Application of GIS image system and remote sensing technology in physical geography land planning

Physical geography is the foundation of urban social and economic development. The quantity and quality of land resources and their distribution directly affect the economic, social, environmental and comprehensive benefits of the city. Now, in our country, low land use efficiency and irrational land structure in physical geography are both prominent phenomena. One of the main reasons is the lack of scientific and reasonable physical geography land use structure and utilization efficiency plan. Using remote sensing and geographic information system technology, this paper conducts a comprehensive and systematic survey of urban land use changes and obtains land use classification maps for two periods, supports further research and correctly guides people to develop and utilize natural resources and protect the ecological environment and provide powerful reference materials for realizing sustainable land use and sustainable social and economic development. In addition, this article will also introduce multi-source remote sensing image data fusion technology, including fusion information representation, fusion principle, fusion system framework model, fusion algorithm, control and application. By combining ETM remote sensing data with large-scale topographic maps, GIS is used as a supporting tool to construct a spatial database of urban environment, which promotes the development of research.


Introduction
Land is the environment for human survival and the pillar of all human activities (Wang et al. 1501). But over the years, uncoordinated development patterns have made the contradiction between people and land in space more and more serious Feichtenhofer et al. 2016). Due to the changes in land use and land cover caused by human activities, it has a great impact on the changes in the earth's environment ).
Research on land use change is an important method for rational use of land and promotion of sustainable development. Natural geographical land is the foundation of urban social and economic development (Tu et al. 2018).
The quantity and quality of land resources and their distribution directly affect the economic, social, environmental and comprehensive benefits of the city (Bilen et al. 1612). Now, in our country, the problem of low land use efficiency and irrational land structure in physical geography is significant. One of the main reasons is the lack of scientific and reasonable physical geography land use structure and utilization efficiency plan. The theme of this thesis is ''The Land Plan Application of Physical Geographical GIS and Remote Sensing Technology Image System.'' It is the first physical geographic remote sensing survey on the status of land use with the support of remote sensing and GIS technology (Karpathy et al. 2014). The GIS system uses the evaluation model to construct the urban land adaptability evaluation, including the survey results, and finally realizes the optimization of land allocation with the support of GIS, and makes reasonable plans for urban land use (Fernando et al. 2017).
Remote sensing technology is often applied in various aspects of land planning (Yang et al. 2012), among which: (1) land use classification and remote sensing changes and (2) dynamic monitoring of land use changes. Through the development of remote sensing technology, it provides a rich data source for the land use plan of physical geography. Remote sensing technology can be divided into aerial remote sensing and satellite remote sensing (Wang et al. 2016). There are many successful examples of aerial remote sensing suitable for urban integrated research (Simonyan 1409). For example, integrated aerial remote sensing surveys have been implemented in large cities such as Beijing, Tianjin and Guangzhou, and good results have been achieved. With the improvement of the spatial resolution of satellite remote sensing technology, this application is becoming more and more popular (Chatfield et al. 1405).

Related work
According to the literature, many reports at home and abroad have introduced satellite remote sensing technology to obtain basic data required for physical geography and land planning (Althloothi et al. 2014). According to the literature, our country's physical geography and land planning departments used satellite remote sensing technology to carry out part of the monitoring work on physical geography of land use surveys and urban changes . The literature points out that the application of satellite remote sensing technology in physical geography and land planning can be summarized in three aspects: urban spatial distribution analysis, urban change monitoring and plan implementation inspection (Wang et al. 2014). In the process of physical geography and land planning, a lot of basic data needs to be mastered. In the past, obtaining these basic data required field investigations, a lot of human and material resources and long-term investigations (Khaire et al. 2018). The literature analyzes a number of applications of GIS technology in the commercial and academic circles and proposes that the development of GIS in the academic field is currently superior to that in the commercial field (Liu and Wang 2018), and mature GIS technology is not applied in the commercial field until it has been verified by the academic field ). 3 Land planning system GIS image system and satellite remote sensing key technology research 3.1 System overall architecture design Before the system function design, we must first design the system according to the overall design system, overall design and then the detailed function design. The overall structure design of the system is shown in Fig. 1. Each functional module is independent of each other and organically combined to form a complete service system.

Remote sensing image data processing for land planning supervision
The noise formula of remote sensing image is: The average filtering algorithm also uses the average value to replace the current point pixel value. In the case of image selection, the size of the template W is W * n, and the average filtered graph g(x, y) can be expressed as Eq. (2): The hybrid filtering algorithm is adopted to eliminate noise, and its algorithm formula is as follows: The filtered image is calculated as: Here, M4X represents the grayscale peak value of the pixel. In the case of a grayscale image, the filtered image is expressed as G(x, y), and the formula for calculating the mean square error of the pixel is as follows: Design the global transformation model: H here is a 3 9 3 two-dimensional transformation matrix, which can be expressed as follows: Similarity transformation is a model that only contains two basic transformations: parallel movement W and rotation. W can be expressed as follows: The affine transformation model includes a variety of transformations. The conversion model is as follows: The projection transformation model is the most common model including the left and right possible transformations, and its formula is generally expressed in a broad sense.
For data preprocessing, including any modification, registration, image process modification and splitting, shading and some spectral normalization links, in addition to these conventional preprocessing, more important preprocessing includes remote sensing for ease of interpretation and classification. The data processing flow, as shown in Fig. 2, is from band coupling, image enhancement and image conversion and so on. From the point of view of physical geography, the main flow chart of land use plan has many research methods of remote sensing data preprocessing at home and abroad, and it fully introduces the diagrams related to remote sensing digital image processing. There are many documents on geometric correction and positioning, among which the geometric correction and positioning of QU1CKBIRD remote sensing satellite image are being studied. In addition, many researches on the fusion of multiple remote sensing images such as Quierkbird image and IKONOS remote sensing data fusion method have been carried out. Regarding the research of remote sensing image enhancement, according to the characteristics of various clouds, Lai Geying et al. use strong remote sensing image processing.
In the database construction process, after the conceptual structure design and logical structure design are completed, the table structure needs to be designed. The design of table structure is directly related to data storage and plays an important role in database design (as shown in Table 1).

System simulation experiment design
Information entropy is a function of the probability distribution of information in the current system. In the case of experiment A with N possible results, N possible results can be expressed as follows: a 1 , a 2 , a 3 …aN, then the probability of their appearance can be P(a i ,)(i = 1,2,...N), then the information entropy function H(A) can be expressed as: Application of GIS image system and remote sensing technology in physical geography land… 8405 When information entropy is applied to image processing, the information entropy H(I) of image I is expressed by the following formula. The joint histogram method is widely used because of its simple and intuitive operation. The combined histogram W can be represented by a two-dimensional matrix: Next, the correspondence of these joint probability densities is expressed in formula (13) as follows The Parsons window method is proved to be more accurate than the joint histogram method, but the calculation is more complicated. The window function is used to directly estimate the probability density. For a sample with N sampling points, the probability density of point A can be expressed as follows.
In this article, the B-spline function is used as the window function of the pulse window method. Calculated as follows: In order to extract the point features of the scaling transformation with invariable robustness, it is necessary to detect and extract them in different differential pyramid spaces. You must first build a sample pyramid. The SIFT inherent point extraction algorithm uses a Gaussian convolution function to construct a scale space pyramid. Specifically, it can be expressed by the following formula.
When the SIFT algorithm is used to construct the proportional space, different ratios of image differences are obtained to improve the stability and firmness of the detected feature points. In order to increase the speed of the algorithm, David G. Low has also been improved, using the Gaussian difference proportional space (Dog, Difference ofsian) Gauss, as follows: Expand the difference of Gaussian function (DOG) Taylor at the extracted extreme point and set its derivative to 0, as shown in the following formula, the correct position of the extreme point can be obtained.
Let oD x;y;r ð Þ X ¼ 0, the extreme point offset can be calculated 9 max The specific steps of remote sensing image fusion are shown below, and the flow chart of the algorithm is shown in Fig. 3. (1)  As shown in Fig. 4, the line connecting the red and blue lines represents the correct matching points of the probability density function matching and the error. You can take r, 0.8, and you can use the filter to exclude more than 90% of the wrong matching points, and the number of matching points modifying the matching error filter below 5% is the ideal threshold, so the evaluation value in this article is 8. In this paper, the K-D tree is recursively established by searching the nearest neighbor defined position and the second neighboring point search in the K-D tree, and the search efficiency is improved through space segmentation, and the matching calculation time is shortened.

Experimental results and analysis
Spectral information can directly distinguish different features on the image. This article compares and analyzes the spectral information of different features in different bands and index characteristics and conducts statistics on the spectral features of different features. It is found that the spectral features of different types of features are different. Same, different features can be distinguished to a certain extent. According to the training samples of vegetation, forest, grassland and construction land in the water and the pre-processed remote sensing image, the spectrum features of the various ground features in the above-mentioned various frequency bands and the scattered index features are shown in Table 2, respectively. Extract the minimum, average and maximum values (Fig. 5).
Therefore, this paper selects 3 9 3, 5 9 5, 7 9 7, 9 9 9…21 9 21 windows to extract the texture features of various ground objects and compares and analyzes the extraction results. Among them, the texture features are shown in the figure (as shown in Table 3).
As can be seen from Fig. 6, as a whole, the average separability of various features varies with the size of the texture window. As a whole, the separability of features increases as the window becomes larger, and individual features can be the degree of separation fluctuates to a certain extent, and the degree of separation of various objects is relatively stable from 7 9 7 window to 17 9 17 window. The most obvious change in separability is woodland. The separability of other ground features is smaller in 3 9 3 and 5 9 5 windows, and the middle window from 7 9 7 to 17 9 17 fluctuates little, among which 9 9 9 window and 13 9 Compared with the 11 9 11 window, the 13 window and 15 9 15 window have a lower separation degree, and the 19 9 19 and 21 9 21 windows have smaller fluctuations.
According to Table 4, the secondary information characteristic amount of the algorithm-based non-grid SURF matching error is about 1 to 2 pixels on average, and the river image matching error is about 2.5 pixels, and the  registration algorithm of the polynomial fitting shape is used for river image registration matching. The error is small, only 1.4 pixels. The visualized shape content registration algorithm based on polynomial approximation can obtain good results for remote sensing images with more obvious curve characteristics and has high robustness to rotation and scaling. As can be seen from Table 5, the large difference between different gray levels and the gray features is based on the shape of the polynomial fitting content feature matching algorithm, because the multi-source remote sensing images from different sensors cannot be displayed on time. The improved SURF algorithm based on the traditional SURF algorithm and the secondary information characteristics of the grid SURF algorithm has obvious advantages.
The cross-correlation coefficient (NCC) and root-meansquare error (RMSE) of the overlapped area of the image  Application of GIS image system and remote sensing technology in physical geography land… 8409 after the registration restoration are estimated, respectively, and the results are shown in the table.
It can be seen from Table 6 that the satellite remote sensing image of Bolivia's capital Lajis has the highest registration accuracy with the Google map image of the same location. The shape content feature registration algorithm based on polynomial coincidence proposed in this paper can be accurately extracted from the database. The Google map image corresponding to the required satellite image is produced, which further proves the reliability and anti-interference of the algorithm in this paper.   channels in the northern part of J Province, which mainly depends on the advantages of dry period irrigation and flood period drainage. The construction land is mainly distributed in rivers and tree shades around the government, and the land of rural residents is also concentrated on flat land. From 1988 to 2000, the changes in the waters increased slightly. The second reason is population. As the population increases, the amount of arable land resources needed for human life and survival is constantly changing, and the shortage of land resources is caused by population pressure. Under this pressure, humans often develop new cultivated land and strive to increase the average amount of cultivated land, resulting in a continuous increase in the total amount of cultivated land resources. Changes in forests and grasslands are also greatly affected by human factors, social and economic development. Due to excessive landfilling of unreasonable production and living activities, the grassland has been damaged and the area has been reduced. Due to the rise of the tertiary industry represented by tourism, people are more and more concerned about the protection of the surrounding environment, the coverage of trees has increased, and the accumulation of forests has increased, leading to the emergence of a large number of unused forests. Another major reason for forest growth is the disappearance of grasslands. It can be seen from the distribution of grassland reduction that most of the conversion of forest is due to the conversion of grassland. The forest area in the mountains is increasing, and on the other hand, the grassland area is drastically decreasing. In addition to shifting to forests other than grasslands, other types of mutual shifts have also taken place locally. Second, there are economic factors (Gaglio et al. 2015). As the population increases, the per capita arable land area continues to decrease. In the early 1990s, food prices continued to rise. Driven by the benefits of the market economy, farmers continued to open up more planting areas and the grassland area decreased. The driving force of this economic benefit is related to the expansion of cultivated land in the survey area. With the development of the economy and the further improvement of transportation and transportation networks, the land for transportation has increased substantially. Under the domination and influence of the market economy, the scale of various construction projects such as ''development zones'' and ''real estate'' has been greatly expanded (Hu et al. 2015). At the same time, the occupation of arable land and grassland is also very significant. The main result of land use changes caused by economic development factors is the adjustment of industrial structure (Triantaphyllou et al. 1998). The primary industry is declining, and the secondary and tertiary industries are on the rise (Anagnostopoulos et al. 2008). Changes in the industrial structure will inevitably lead to the adjustment of land resources between industries and increase the area of land used for the construction of the tertiary industry. Finally, there are policy factors. Political and economic policies directly affect land use and its structure through the land ownership system, price system and management system (Du et al. 2012).

Monitoring business and technical process of land planning remote sensing system
The specific process of physical geographic land planning supervision business is shown in Fig. 7.
The dynamic degree of land use can quantitatively explain the rate of change of land use in a region, compare the regional differences of land use change and play a positive role in predicting the future trend of land use change. The dynamic degree of a single land use type represents the change in the amount of a specific land use type in a specific study area during a specific period, and its formula is as follows: Mathematically integrate the comprehensive indicators of land use to form a continuously distributed comprehensive index Comprehensive land use degree change model: The plan content of the land use plan mainly includes the adjustment of the land use structure and layout, as well as the land use plan of the main project. According to the basic situation of the collected data, combined with the basic characteristics of a certain area and the basic requirements of LUPEA work, the evaluation of the land use status and the evaluation of the land plan should have the same evaluation criteria. According to LUPEA's environmental characteristic elements and structure, as well as the recommendations of relevant experts, the indicator system of the Environmental Impact Assessment (LUPEA) of the Urban Land Use Plan is defined as shown in Table 7.

Land use planning plan adjustment and environmental mitigation measures
The environmental impact assessment program of the land use plan based on the national framework can not only comprehensively evaluate the results, but also provide the adjustment of the land use plan based on the fuzzy factor model and suggestions for environmental mitigation policies. By analyzing the index value of each element in the environmental impact assessment factor model of the urban land use plan, it is possible to determine the elements that have a bad impact on the environment at the stage of this plan and to adjust the plan and analyze the environmental mitigation policy. According to the analysis of the changes in the quality of the ecological environment in each region, the environment has changed to a certain extent before and after the implementation of the land use plan, but the situation in each region is different. Therefore, when adjusting land use plans and proposing environmental mitigation policies, concentrated measures must be proposed to specific towns and villages. Based on the above analysis, when the general land use plan is revised in a new round, the city should adjust the last round of the general land use plan, and the basic principles for adopting environmental mitigation measures should be reflected in the following aspects:  Humans must live in the ecological environment of the earth, and humans themselves are also members of the environment. When using land, humans must follow the basic laws of natural development. Specifically, in the concept of the plan, the laws of nature should be followed, and measures adapted to local conditions should be adopted to avoid excessive interference in human activities. The economic and intensive use of land is mainly manifested in the same effect of land output, with less land resource input. By using land economically and intensively, and keeping the amount of new construction land to a minimum, pollution sources can be effectively reduced.
The results of zoning considering the environment can not only realize the preliminary assessment of land use plan, but also can be used as a guide for future land use zoning. First of all, the search for more intense human activities of land use types is to move from the unavoidable environment of land use types such as planning and transportation land to moderately sensitive parts, and efforts should be made to avoid layout in the environment of severe and moderately sensitive areas., The type of land used may have an impact on the environment, and measures should be taken to delay the impact on the environment. When the new construction land is located in a medium-sensitive area, strictly manage the possibility of environmental impact, and do not arrange polluting industrial projects. It is necessary to take measures to reduce the impact of noise and pollutants on the surrounding environment for traffic land that traverses areas susceptible to moderate impacts. At the same time, in order to avoid adverse effects on species diversity, sufficient biological corridors and green roads must be ensured for transportation land that crosses environmental protection areas.
A scientific and reasonable land use model will not have a bad impact on the environment and will also have a good impact on the environment to a certain extent. Therefore, we should pay attention to the ecological use of land. The above content can be reflected in the guarantee measures for the implementation of the land use plan. In order to reduce environmental purification and concentration of pollutants, water areas and forest land types have played an important role. Therefore, on the basis of existing water resources, the layout of existing water resources, the occupation of forest resource, and the proportion of forest resources should be avoided as much as possible as the improvement and optimization of the layout. Establish a pollutant emission control mechanism, use economic, policy and taxation methods to reduce pollutant emissions, implement a total pollution source control system and establish a mechanism for eliminating industrial projects with high energy consumption and high pollution. Strengthen environmental management, increase capital investment and comprehensively treat pollutants before discharge. According to the above principles, after adjusting the new round of urban land use plan, the environmental impact assessment is implemented again, and the land use plan is adjusted to achieve the goal of improving the environmental quality. The land use plan can be realized in the improved environment of the city.

Conclusion
In this paper, GIS technology and remote sensing technology are used to construct a land resource planning and utilization model in the study area, and the empirical research and analysis of the environmental impact assessment of the regional land use plans are implemented. The state framework theory is used to establish an environmental impact assessment model for urban land use plans, and a corresponding indicator system is established. With the support of RS, GIS spatial information technology and MATLAB environment, the environmental impact assessment of the land use plan based on a certain area is completed.
Funding The authors have not disclosed any funding.
Data availability Data will be made available on request.

Declarations
Conflict of interest The authors declare that they have no conflict of interests.
Ethical approval This article does not contain any studies with human participants performed by any of the authors.