(Zhengzhou 4500 16, Information Center of Henan Provincial Department of Land and Resources)
Through the research and exploration of high-resolution satellite remote sensing image data (SPOT5) processing and database construction technology, the technical requirements of high-resolution satellite remote sensing image data processing and land use database construction and the provincial land use database standard based on 1 ∶ 10000 remote sensing image were formulated. The digital orthophoto map of 1 ∶ 1 10,000 covering the whole province of Henan was made, and the graphic image database of GPS image control points based on SPOT 5, high-resolution satellite image database and Henan land use database based on image information were established, which made a beneficial exploration for making the basic map of the second national land use survey.
Keywords: image control points of remote sensing database of land and resources satellite images
0 quotation
With the rapid development of information technology, satellite remote sensing image processing technology has made a breakthrough, and the application of high-resolution satellite images in land resources investigation and evaluation, land use dynamic remote sensing monitoring, land law enforcement supervision, land change investigation and large-scale topographic mapping has achieved remarkable results.
Aiming at the task of high-resolution remote sensing image data processing and database construction in Henan Province, the project team proposed to use GPS field static measured coordinates as control data for image data correction, and formulated Technical Requirements for High-resolution Remote Sensing Image Data Processing and Land Use Database Construction and Provincial Land Use Database Standard Based on1:110,000 Remote Sensing Image. According to the requirements of the project task, a land use classification system based on remote sensing image information suitable for Henan Province is established. At the same time, through the development of the project, the digital orthophoto map (DOM) covering Henan Province was made, and the graphic image database of GPS image control points in Henan Province based on SPOT 5 was established, which made a beneficial exploration for making the basic map of the second land use survey.
1 technical route of image data processing and database construction
(1) Multi-source remote sensing information. Multi-spectral images with the best band combination are fused with high-resolution panchromatic images to produce fused images with high-resolution spatial information and rich spectral information.
(2) The combination of GPS image control points, basic map (database) and DEM. According to the actual situation, GPS image control points are used, and 1∶50000 DEM is used for orthorectification of remote sensing images.
(3) The combination of human-computer interaction and computer automatic extraction. Based on human-computer interactive interpretation, land classification information is extracted.
(4) Combining remote sensing interpretation with ground investigation. The extracted land map information is verified on the spot, and the indoor uncertain land map points are investigated on the spot.
Establishment of graphic image database of GPS image control points
In order to ensure the selection accuracy of image control points, firstly, according to the technical requirements of image control point selection, 25 image control points are evenly selected in each scene, and the static measurement of image control points is carried out in the whole field. Edit the attribute structure of image control points under MapGIS platform, establish the graphic image database of GPS image control points, and save the field measurement results table of image control points in the attribute table in the form of pictures. As shown in figure 1.
Figure 1 Schematic diagram of image control point graphic image database
2. Selection of control points for1GPS images
In order to ensure the field measurement accuracy of image control points, when selecting image control points, the distribution of points should be relatively uniform, the characteristics should be obvious, the traffic should be convenient and the number should be sufficient. Image control points should be selected on full-color images as far as possible to avoid interference factors such as high-voltage lines and large areas of water.
In order to improve the efficiency of field survey, the selected image control points to be measured are made into "Field Survey Results Table of Image Control Points", which includes the number of image control points, the schematic diagram of points and enlargement, three sets of coordinates and point descriptions of WGS84, 1954 Beijing and 1980 An, etc.
2.2 GPS image control point field survey
The field survey of image control points adopts the attached line method, and the average distance between image control points is about 13 km. The image control points and C-class GPS control points form a GPS control network. In the field survey of GPS image control points, three groups of data (WGS 84, 1954 Beijing coordinates and 1980 Xi 'an coordinates) obtained by the C-class GPS control network in Henan Province are used as the initial data. According to the global positioning system (GPS) measurement specification, three sets of GPS receivers are used for synchronous observation in a static manner, and the observation period is not less than 45 minutes, and the satellite altitude angle. South surveying and mapping software is used for baseline calculation, adjustment and elevation fitting of survey data. Finally, three sets of data and fitting elevation of image control points are calculated based on three sets of coordinate systems.
2.3 Establishment of graphic image database of GPS image control points
The graphic image database of GPS image control points takes the geographical base map of Henan Province/Kloc-0: 500,000 as the working base map, inputs the spatial coordinates of image control points, collects the attributes and graphic information of image control points, and establishes a graphic image file of image control points with a unified mathematical basis. The graphic image information of the image control point includes not only the geographic coordinate information of the image control point, but also the texture information and resolution information of the feature objects related to the image to be corrected.
3 image data processing
Image data processing includes registration and fusion of satellite panchromatic data and multispectral data, orthorectification of image data, mosaic and production of orthographic images, etc. The SPOT 5 data used in this project is 1A data provided by Shibao Company, which is only balanced by detectors. In order to combine various data and make orthophoto map, it is necessary to correct orthophoto map and establish geographical coordinates. The technical process of image data processing is shown in Figure 2.
Fig. 2 Technical process of image data processing
3. 1 image registration
Single scene multi-spectral data and panchromatic data used in this project are received synchronously, and their graphics have good geometric correlation. The registration of multi-spectral data and panchromatic data is difficult and the accuracy is not high, so the relative registration method is adopted. The band combination of SPOT 5 multispectral data is in the form of XS2 (red), XS3 (green), XS 1 (blue), the image resampling interval is 2.5 m, and the resampling method is bilinear interpolation. Taking the scene as the registration unit and SPOT 5 panchromatic data as the registration basis, the registration control points are evenly selected. For areas with large viewing angle on the receiving side and serious influence of terrain fluctuation on registration, the density of control points is correspondingly increased, so that SPOT 5 multispectral data can be accurately registered with it, and points with the same name are randomly selected for inspection on the registered panchromatic and multispectral data to ensure the registration accuracy of the data.
3.2 Image Fusion
In image fusion, the most basic product merging algorithm is used to directly synthesize remote sensing data with two spatial resolutions. After the fusion, through histogram adjustment, USM sharpening, color balance, chroma saturation adjustment and contrast enhancement, the color of the whole image is uniform, moderate and clear, and the thematic information, especially the texture information, is enhanced.
3.3 Image Orthophoto Correction
Image orthorectification adopts LPS orthorectification module of ERDAS and SPOT 5 physical model. Twenty-five image control points in each scene are evenly distributed in the whole scene image, and there are more than two * * * points in the overlapping area of adjacent scenes. Orthorectification is based on the measured points and 1∶50000 DEM, and the fused data is orthorectified in the unit of scene, and the sampling interval is 2.5 m. ..
3.4 Image stitching
Image mosaic adopts batch processing module in LPS orthographic projection module of ERDAS, and more than two points are collected from two adjacent images, which greatly improves the accuracy of image mosaic. In order to verify the mosaic accuracy, more than 25 checkpoints are randomly selected in the mosaic area of a county (city, district) to check the mosaic accuracy.
3.5 Digital orthophotomap Production
Digital orthophoto map (DOM) is made by using Image Info tool, and it is cut according to the standard of 1 ∶ 1 10,000, covering a complete county-level administrative area. According to the Technical Requirements for High Resolution Image Data Processing and Database Construction, the standard map of 1 ∶ 1 10,000 is completed by using MapGIS database platform and according to Beijing coordinate system 1954 and national elevation datum 1985.
4 innovation achievements
On the premise of successfully completing the project tasks, the project team creatively carried out the work in combination with the project schedule and land management needs. Summarizing the progress and achievements of the project, the innovative achievements are mainly reflected in:
Using the measured data of (1) image correction control point GPS field as image correction control data, the traditional method of using topographic map and land use status map (database) as control data is changed, which greatly improves the accuracy of image correction and saves engineering investment.
Covering the standard topographic map of 1∶ 1 10,000 in Henan Province, there are only over 5,600 topographic maps. On the premise of seeking the consent of the research group of the Ministry, the project team put forward the idea of using GPS field survey control points as image correction control data. Based on this idea, the project team conducted a series of research and demonstration, formulated the technical requirements for GPS field survey, and selected 25 control points for each SPOT 5 satellite image with relatively uniform coverage in the whole province. The principle is that there are no less than 2 control points in adjacent scenes. A total of 142 1 image correction control points and 94 GPS geodetic control C points were selected in the whole province. According to the receiving time of the image data and the progress of the project, * * is divided into 13 survey area, and all the control points are statically measured by the attached line method, and three sets of coordinates of WGS84, 1954 Beijing and 1980 Xi 'an of each control point and checkpoint are calculated respectively.
(2) The establishment of graphic image database of image control points in Henan Province provides technical support for remote sensing monitoring of land use and law enforcement supervision of satellite films in Henan Province in the future.
In order to preserve the field survey results for future use for a long time, the project team established a graphic image database of GPS image control points on the MapGIS platform based on the geographical map of Henan Province/KLOC-0: 500,000. The establishment of the graphic image database of GPS image control points not only meets the correction accuracy requirements of SPOT 5_2.5 m high-resolution satellite images, but also lays a foundation for the future land use remote sensing monitoring, satellite photo law enforcement inspection and mine environment monitoring in Henan Province.
(3) Exploring the large-area global orthorectification and mosaic processing technology of high-resolution image data is conducive to the popularization of image data batch processing technology.
Due to the wide range of image processing and heavy workload in this pilot project, the project team explored and used the large-area overall orthorectification and image mosaic processing functions provided by the LPS module of professional remote sensing image processing software ERDAS on the premise of ensuring the accuracy of image correction, and achieved good application results.
In view of the fact that the image data used in this pilot project are all synchronously received SPOT 5 multispectral and panchromatic data, and the geometric correlation of their graphics is good, and the registration of multispectral data and panchromatic data is difficult and accurate, therefore, the image data processing adopts the technical process of single scene fusion first, then large area orthorectification, and finally large area mosaic registration.
Orthophoto correction adopts LPS batch orthographic module of ERDAS. The physical model of SPOT 5 is used for correction, and the control points are evenly distributed in the whole scene image, each scene has 25 control points, and the overlapping area of adjacent images has more than 2 * * * points. Orthophoto correction is based on the GPS control points measured in Henan Province and the preprocessed1∶ 50,000 DEM, and the SPOT 5 fusion data is corrected in batches with a sampling interval of 2.5 m. The image stitching adopts the LPS batch processing module of ERDAS. Because more than two * * * points are collected in each adjacent scene image, the accuracy of image mosaic is greatly improved.
(4) The establishment of land use classification standard system based on remote sensing image information has made a beneficial exploration for the national and provincial level to quickly grasp and extract land use change information.
According to the requirements of the research group and the needs of national and provincial land management, the project team has formulated a land use classification standard based on remote sensing image information, which is suitable for Henan's reality and meets the needs of the pilot project of "high-resolution image data processing and database construction". In this standard, land use types are divided into three categories: agricultural land, construction land and unused land. Cultivated land, garden land, other agricultural land, urban land, urban land, rural residential land, railway land, highway land, other construction land and unused land are classified as 10. In addition, according to the characteristics of individual land types, five three-level categories such as highway forest belt, agricultural water conservancy land, water conservancy facilities land, unused water surface and Yellow River beach are divided into agricultural land, construction land and unused land respectively.
(5) The construction of land use database based on remote sensing images provides powerful electronic data of land use for national and provincial macro-management of land, and provides a technical basis for similar work in China.
Considering the needs of national and provincial land macro-management, according to the "land use classification system based on remote sensing images" formulated by the project, combined with the framework of MapGIS land use database management system of China Land Corporation, the project team is based on MapGIS database management system platform. The Technical Requirements for High Resolution Image Data Processing and Database Construction and the Land Use Database Standard Based on 1 ∶ 1 10,000 remote sensing images were formulated respectively, which defined the land classification, file naming rules, data layered format and requirements based on remote sensing images, ensured the consistency of data standards and data formats and the quality of database construction, and provided accurate land use status data for the whole country and the whole province.
5 conclusion
With the rapid development of remote sensing technology and computer technology, the application of high-resolution remote sensing image data in land management is becoming more and more common. At the same time, the technology of remote sensing image data processing is becoming more and more scientific and advanced, especially with the comprehensive development of the second national land survey, the application of remote sensing image in land management has been pushed to an unprecedented height. Therefore, in the process of image data processing, it is particularly important to reduce the input of manpower and financial resources as much as possible. In order to solve the above problems, in the process of scientific research and production, this project proposes to use GPS field survey control points as image correction control data, establish a database of GPS image control points and a standard of land use database based on remote sensing image information, and make a good interpretation according to the needs of national and provincial land management, which is a useful exploration for similar work in the future.
Take the exam and contribute.
Chang Qingrui et al., 2004. Introduction to remote sensing technology [M]. Beijing: Science Press.
Chen, wait. Research on Remote Sensing Information Mechanism [M]. Beijing: Science Press.
Don Rong 'an et al., 2003. Remote sensing image processing method of ERDAS imaging [M]. Beijing: Tsinghua University Publishing House.
Guoan Tang et al., 2004. Remote sensing digital image processing [M]. Beijing: Science Press.
Xu Dao. 1988. Orthographic projection technology and image map [M]. Beijing: Surveying and Mapping Press.
You Shuzhu, Liu Shunxi. 2002. Application of GPS in land change survey [J]. Bulletin of Surveying and Mapping (5): 1 ~ 3
Zhang Jixian et al. 2000. Database of graphic image control points and its application [J]. Bulletin of Surveying and Mapping (1)
(Originally published in Bulletin of Surveying and Mapping,No. 10, 2008)