THE LAND-USE AND LANDCOVER CHANGE ANALYSIS IN BEIJING HUAIROU IN LAST TEN YEARS

With eCognition software, the sample-based object-oriented classification method is used. Remote sensing images in Huairou district of Beijing had been classified using remote sensing images of last ten years. According to the results of image processing, the land use types in Huairou district of Beijing were analyzed in the past ten years, and the changes of land use types in Huairou district were obtained, and the reasons for its occurrence were analyzed. * Corresponding author 1. SATELLITE REMOTE SENSING TECHNOLOGY 1.1 General Instructions Satellite remote sensing technology has the advantages of fast acquisition, wide range and lossless information. It is widely used in land use, resource investigation, environmental testing and disaster assessment. Based on the characteristics of multi-temporal and high resolution of remote sensing data, the role of remote sensing technology in the dynamic monitoring of land use can be fully utilized, which has the characteristic of scientific and accurate, interpretation at any time and smart auxiliary decision-making. 1.2 Huairou District Huairou District is the ecological barriers and important resources guarantee in capital of China. It is a important district in building the integration of urban and rural areas in Beijing. It is also important area of the adjustment and optimization of industrial structure. In this paper, Select Huairou District of Beijing as the research object, as one of the five ecological conservation areas in Beijing. So it's a good study. 1.3 Object-oriented Method Remote sensing technology has developed rapidly and gradually been applied to many fields, as the high resolution satellite launch, how to extract abundant information from high resolution images and satisfy a certain accuracy requirement has become the research The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-3, 2018 ISPRS TC III Mid-term Symposium “Developments, Technologies and Applications in Remote Sensing”, 7–10 May, Beijing, China This contribution has been peer-reviewed. https://doi.org/10.5194/isprs-archives-XLII-3-2391-2018 | © Authors 2018. CC BY 4.0 License. 2391


Huairou District
Huairou District is the ecological barriers and important resources guarantee in capital of China.It is a important district in building the integration of urban and rural areas in Beijing.It is also important area of the adjustment and optimization of industrial structure.
In this paper, Select Huairou District of Beijing as the research object, as one of the five ecological conservation areas in Beijing.So it's a good study.

Object-oriented Method
Remote sensing technology has developed rapidly and gradually been applied to many fields, as the high resolution satellite launch, how to extract abundant information from high resolution images and satisfy a certain accuracy requirement has become the research The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-3, 2018 ISPRS TC III Mid-term Symposium "Developments, Technologies and Applications in Remote Sensing", 7-10 May, Beijing, China hot spot.

Function
ECognition is a commercially available software for high-resolution telemetry, using the "image object" as a processing unit, using a object-oriented method to categorize data and extract information.

The TM / GF-1 satellite remote sensing data
With the support of remote sensing images, the TM / GF-1 satellite remote sensing data of the three phases (2006-16, 2016 and 2016) in the last ten years (06-16) were selected as the information source.First of all, preprocessing the remote sensing images, mainly includes image fusion, orthorectification, image registration and image splicing.The next step is image digitization, including vectorization processing, establishing of points, lines, Polygon layers, checking the topology error.

obtain the object
Object-oriented image classification technology through the image multi-scale segmentation to obtain the object, when we classify remote sensing image not only rely on the spectral features of the object corresponding to the target features, but also use its geometric information and structural information, the subsequent image analyzing and processing are also based on the object.And then, combining with the field survey and colour texture analysis, the man computer interactive land classification method was used to classify the land use in Huairou district of Beijing.And after the classification and do some post processing (such as cluster analysis).

Land use types distribution map
Then we obtain the classification results, including the step of choosing a reasonable color scheme, according to the method of visual interpretation, we got the land use types distribution map through the correlation

Advancing Urbanization
With the support of ArcGIS10.2,by calculating the index of land use pattern, we got the land use structure and evolution characteristics of Huairou district in recent ten years.Urbanization is advancing, and urban sprawl is very obvious to green space erosion.Due to the management system of the construction site, the trend of the land to be replaced by the plain of the plains is more obvious.The total amount of woodland The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-3, 2018 ISPRS TC III Mid-term Symposium "Developments, Technologies and Applications in Remote Sensing", 7-10 May, Beijing, China is increasing year by year.According to the Beijing policy, every year the government has the task of afforestation.The construction of the city park in Beijing, the planning of the country parks, the reforestation of the farmland, the afforestation of the plains and the afforestation of the mountainous areas have promoted the total amount of forestland.
However, we can clearly see that the average distribution of forestland has become a small distribution in the inner suburbs of the city, and the outer suburbs are more dense, and the average degree of forest distribution has decreased significantly, among which, the decrease of the suburbs is obvious, and the urban area has a small increase.

Figures
Satellite remote sensing technology has the advantages of fast acquisition, wide range and lossless information.It is widely used in land use, resource investigation, environmental testing and disaster assessment.Based on the characteristics of multi-temporal and high resolution of remote sensing data, the role of remote sensing technology in the dynamic monitoring of land use can be fully utilized, which has the characteristic of scientific and accurate, interpretation at any time and smart auxiliary decision-making.
With the development of Object-oriented technology, the classification algorithm based on object-oriented image segmentation has developed to maturation.The most important characteristic of the classification algorithm is that the smallest unit of classification is the homogeneity Image object (patch), instead of a single pixel, this method can achieve a higher level of remote sensing image classification and extraction of object features.This method is used to segment remote sensing images of Huairou District.Firstly, the homogeneous regions (or elementary unit) are detected and extracted according to the requirements of the remote sensing image classification or the extraction of the target features, (such as spectrum, shape, size, structure, texture, shadow, spatial location, related layout, etc.), so we can achieve the purpose of classification or target feature extraction of remote sensing images.
analysis.And do carry on the field research, including selecting the sampling points based on the distribution, designing sampling routes, recording the sampling point at which land use types, and positioning by GPS.Finally, do the accuracy verification, including the accuracy of land use types based on field research data, by selecting sample points in Google Earth and loading them to check the accuracy of the classification in ArcGIS 10.2.