2.1. Composition and Characteristics of GIS

GIS is a spatial information system, it can not only collect, store, manage, analyze and describe the data related to spatial and geographical distribution of the whole or part of the earth’s surface, The main feature of this system is the joint processing and storage of spatial data and attribute data, which fundamentally solves various inconveniences of manual cadastral survey and management, such as manual calculation and manual drawing of cadastral map. It is not only inefficient, but also difficult to store various tables and maps in the form of “paper” ^[6]. In addition, GIS can also meet the requirements of huge data management, cadastral information status, cadastral map production and rapid update, land information consultation and land use state prediction, etc. Therefore, using new technology to research and develop a cadastral management information system based on GIS not only provides efficient, real-time and accurate information service for land resource planning and comprehensive utilization in objective decision-making and micro application, but also promotes China’s economic construction and social development. GIS, an acronym for Geographic Information System, represents a spatial information system tailored to meet specific application objectives. Leveraging computer hardware, software, and networking capabilities, GIS constitutes a comprehensive technical framework that encompasses preprocessing, input, storage, querying, retrieval, processing, analysis, visualization, updating, and the application of pertinent spatial data ^[7]. GIS (Geographic Information System) stands as a pivotal and interdisciplinary field, rooted in the rich soils of Earth science and information science. It constitutes a comprehensive and intricate technical discipline system, seamlessly integrating the depths of geography, cartography, remote sensing, CAD (Computer-Aided Design) technology, database technology, and numerous other disciplines and technologies. A fully-fledged GIS comprises four integral components: a robust computer hardware system, a sophisticated computer software system, a vast repository of geographic data (or spatial data), and a team of skilled system management operators ^[8]. This holistic framework enables GIS to harness the power of diverse fields, fostering innovation and advancing our understanding of the Earth’s systems. The core part of GIS is computer system (software and components). Spatial data reflects the geographic content of GIS, while managers and users decide the working mode and information representation mode of the system. The arithmetic averaging method is used as the subjective weight, and the exponential weight and information entropy calculation formulas are shown in Eqs. (1) and (2).

The biggest difference between GIS and general information system is that it cannot only store, analyze and express the attribute information of each object in the real world, but also deal with its spatial positioning characteristics. It can combine spatial information and attribute information organically, query, retrieve and analyze each object in the real world from two aspects of space and attribute, and express the results vividly and accurately in various intuitive forms. Its main characteristics are as follows:

It meticulously organizes data in a highly structured fashion, firmly anchored upon the bedrock of geographical coordinates. This intricate process involves the meticulous construction of GIS frameworks, tailored to either longitude and latitude grids, the universal UTM coordinate systems, or the Gauss-Kruger coordinate networks, each meticulously designed to cater to the unique needs of specific administrative regions or watersheds. Within this intricate web of coordinates, every attribute data point within the map is intricately intertwined with its precise geographical location, fostering a seamless fusion of spatial and non-spatial information, ensuring a holistic and comprehensive understanding of the landscape.

It has the characteristics of multi-dimensional structure. Geographic Information System (GIS) uses geographic coordinates to form the first and two-dimensional information of spatial entities, and uses attributes of various thematic contents to form the third-dimensional information. The connection between thematic information and spatial positions is carried out by attribute codes. This provides the possibility for the comprehensive study of the internal information of spatial entities, and also provides convenience for the multi-level analysis and screening of information. The calculation formula of the consistency index is shown in Eq. (3).

The data involved in the processing have been standardized and digitized. In order to meet the needs of computer input and output, and facilitate the comparison, operation and related analysis among multiple elements, whether it is statistical data, maps or images, or some descriptive information, before participating in the processing, they are standardized and digitized according to the unified format or standard requirements, and become digital forms acceptable to computers ^[9].

The GIS environment is inherently temporal, as spatial data is inherently captured or computed within a defined point in time or a specific time period. Consequently, GIS typically encompasses several key components. One such component is data input and editing, where geographic features like points, lines, and areas within a map are transformed into vector data with precise spatial coordinates, adhering to specified geographic coordinates and boundaries, utilizing vectorization tools (such as digitizers and vectorization software). This process also involves establishing topological relationships among spatial data, enabling functionalities for topology modification, retrieval, and correction, thereby ensuring the integrity and accuracy of spatial representations ^{[10, 11]}. At the same time, the input points, lines and regions are edited by editing tools, and the spatial positions and attributes of each graph are modified. Geographic database. Geographic database is the core of GIS software and the basis and source of all geographic operations. The choice of its data model determines the effectiveness and running efficiency of GIS software. The commonly used models are hierarchical data model, network data model and relational data model. Most GIS software adopt relational data model. Data processing and analysis form the core of GIS software, setting it apart from other database and mapping software. Spatial data processing is a nuanced process that encompasses two distinct yet complementary facets: editing processing and user processing. Editing processing serves as a crucial step in refining and ‘purifying’ the spatial database, ensuring its accuracy and integrity. Conversely, user processing focuses on tailoring the data to meet the specific needs and preferences of the end-users, resulting in the creation of data files that are both comprehensive and user-friendly.

Fig. 1. Neural network TOPSIS algorithm in the water quality resource image preprocessing process.

Fig. 1 shows neural network TOPSIS algorithm in the water quality resource image preprocessing process. Data analysis mainly includes analysis type, superposition type, polygon inclusion analysis, shortest path analysis, distance, adjacency, contact analysis and buffer analysis. Data display and output. GIS can display different elements and locations such as water system and residential areas like ordinary maps; It can also be expressed by thematic map, which contains only one kind of attribute data information, and the information represented by shape and color can be expressed. Using printer, plotter can output the information needed by users. Interface. It is divided into user interface and program interface. User interface refers to user input information through keyboard and mouse and output information through display and printer. Program interface refers to the cooperation between GIS software and other software, such as remote sensing technology (RS), global positioning technology (GPS) and animation technology ^[12]. Users with certain geoscience knowledge can extract the spatial and temporal features of different sides and levels of real spatial models with the support of GIS, and quickly simulate the evolution of natural processes to predict or verify the results, and then select and optimize the scheme. This kind of rapid information simulation is almost cost-free for the existing GIS, and can avoid the loss caused by wrong decisions. Therefore, it can be said that GIS is a technical system for comprehensive processing and analysis of spatial data ^[13].

2.2. Data Organization Scheme and Data Storage in GIS

Data is a vital cornerstone of Geographic Information Systems (GIS). Prior to developing a tailored GIS, it is imperative to ascertain the content, format, and required level of accuracy of data, aligned with specific project demands ^[14]. Subsequently, the identification of appropriate data sources becomes paramount. In scenarios where the desired data content is absent from existing sources, a meticulous assessment of the feasibility of deriving the necessary content from alternative data sources is imperative. This approach underscores the importance of a holistic approach to data resource utilization, ensuring optimal integration and exploitation of available data. Should the existing data prove insufficient for generating the required content, the necessary data must be gathered independently and diligently. Geographic data can be divided into spatial data and attribute data. Attribute data is generally simple and has obvious relationship characteristics, while spatial data is complex in structure and various in variety. High quality and reasonable organization of geographic data is the key to establish GIS system, and the emphasis is also on the organization and processing of spatial data. TOPSIS The impact strength calculation formula is shown in Eq. (4).

GIS completely describes the state of spatial entities or phenomena, and abstracts three basic characteristics of the real world with data, namely spatial characteristics, temporal characteristics and thematic characteristics ^{[15, 16]}.

Fig. 2. Water quality image feature extraction and neural network training process.

Fig. 2 shows water quality image feature extraction and neural network training process. For GIS, temporal features and thematic features are often regarded as non-spatial features. These three characteristics are described below: Spatial features refer to geometric features such as position, shape and size of spatial objects, as well as topological relations with adjacent objects, also known as set features or positioning features. The distinctive characteristics of geographic or spatial information systems lie in their emphasis on location and topological features. Spatial location can be articulated through a myriad of coordinate systems, encompassing longitude and latitude coordinates, standardized map projection systems, and even arbitrary rectangular coordinates. Among the pivotal functions of GIS lies the capability to transform coordinates seamlessly across various systems, facilitating the integration and analysis of spatial data across diverse frameworks. For GIS, coordinates are the simplest and most direct spatial positioning method, and topological relations are established by calculation on the basis of spatial coordinates; Thematic characteristics refer to other characteristics of spatial phenomena besides temporal and spatial characteristics, such as slope and direction of topography, annual rainfall of a certain place, land pH value, land cover type, population density, traffic flow and air pollution degree, etc. These features can be stored and processed in other types of information systems. The spatial representation methods of these features are described in detail in traditional cartography. At present, the output methods of thematic features in GIS mostly follow the traditional thematic mapping methods, such as hierarchical (layer) coloring method and symbolic method; Time characteristics. Refers to the change of phenomena or objects with time. However, how to effectively use time for indexing and spatio-temporal analysis in GIS is still in the research stage.

2.3. Spatial Data Type

There are three types of data in GIS: Geometric data. It comes from various types of maps and measured geometric data; It not only reflects the geographical position of spatial entities, but also reflects the spatial relationship between entities. Image data. It mainly comes from satellite remote sensing and aerial remote sensing ^[17]. Attribute data, alternately referred to as statistical or thematic data, provides a nuanced depiction of the characteristics of entities beyond their spatial attributes. It encapsulates a detailed description of the target’s identity and specific attributes, offering a comprehensive understanding of the target’s nature. The precise definition of the object type stands as a cornerstone for every spatial object, rendering it an indispensable element in the definition of ground object types. Generally, attribute data is meticulously entered through keyboard input, guaranteeing the utmost accuracy and precision in capturing the distinctive characteristics of each spatial entity. Two complementary methods of input exist: the first involves direct input against graphics, allowing for instantaneous association between the data and its visual representation. The second approach entails setting up an attribute table, either by manually inputting attributes in advance or by importing them from other statistical databases, subsequently facilitating the automatic linking of these attributes with the corresponding graphic data through the use of keywords, ensuring a seamless integration of data sources.

Fig. 3. Decision process of water quality image classification based on the TOPSIS algorithm.

Fig. 3 shows decision process of water quality image classification based on the TOPSIS algorithm. Attribute data primarily comprises strings and numerical values, offering a structured representation of entities. However, the advent of multimedia has broadened the scope, with images, sounds, and textual descriptions frequently serving as descriptive features for spatial objects. Consequently, these multimedia elements can also be incorporated into attribute data collection and processing ^[18]. Topographic data, derived from the digitization of topographic contour maps, grid digital elevation models (DTMs), or alternative representations like Triangulated Irregular Networks (TINs), plays a crucial role in GIS. To facilitate the retrieval of non-spatial attribute information associated with spatial entities, it is imperative to establish a robust connection between spatial and non-spatial data, ensuring seamless integration and efficient querying capabilities within the GIS environment. Generally, the method is to attach a feature code and identifier directly to spatial entities, but it is too inefficient to input a large number of complex non-spatial data interactively. A better way to connect spatial data and non-spatial data is to connect non-spatial attribute data with digitized spatial entities such as points, lines and planes by special programs. In this way, only the spatial entity is required to have a unique identifier, which can be entered manually or automatically generated by the program and stored together with the coordinates of the graphic entity.

3.1. Exploration on Optimization of Monitoring Area Distribution

As a data acquisition point, the establishment of each monitoring point needs to invest a lot of hardware equipment. If the distribution of monitoring points is unreasonable, it will greatly waste financial and material resources ^[19]. Therefore, for the whole Three Gorges Basin, we consider selecting a certain number of representative monitoring points to ensure the integrity of water quality information in the whole water area, and reduce the number of monitoring points as much as possible on the premise of ensuring sufficient information, so as to make the layout of monitoring points more scientific and reasonable.

Fig. 4. Application Process of neural network TOPSIS algorithm in image evaluation of water quality resources.

Fig. 4 shows application process of neural network TOPSIS algorithm in image evaluation of water quality resources. The principle of optimizing distribution necessitates a strategic control over the number of sampling points to guarantee an adequate level of information capture while minimizing redundancy. In the context of water monitoring, the distribution of optimized points should be rational, ensuring that the average concentration of pollutants, as characterized by these points, aligns closely with the overall average concentration prior to optimization. Additionally, the strategic placement of fixed monitoring points at key locations is essential for targeted and effective monitoring. The relative closeness of Ti to the positive ideal solution and the formula of the i. i index are shown in Eqs. (5) and (6).

Initially, the monitoring points are systematically categorized into three distinct types: the first type, categorized as Nk, is segmented based on the monitoring sections within the basin; the second type encompasses source and key pollution sources, designated as Hj; lastly, the third type comprises locations where the establishment of fixed monitoring points is impractical, necessitating the temporary collection of data, Ft, via ship-based operations. This classification is meticulously designed to minimize data loss while maximizing real-time reflections of water quality changes, taking into account the number of input points to ensure the economical operation of the system. Subsequently, the first type of monitoring points undergoes optimization utilizing the TOPSIS method, resulting in a refined set of optimized monitoring points, denoted as Si.

3.2. Study on TOPSIS Method for Optimal Distribution of Points

TOPSIS method is a common technique for multi-objective decision analysis of finite schemes. In this paper, TOPSIS improvement is used to optimize the distribution of monitoring points.

(1) The improved TOPSIS optimization model ^[20] establishes the target distance matrix: If there are n target units and each target unit has m attribute indexes, The optimization matrix X is shown in Eq. (7).

(2) Normalized target range matrix

When discussing the practical multi-objective decision-making problem, because the dimensions of each index are different and the variation range of each index is large or small, in order to reflect the actual situation of the change of the index, each index is treated dimensionless and forms a standard distance matrix Z. The calculation formula is shown in Eq. (8).

(3) Calculate the weight wj of each index

The weight wj of the evaluation index in the target set is determined according to the relative change rate of the index value, and the specific calculations are shown in Eqs. (9) and (10).

Calculate the distance from each target element to the positive ideal solution and the negative ideal solution. The distance from each target element to the positive and negative ideal solutions is shown in Eqs. (11) and (12).

3.3. Remote Monitoring

Remote monitoring means that the local computer monitors and controls the remote equipment through the network, so that the staff can know the actual operation of the equipment without going to the site in person, and can send control instructions to it, thus achieving the same working effect as the field staff operating the equipment ^[21]. Our innovative online water quality monitoring system, tailored specifically for the Three Gorges Reservoir Area, epitomizes the pinnacle of remote monitoring information processing systems. Faced with the vast dispersion of water quality monitoring points, manual data collection becomes a formidable obstacle. To uphold the imperative of continuous monitoring, an extensive amount of data must be expeditiously transmitted to the monitoring center, thereby underscoring the need for a user-friendly human-machine interface that seamlessly integrates technology with operational efficiency. The Internet, with its unparalleled convenience and openness, emerges as the premier choice for constructing a robust remote monitoring network that fulfills these requirements ^[22].

Fig. 5. Image ISON of water quality resources by neural network optimization TOPSIS algorithm.

Fig. 5 shows image ison of water quality resources by neural network optimization TOPSIS algorithm. It is one of the core tasks of the system to use Internet to collect data remotely. At present, there are mainly two application modes of Internet-based applications: Client/Server and Browser/Server. The latter separates transaction processing logic from data processing logic and concentrates software programs, databases and other components on the server side. It has the advantages of simple use, low client configuration (only need to install a general browser), cross-platform and easy maintenance, so it has been widely used ^{[23, 24]}. For applications based on Browser/Server mode, a complete process for remote monitoring includes: collecting real-time data from the data site. Pass the data to the Web application on the Web Server side. Web applications store data into a database for processing. A web application program initiates a data request, which prompts the database to process the information and subsequently return the results. These results are then dynamically displayed on the webpage, allowing for periodic refreshes to keep the information up-to-date. To enhance user experience, animation effects are employed to showcase the dynamic display of data. On the webpage, users can interact by clicking on an image representing a controlled object and inputting relevant control parameters. This action triggers the return of the updated controlled object representation and parameters to the web server, which subsequently forwards the data to the appropriate program for execution. At present, there are two ways for Web applications to extract data. One is to use database technology to put the data extracted from the industrial field into the database, and let the Web application operate the database. Another method is to use component technology, use Visual Basic or Visual C++ to write components, let the components to call industrial real-time data, put the data into memory, and Web applications directly contact with components, call the data in memory. Components can call industrial real-time data through DDE or OPC ^[25].

3.4. Web Database Access Technology

The realm of Web database access technologies encompasses a diverse array, including CGI, ASP, JSP, and PHP. In this paper, we undertake a comparative analysis of CGI, ASP, and PHP, with a particular focus on CGI. CGI, or Common Gateway Interface, serves as a pivotal interface protocol specification, facilitating seamless interaction between web servers and diverse database servers. This robust framework relies on a standardized set of parameter formats and environment variables, ensuring compatibility and efficiency across platforms ^[26]. At the heart of CGI’s functionality lies its capability to augment the capabilities of web servers, establishing a bridge between them and database servers. Furthermore, CGI programs enable dynamic interaction with browsers, enhancing the user experience by responding to user inputs and queries in real-time. This synergy between server-side processing and client-side interactivity underscores the vital role CGI plays in modern web-based applications. For example, a CGI program can obtain data from a database server, convert it into HTML documents and send them to the browser, or store the data obtained by the browser to the database server. In this way, client users can conveniently interact with the WEB server and realize data processing. CGI programs can be written in any language, usually C, Perl, C + +, Java, VB and so on.

Fig. 6. Plot of recognition accuracy of neural network TOPSIS algorithm under different water quality parameters.

Fig. 6 shows plot of recognition accuracy of neural network TOPSIS algorithm under different water quality parameters. The specific process is that the client uses TCP/IP protocol to establish a connection with the WEB server through port 80 and send requests. The WEB server assigns values to several environment variables, then runs CGI program, and transmits the information in the request body received from the client to CGI program. The CGI program makes a request to the database server. The database server returns the request result to the CGI program. The CGI program returns the processing results to the WEB server. The WEB server returns the result to the client and disconnects. CGI performance analysis: CGI can be written in almost any program language, and almost all WEB servers support CGI, therefore, CGI for WEB server versatility is better. However, CGI is an external program of WEB server, and the communication between CGI and WEB server belongs to inter-process communication. Whenever clients produce HTTP requests, they must start a CGI process in the server and allocate new resources, which costs a lot. When the user traffic increases, the performance of WEB server drops sharply. Secondly, the inefficiency in accessing the database arises from the fact that every database query necessitates the establishment of a new connection followed by its immediate termination. This repetitive process, even when the web server has frequently interacted with the database before, results in a notable slowdown. Thirdly, the system’s limited portability poses a formidable challenge, necessitating the creation of bespoke CGI programs tailored exclusively to individual database servers. This approach not only prolongs the development cycle but also demands a high degree of expertise from developers, who must navigate the intricacies of each server’s unique requirements. Turning our attention to ASP (Active Server Pages), this open and dynamic page construction technology offered by Microsoft revolutionizes web development.

Fig. 7. Distribution map of water quality image classification and evaluation results based on the neural network TOPSIS algorithm.

ASP seamlessly blends HTML, scripts, and reusable ActiveX Server components, empowering developers to create efficient, dynamic, and web-centric database application access environments. By embedding scripts within HTML text, ASP integrates dynamic content seamlessly into static pages, fostering a seamless and intuitive user experience. At the same time, ASP adopts the object-oriented feature and expands ActiveX control, which can realize the dynamic access to Web database. ASP uses ADO (Active Data Object) to access the database, that is to say, the details of accessing the database are highly abstracted by establishing objects.

Fig. 7 shows distribution map of water quality image classification and evaluation results based on the neural network TOPSIS algorithm. The core ADO of ASP method includes seven objects, which are: Command (which defines the Command to operate on the database), Connection (which indicates the Connection to establish a data source), Err (which provides the information of data access errors), Field (which indicates the fields of general data types), Parameter (which indicates the parameters of Command objects), Property (which indicates the dynamic characteristics of ADO objects defined by the data provider) and Recordset (which indicates all the result sets generated by database commands). Among the seven objects, Recordset is the main interface for data. ASP features that ASP scripts are integrated into HTML. It is easy to generate, the development of the page is exactly the same as HTML, with a common editor can be, without compiling and linking can be directly interpreted and executed. Browser independent. As long as the client uses a browser that can interpret conventional HTML code, it can browse the homepage content designed by ASP. The ASP script is executed on the server side of the site, and the browser on the client side does not need to support it. ASP scripts leverage object-oriented and extensible ActiveX Server Components, enabling seamless integration with system and built-in ASP components. Custom ActiveX Server Components can be easily crafted to expand functionality, offering unparalleled versatility. Compatible with any ActiveX scripting language, ASP files are identified by the asp extension, differentiating them from HTML files that may also embed scripts. The source code remains secure as ASP scripts execute on the server, transmitting only the resulting, conventional HTML code to the user’s browser, thus mitigating the risk of source program theft.