Resolution Remote Sensing Image Change Detection Based On Applied Research

Timely and accurate grasp of the changes in the waters of Poyang Lake,contribute to a correct evaluation of the climate factors and the impact of human activities on the change of water resources in the region. It is very important to the rational allocation of water resources、maintenance of ecological balance and sustainable development of water ecosystems. With remote sensing image change detection technology has been widely applied, change detection based on remote sensing image classification in various fields of water resources survey, urban resource management, environmental disaster prediction and environmental mitigation, will have a more important role. Due to the high-resolution satellite imagery with a wealth of detail and spatial information with obvious structural information, Making high-resolution remote sensing image data processing to get better results visualization, levels of information are better rich, to further enrich the content of the extracted and improve the robustness and accuracy of the study area extraction.This paper describes the geometric correction, image fusion and classification of the key technologies in change detection pretreatment. This paper chose the five phase of the high resolution remote sensing image in the same area of the waters of Poyang Lake. First apply a rational function model for image geometric correction as image required test data,thus comparison of six kinds of fusion to get. In this paper, the wavelet transform is the best fusion experiments. And comparative analysis of the characteristics of supervised classification and unsupervised classification of images for supervised classification, obtained experimental video of change detection. Summarizes the strengths and weaknesses of the four methods of change detection: Algebra change detection method, Transformed image change detection method, Image classification change detection method and Characteristic change detection method. In this paper, the characteristics of high-resolution remote sensing image applied to Image classification change law, and separate category of two images before and after. In the area has been classified sequentially comparing each pixel to get attribute information changes. By subjective and objective analysis of the Poyang lake waters four groups change detection results from January to May, May to July, July to November, November to December. Experiments show, Using classification change detection method of comparative law analysis of the resulting water changes And together with the waters of Poyang Lake hydrological and meteorological data to make a correlation analysis are in agreement.Multi-temporal remote sensing image change detection by this method will not cause the problem that environmental conditions and sensor acquisition brings, such as radiation normalized geometric registration, etc. Also eased the "synonyms spectrum with the spectrum foreign body" problem caused by external factors in multi-temporal image change detection work. After comparing post-classification comparison method directly get the changing type, location and quantity. In change detection process, This method is not require much for original data, And there is no limitation of the number of times.