of model evaluation, implement the model prediction research area economic income.(3) Urban regional income level estimation model in the application of the differences between China and the United States, including: high resolution remotesensing image of city scale community classification comparison analysis research,urban regional income estimate model comparative analysis research, urban regional income distribution analysis research.