Classification method for new pollutants in game scene, medium and science popularization equipment

Abstract

The invention provides a classification method for new pollutants in a game scene, a medium and science popularization equipment, and the method comprises the steps: randomly generating various types of new pollutant icons and various virtual collection devices when a user enters a target game scene; in response to a dragging operation of dragging a first new pollutant icon selected from the target game scene to a first virtual collection device by a user, judging whether the first virtual collection device can collect the first new pollutant icon or not; if yes, it is judged that the classification of the user on the first new pollutant icon meets the requirement; and when it is determined that the classification of all the new pollutant icons meets the requirements, it is judged that the user completes the classification task of the new pollutant icons, so that the learning interest and enthusiasm of the user are improved through a gamification mode, and the new pollutant classification education effect and the environmental health protection awareness are improved.

Description

Technical Field

[0001] The present application relates to the technical field of game interaction. Specifically, the present application relates to a method for classifying new pollutants in a game scene, a medium, and a popular science device. Background Art

[0002] From the perspective of improving the quality of the ecological environment and environmental risk management, new pollutants refer to those toxic and harmful chemical substances that have characteristics such as biological toxicity, environmental persistence, and bioaccumulation. After entering the environment, they pose a relatively large risk to the ecological environment or human health, but have not been included in environmental management or the existing management measures are insufficient. Currently, the new pollutants that have received extensive attention at home and abroad mainly fall into four categories: persistent organic pollutants, endocrine disruptors, antibiotics, and microplastics. Compared with the conventional pollutants that people are familiar with, such as sulfur dioxide, nitrogen oxides, and PM2.5, people have a relatively shallow understanding of new pollutants.

[0003] Therefore, in order to enable people to have a deeper understanding of new pollutants, in the current field of environmental protection education and awareness improvement, educating people about the classification of new pollutants has become an important trend. However, the existing methods for classifying new pollutants rely on theoretical explanations, picture or video displays, which are difficult to stimulate people's learning interest and lack interactivity and practicality, resulting in poor classification education effects of new pollutants. Summary of the Invention

[0005] The main purpose of the present application is to provide a method for classifying new pollutants in a game scene, a medium, and a popular science device, so as to conduct classification education of new pollutants by means of computer technology and gamification, improve interactivity and practicality, stimulate people's learning interest, and thus improve the classification education effect of new pollutants.

[0006] To achieve the above-mentioned invention purpose, the present application provides a method for classifying new pollutants in a game scene, including: When a user enters a target game scene, randomly generate various types of new pollutant icons and various virtual collection devices; In response to a drag operation in which the user drags a first new pollutant icon selected from the target game scene to a first virtual collection device, determine whether the first virtual collection device can collect the first new pollutant icon; If so, determine that the user's classification of the first new pollutant icon meets the requirements; When it is determined that the classification of all new pollutant icons meets the requirements, determine that the user has completed the classification task of new pollutant icons.

[0007] Preferably, the randomly generating various types of new pollutant icons includes: Determine all types of emerging pollutants and obtain the real images of each type of emerging pollutant; Digitally process the real image to obtain a standard real image; Extract the characteristic elements of each type of emerging pollutant according to the standard real image, and generate virtual icons containing the corresponding characteristic elements for each type of emerging pollutant, obtaining multiple types of emerging pollutant icons, where the characteristic elements are the elements in the standard real image that best represent the characteristics of the emerging pollutant.

[0008] Preferably, the determining all types of emerging pollutants includes: Obtain the attribute information of all emerging pollutants; Extract the attribute name and pollution damage value from the attribute information of each emerging pollutant respectively; Analyze the attribute name and pollution damage value of each emerging pollutant using a pre-trained neural network model to obtain the type of each emerging pollutant.

[0009] Preferably, the randomly generating multiple types of emerging pollutant icons includes: Obtain the real image of each type of emerging pollutant. For the emerging pollutant icon of each type of emerging pollutant, perform edge detection on the real image to obtain an edge point sequence. In the edge point sequence, randomly select a continuous segment of edge points, calculate the angle change rate between every three consecutive edge points, and divide the edges into high-curvature edges and low-curvature edges according to the magnitude of the angle change rate; Depict the concave and convex shapes of the high-curvature edges according to the spline curve interpolation method, connect the low-curvature edges with a smooth curve, and connect the contours of all edge parts to obtain the emerging pollutant contour; Determine the color interval in the real image that best represents the emerging pollutant according to the color histogram analysis method to obtain the main color and color distribution, and extract the texture features of the surface of the emerging pollutant from the real image; Render the corresponding emerging pollutant contour according to the main color, color distribution and texture features of each type of emerging pollutant to obtain multiple types of emerging pollutant icons.

[0010] Preferably, the randomly generating multiple types of emerging pollutant icons includes: Determine the current game level of the user and randomly generate multiple types of emerging pollutant icons that match the level difficulty of the game level; Add a label reflecting the level difficulty to the emerging pollutant icon.

[0011] Furthermore, after determining that the user's classification of the first emerging pollutant icon meets the requirements, it further includes: When it is determined that the classification of at least one new pollutant icon does not meet the requirements, the new pollutant icon whose classification does not meet the requirements is used as the second new pollutant icon; For each second new pollutant icon, determine the key information of the new pollutant corresponding to the second new pollutant icon, and search for the popular science video of the new pollutant corresponding to the second new pollutant icon from the video library according to the key information; After screening the popular science videos, obtain the candidate popular science videos corresponding to each second new pollutant icon; After cropping and re - splicing all the candidate popular science videos, obtain the target popular science video; Display the target popular science video in the target game scene.

[0012] Preferably, the judging whether the first virtual collection device can collect the first new pollutant icon includes: Determine all the characteristic information of the new pollutant icons that the first virtual collection device can collect, and obtain a plurality of first characteristic information; Obtain all the characteristic information of the first new pollutant icon, and obtain a plurality of second characteristic information; Respectively perform vectorization processing on the plurality of first characteristic information and the plurality of second characteristic information to obtain a plurality of first characteristic vectors and a plurality of second characteristic vectors; Perform clustering processing on the plurality of first characteristic vectors and the plurality of second characteristic vectors respectively, and screen to obtain a first target characteristic vector and a second target characteristic vector. The clustering processing includes: Randomly select a vector from all the characteristic vectors as the initial clustering center, calculate the Euclidean distance between all the characteristic vectors and the initial clustering center, allocate all the characteristic vectors to different clusters according to the Euclidean distance, and calculate the average vector of the plurality of characteristic vectors as the new clustering center. Replace the initial clustering center with the new clustering center, and return and repeat the step of calculating the Euclidean distance between all the characteristic vectors and the initial clustering center until the change of the new clustering center is less than the preset threshold, and use the new clustering center as the first target characteristic vector or the second target characteristic vector; Calculate the cosine distance between the first target characteristic vector and the second target characteristic vector. When the cosine distance is greater than the preset cosine distance, it is determined that the first virtual collection device can collect the first new pollutant icon.

[0013] Further, after judging whether the first virtual collection device can collect the first new pollutant icon, it further includes: If not, determine the pollution type and pollution damage value of the new pollutant corresponding to the first new pollutant icon; According to the position of the first new pollutant icon in the target game scene, determine the rendering center point of the pollution special effect; Selecting a matching special effect object according to the pollution type of the new pollutant corresponding to the first new pollutant icon, and setting basic properties of the special effect object; The state of the special effect object is dynamically adjusted according to the pollution damage value, and the dynamically adjusted special effect object is drawn in a semi-transparent form to a rendering center point in the target game scene using a rendering system of the game engine.

[0014] The present application also provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for classifying new pollutants in a game scene as described in any of the above items is implemented.

[0015] The present application also provides a popular science device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method for classifying new pollutants in the game scene described in any one of the above items are implemented.

[0016] The present application provides a classification method, medium and popular science equipment for new pollutants in a game scene. When the user enters the target game scene, various types of new pollutant icons and various virtual collection devices are randomly generated. This random generation mechanism can simulate the diversity and complexity of new pollutants in the real world and improve the repeatability and fun of the game. In response to the user dragging the selected new pollutant icon to the virtual collection device, it is judged in real time whether the collection device can collect the new pollutant icon. This real-time interaction and feedback mechanism can not only accurately judge whether the user classification is accurate, but also enhance the user's sense of participation and learning effect. In addition, by determining whether the user's classification operation meets the requirements and completing the task setting when all classifications are correct, it can effectively guide users to learn the classification knowledge of new pollutants, thereby improving the user's learning interest and enthusiasm through gamification, and improving the classification education effect of new pollutants and environmental health protection awareness. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a flow chart of a method for classifying new pollutants in a game scene according to an embodiment of the present application; Figure 2 A schematic diagram of a game interface for a method for classifying new pollutants in a game scene according to an embodiment of the present application; Figure 3 A schematic diagram of a game interface for a method for classifying new pollutants in a game scene according to another embodiment of the present application; Figure 4 This is a schematic block diagram of the structure of a new pollutant classification device in a game scene according to an embodiment of the present application; Figure 5Structural schematic block diagram of a popular science device according to an embodiment of the present application.

[0018] The realization of the purpose of the present application, functional features and advantages will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0020] It should be noted that currently, new pollutants mainly include persistent organic pollutants, endocrine disruptors, antibiotics and microplastics. Among them, Persistent Organic Pollutants (POPs) refer to a class of organic chemical substances with long-term persistence, bioaccumulation, semi-volatility and high toxicity. They can persist in the environment for a long time, migrate long distances through media such as the atmosphere, water and organisms, and cause serious impacts on human health and the environment. The hazards to human health and the environment are mainly reflected in neurological toxicity, immunotoxicity, reproductive and developmental toxicity, and carcinogenicity. Their common sources include pesticides in agriculture: insecticides such as DDT, chlordane, mirex, etc.; industrial chemicals: polychlorinated biphenyls (PCBs), hexachlorobenzene, decabromodiphenyl ether, short-chain chlorinated paraffins, etc.; unintentionally produced industrial by-products: in some industrial production processes, such as waste incineration, chemical production, metal smelting, etc., POPs such as dioxins, furans, hexachlorobenzene will be accidentally produced. In addition, the use of chlorine-containing compounds, such as chlorophenols, PCBs, chlorinated phenyl ether pesticides and binapacryl.

[0021] Endocrine Disrupting Chemicals (EDCs), also known as Environmental Hormones, are a class of chemical substances that can interfere with the normal functions of the endocrine systems of organisms. They enter the body through channels such as food and drinking water. Although they do not directly cause acute hazards like toxic substances, they have effects similar to hormones, can interfere with normal endocrine functions, and thus have adverse effects on health. Such substances can cause damage to the reproductive systems of animals and humans, neurodevelopmental disorders, decreased immune system function, metabolic disorders, cancer risks, etc. They mainly come from plastic products, personal care products and cosmetics, household products, pesticides and herbicides, industrial pollution, food packaging and processing, etc.

[0022] Antibiotics are secondary metabolites produced by bacteria, molds, or other microorganisms, or synthetic analogs. They are widely used in the treatment of infectious diseases in humans and animals, and are also extensively used in aquaculture and livestock farming. Antibiotics cannot be fully absorbed in the human and animal bodies, and most of them will enter sewage treatment plants or directly enter the environment in the form of the original or active metabolites with excreta. Although the half-lives of most antibiotics are short, due to their frequent use and entry into the environment, a "pseudo-persistence" phenomenon has been formed, which has gradually become a new pollutant that all environmentalists are concerned about, posing potential ecological risks to the ecological environment and human health.

[0023] Microplastics are tiny particles formed by the decomposition of plastic products under the influence of environmental and human factors. During the use of plastic products, it is inevitable to cause the generation and release of microplastics. Their diameters range from 1 micrometer to 5 millimeters, which are very small, but they are numerous and widely distributed. Research shows that microplastics exist in tap water, bottled water, the atmosphere, soil, plants, and animals, especially in marine organisms, and they are also present in human tissues such as blood and lungs. Microplastics are not only difficult to decompose themselves but also easily adsorb other toxic pollutants, further exacerbating environmental and health risks.

[0024] In fact, the sources of new pollutants are the same as those of conventional pollutants, originating from industrial production, daily life, and agricultural activities. The national list of new pollutants has listed 14 new pollutants, including PFOS (Perfluorooctane Sulfonates), PFOA (Perfluorooctanoic Acid), PFHxS (Perfluorohexane Sulfonic Acid), short-chain chlorinated paraffins, dichloromethane, chloroform, nonylphenol, antibiotics, and the phased-out chlordane, etc.

[0025] In addition, new pollutants also have four major characteristics: 1. Concealment: The risks of most new pollutants are long-term and concealed, and their short-term hazards are not obvious. Once their hazards are discovered, new pollutants may have entered the environment or even the human body through various channels.

[0026] 2. Persistence: New pollutants have high stability, are difficult to degrade in the environment, and are easily enriched in the ecosystem. They can be accumulated in the environment and organisms for a long time, and can migrate over long distances through media such as the atmosphere, water, and soil, or spread along the food chain.

[0027] 3. High harm: Many emerging pollutants have various biological toxicities such as organ toxicity, neurotoxicity, reproductive and developmental toxicity, immunotoxicity, endocrine disruption effects, carcinogenicity, and teratogenicity, posing potential hazards to the ecological environment and human health. For example, long-term abuse of antibiotics may lead to resistance gene pollution, and even result in some diseases being incurable. Unqualified water bottles containing bisphenol A pose a threat to the health of fetuses and children.

[0028] 4. Difficult to treat: Some emerging pollutants are newly synthesized substances by humans, with excellent product characteristics, and it is difficult to develop their substitutes and alternative technologies. Some emerging pollutants are widely used by humans, involving a wide range of industries and a long industrial chain. Emerging pollutants have low contents and are dispersed in the environment, making it difficult to control their production, use, and pollution treatment. In addition, the research foundation on the migration, transformation, and natural purification laws of emerging pollutants is weak, all of which lead to emerging pollutants being difficult to treat.

[0029] Therefore, in order to enable people to have a deeper understanding of emerging pollutants, this application proposes a classification method for emerging pollutants in a game scenario, with the execution entity being a science popularization device. The science popularization device can be any type of mobile or stationary computer device, including mobile computers or stationary computer devices such as desktop computers or PCs. The science popularization device can also be terminal devices such as mobile phones, tablets, and game consoles. Among them, the components of the computer device include but are not limited to a memory and a processor, the processor is connected to the memory through a bus, and the database is used to store data.

[0030] Reference Figure 1 As shown, in one embodiment, this application provides a classification method for emerging pollutants in a game scenario, and the method includes: S100. When the user enters the target game scenario, randomly generate various types of emerging pollutant icons and various virtual collection devices; S200. In response to the user's drag operation of dragging the first emerging pollutant icon selected from the target game scenario to the first virtual collection device, determine whether the first virtual collection device can collect the first emerging pollutant icon; S300. If so, determine that the user's classification of the first emerging pollutant icon meets the requirements; S400. When it is determined that the classification of all emerging pollutant icons meets the requirements, determine that the user has completed the classification task of emerging pollutant icons.

[0031] When the user enters the target game scene preset by the system, the system will automatically and randomly generate various types of new pollutant icons and various virtual collection devices. Among them, the number of new pollutant icons of each type can be the same or different. Each new pollutant icon represents a type of new pollutant. For example, persistent organic pollutant icons, endocrine disruptor icons, antibiotic icons, and microplastic icons. The virtual collection devices are used to collect new pollutants, and each virtual collection device is used to collect specified and identical types of new pollutants. The arrangement order of the icons is random and distributed in multiple layers, increasing the challenge and interest of the game and preventing the user from easily completing the classification task.

[0032] For example, referring to Figure 2 As shown, in this target game scene, the system randomly generates 3 types of new pollutant icons, and the number of these new pollutant icons is 5 (new pollutant icons 1, 2, 3, 4, 5). Among them, new pollutant icon 1 and new pollutant icon 4 are the same type of the first new pollutant icon, new pollutant icon 3 and new pollutant icon 5 are the same type of the second new pollutant icon, and new pollutant icon 2 is the third new pollutant icon. At the same time, 4 virtual collection devices (virtual collection devices 1, 2, 3, 4) are generated.

[0033] When the user selects the first new pollutant icon from the target game scene and drags it to the first virtual collection device, the system will respond to this drag operation of the user and determine whether the first virtual collection device can collect the first new pollutant icon. Among them, the first new pollutant icon is the new pollutant icon selected and dragged by the user, and the first virtual collection device is the virtual collection device to which the first new pollutant icon is dragged. The judgment process involves the preset rules for the association relationship between the new pollutant icon and the virtual collection device. For example, the system has predefined that a certain virtual collection device can only collect specific types of new pollutants. When the user drags the new pollutant icon of this type to this device, the system determines it as collectable; otherwise, it is determined as non-collectable.

[0034] For example, referring to Figure 3 As shown, when the user drags the selected new pollutant icon 3 to the virtual collection device 1, the system will respond to this drag operation of the user and determine whether the virtual collection device 1 can collect the new pollutant icon 3.

[0035] If the system determines that the first virtual collection device can collect the first new pollutant icon, then it is considered that the user's classification of the first new pollutant icon meets the requirements. This step is to judge whether the user's operation is correct. If it is correct, it indicates that the user has successfully completed the classification of this new pollutant icon, indicating that the user has a correct understanding of this type of new pollutant and knows which collection device it should be classified into.

[0036] When the system determines that the classification of all new pollutant icons meets the requirements, that is, the user has dragged all new pollutant icons to the corresponding virtual collection device according to the correct rules, then the user is judged to have completed the classification task of the new pollutant icons, indicating that the user has a comprehensive and correct grasp of the classification knowledge of the new pollutant and can successfully pass this game scene. By integrating the classification knowledge of new pollutants into the game scene, users can learn relevant knowledge about new pollutants in a vivid and interesting way during the game process, improve their cognition of new pollutants and environmental awareness, and make users pay more attention to environmental protection issues in real life. At the same time, compared with the traditional knowledge explanation method, this game-based interactive learning method can allow users to participate in learning more actively, deepen their memory of the classification knowledge of new pollutants, and enable users to remember these important environmental protection knowledge for a longer time.

[0037] The present application provides a method for classifying new pollutants in a game scene. When a user enters a target game scene, various types of new pollutant icons and various virtual collection devices are randomly generated. This random generation mechanism can simulate the diversity and complexity of new pollutants in the real world, and improve the repeatability and fun of the game. In response to the user dragging the selected new pollutant icon to the virtual collection device, it is judged in real time whether the collection device can collect the new pollutant icon. This real-time interaction and feedback mechanism can not only accurately judge whether the user's classification is accurate, but also enhance the user's sense of participation and learning effect. In addition, by determining whether the user's classification operation meets the requirements, and completing the task setting when all classifications are correct, it can effectively guide users to learn the classification knowledge of new pollutants, thereby improving the user's learning interest and enthusiasm through gamification, and improving the new pollutant classification education effect and environmental health protection awareness.

[0038] In one embodiment, the randomly generated multiple types of new pollutant icons include: Identify all types of new pollutants and obtain a realistic picture of each type of new pollutant; Digitally processing the real image to obtain a standard real image; According to the standard real image, characteristic elements of each type of new pollutant are extracted, and a virtual icon containing the corresponding characteristic elements is generated for each type of new pollutant to obtain multiple types of new pollutant icons. The characteristic elements are the elements in the standard real image that best represent the characteristics of the new pollutant.

[0039] For each new pollutant type, obtain its real image. This real image can be an actually taken image, such as taking a partial picture of an electronic product containing persistent organic pollutants, or taking a picture of a medicine bottle containing antibiotic tablets, etc. By obtaining the real image, it provides the basic material for digital processing and feature extraction.

[0040] Perform a series of digital processing operations on the obtained real image, including image cropping, scaling, denoising, color correction, etc., to eliminate the interference factors existing in the image, making the image clearer and more standardized for feature analysis. For example, for an image of persistent organic pollutants taken in a complex environment, crop out the irrelevant objects in the background and adjust the brightness and contrast of the image to make the pollutants themselves more prominently displayed, thus obtaining a clear and standard real image that meets the processing requirements.

[0041] Based on the standard real image, extract the feature elements that can most significantly represent the new pollutant type. Taking persistent organic pollutants as an example, if they are manifested as particulate matter of a certain specific shape or a special texture pattern in the standard real image, then these shapes and textures are the feature elements. For antibiotics, the shape, color of its tablets, or the text style on the packaging, etc. are the feature elements. After extracting the feature elements, use an image generation algorithm to generate virtual icons for each type of new pollutant according to the feature elements. The generated virtual icons can intuitively reflect the typical characteristics of the new pollutant, enabling users to quickly identify the type of new pollutant represented by the icon's appearance during the game process.

[0042] For example, assume that a virtual icon is to be generated for persistent organic pollutants. First, determine its real image by taking a picture of the internal parts of an electronic product containing the flame retardant to obtain an image containing the flame retardant. Then perform digital processing on this image, crop out the redundant background, adjust the image to an appropriate size, and correct the color to make the image clearer, obtaining a standard real image. Next, extract the feature elements from the standard real image and find that the flame retardant is manifested as a regular granular distribution on the surface of the electronic parts, the granules are light yellow and have fine textures. Generate a virtual icon based on this feature element, with a pattern of light yellow regular granules shown on the icon and a background simulating the surface texture of the electronic parts around it. The finally obtained virtual icon represents the persistent organic pollutants and can be used for classification tasks in the game scenario.

[0043] In this embodiment, by extracting feature elements from real images to generate virtual icons, the virtual icons are highly correlated with the appearance characteristics of actual new pollutants, can more accurately reflect the characteristics of new pollutants, enable the images that users come into contact with in the game to have a strong connection with new pollutants in reality, and enhance users' intuitive understanding of new pollutants. The generated virtual icons can highlight the core characteristics of each new pollutant, help users quickly and accurately identify different types of new pollutant icons, thereby improving the efficiency of classification tasks during the game and making it easier for users to understand and master the knowledge of new pollutant classification when learning. In addition, the icons generated based on real images are more scientific and accurate, avoiding misleading users' understanding of new pollutants due to unreasonable icon design.

[0044] In one embodiment, determining all types of new pollutants includes: Obtaining the attribute information of all new pollutants; Respectively extracting the attribute names and pollution damage values from the attribute information of each new pollutant; Using a pre-trained neural network model to analyze the attribute names and pollution damage values of each new pollutant to obtain the type of each new pollutant.

[0045] In this embodiment, various data related to the attributes of all new pollutants are collected. The attribute information may include the chemical properties of new pollutants (such as molecular structure, chemical stability, etc.), physical properties (such as form, solubility, etc.), sources (such as industrial production processes, additives in daily necessities, etc.), distribution in the environment (such as existence forms in water, soil, air), toxic effects on organisms (such as types of diseases caused, biological systems affected, etc.), and their pollution damage values, etc. This attribute information can be obtained from channels such as research literature related to environmental science, reports of environmental monitoring agencies, chemical substance databases, international environmental protection convention documents (such as relevant materials of the Stockholm Convention), and data released by specialized scientific research institutions. For example, for persistent organic pollutants, their attribute information will include their characteristics of long-term existence and difficulty in decomposition in the environment, and their toxic characteristics of affecting human health through food chain enrichment; for endocrine disruptors, the specific mechanisms of disrupting the biological endocrine system and their extensive distribution in daily items will be recorded. From the rich attribute information collected, two key parts are screened out: the attribute name and the pollution damage value. The attribute name is the specific description name of various characteristics of the new pollutant. For example, for a certain persistent organic pollutant, its attribute names can be "chemical stability", "bioaccumulation", "carcinogenicity", etc.; for an endocrine disruptor, the attribute names can be "endocrine disruption", "reproductive toxicity", "developmental toxicity", etc. The pollution damage value is a numerical index used to quantify the degree of harm caused by the new pollutant to the environment or organisms, and the value can be obtained through comprehensive calculations of a series of complex environmental risk assessment models, toxicological experimental data, etc. For example, based on data such as the decline in the number of biological populations and the degradation of ecosystem service functions caused by a new pollutant in a certain area, its pollution damage value is calculated to be 8.5 (assuming a full score of 10, the larger the value, the more serious the harm).

[0046] Use a neural network model that has been trained with a large amount of data in advance to analyze the attribute names and pollution damage values of each new pollutant extracted. This neural network model is an artificial intelligence algorithm model that learns based on a large amount of input-output data and can discover complex relationships and patterns in the data. In this process, the attribute names of the new pollutant and the corresponding pollution damage values are used as input data and input into the neural network model. Inside the model, complex operations and processing are performed on the input data through a multi-layer neuron structure. According to the knowledge and rules learned during the previous training, it outputs the type to which the new pollutant belongs, such as persistent organic pollutants, endocrine disruptors, antibiotics, or microplastics, etc. This classification process is based on the neural network's learning of the attribute names and pollution damage values of a large number of known types of new pollutants, so as to accurately judge and classify new and unknown types of new pollutants.

[0047] This embodiment can use a neural network model to judge the type of new pollutants, can comprehensively consider various complex attributes and pollution damage situations, can accurately determine the type of new pollutants, and avoid classification errors caused by human factors or one-sided consideration of certain characteristics. At the same time, in the face of numerous new pollutants and their complex attribute information, the neural network model can quickly process this data and give classification results, greatly improving the efficiency of new pollutant classification. Especially in an environment where new pollutants are constantly emerging and need to be managed and controlled in a timely manner, this efficient classification ability is crucial and helps to promptly grasp the type distribution of new pollutants.

[0048] Preferably, randomly generate various types of new pollutant icons, including: Obtain the real image of each type of emerging pollutant. For the emerging pollutant icon of each type of emerging pollutant, perform edge detection on the real image to obtain a sequence of edge points. In the sequence of edge points, randomly select a continuous segment of edge points, calculate the angle change rate between every three consecutive edge points, and classify the edges into high-curvature edges and low-curvature edges according to the magnitude of the angle change rate; Depict the concave-convex shape of the high-curvature edges according to the spline curve interpolation method, connect the low-curvature edges with a smooth curve, and connect the contours of all edge parts to obtain the emerging pollutant contour; Determine the color interval that best represents the emerging pollutant in the real image according to the color histogram analysis method, obtain the main color and color distribution, and extract the texture features of the surface of the emerging pollutant from the real image; Render the corresponding emerging pollutant contour according to the main color, color distribution, and texture features of each type of emerging pollutant to obtain emerging pollutant icons of multiple types.

[0049] First, obtain the real image of each emerging pollutant type as the basic material. The real image can be obtained by shooting or from a relevant database. Then perform edge detection on the real image to find the boundaries of the objects in the image and obtain a sequence of edge points composed of many edge points. For example, for a real image of microplastics, its granular boundary contour can be obtained after edge detection.

[0050] Next, randomly select a continuous segment of edge points from the sequence of edge points to introduce a certain degree of randomness and diversity in subsequent analysis. Then, calculate the angle change rate between every three consecutive edge points. The angle change rate reflects the degree of bending of the edge in a local area. For example, three consecutive edge points form a broken line, and calculate the angle change rate of its bending. According to the magnitude of the angle change rate, the edges are classified into high-curvature edges (a large angle change rate indicates obvious edge bending) and low-curvature edges (a small angle change rate, the edge is relatively straight). For example, the edges at the relatively sharp corners of microplastic particles will be classified into high-curvature edges, while the relatively smooth side edges belong to low-curvature edges.

[0051] Among them, the spline curve interpolation method is used to generate a smooth curve between known points. For high-curvature edges, due to their obvious concave and convex changes, the spline curve interpolation method can depict the concave and convex shape according to the selected edge points, so that the generated contour can accurately reflect the true shape of the new pollutant in the high-curvature area. For example, for the pits or protrusions on the surface of microplastic particles, the spline curve interpolation method is used to depict their shapes. For low-curvature edges, smooth curves are used for connection to ensure a smooth transition of the edges, avoid the appearance of rigid broken lines, and make the entire contour more natural and coherent. Finally, the contours of all edge parts (including the parts after processing high-curvature and low-curvature edges) are connected to form a complete contour of the new pollutant. For example, the concave and convex parts and the smooth side edges of the microplastic particle are connected to obtain a complete contour of the microplastic particle.

[0052] Among them, the color histogram analysis method is used to count the distribution of different colors in the image. By performing color histogram analysis on the real image, the color intervals with higher frequencies and most representative of the new pollutant are found, so as to determine the main color tone and color distribution. For example, for a real image containing persistent organic pollutants, if it mainly presents a light yellow tone, then the color histogram analysis can determine the light yellow interval as the main color tone interval and also see the distribution of this color in the entire image. In addition, the texture features on the surface of the new pollutant are extracted from the real image, and the texture features can reflect information such as the roughness and texture direction of the object surface. For example, the surface of microplastics may show a certain granular texture, and persistent organic pollutants may have fine fibrous textures, etc.

[0053] Apply the main color, color distribution, and texture features of each type of new pollutant to the previously generated outline of the new pollutant for rendering. Rendering is a process of filling two-dimensional outlines with visual elements such as colors and textures, making the outlines have richer visual information and be closer to the actual appearance of the new pollutant. For example, according to the main color (such as light blue) and color distribution of microplastics, fill the corresponding light blue inside its outline and draw it according to its texture features (granular texture). Finally, a realistic microplastic icon is obtained. Thus, by extracting and processing from multiple aspects such as the edges, colors, and textures of real images, the generated new pollutant icons can truly reflect the actual appearance characteristics of the new pollutants, making the icons that users see in the game scene just like seeing the real new pollutants, enhancing the visual realism and cognitive effects. At the same time, due to the icons having real colors, textures, and outline features, there are obvious visual differences between different types of new pollutant icons, facilitating users to quickly and accurately identify and classify them, improving the playability of the game and the effectiveness of knowledge dissemination. In addition, realistic icons can enable users to more deeply perceive the existence and characteristics of new pollutants, thereby deepening the understanding of the harms of new pollutants, enhancing environmental awareness, and increasing the attention to the treatment of new pollutants.

[0054] In one embodiment, randomly generate multiple types of new pollutant icons, including: Determine the current game level of the user and randomly generate multiple types of new pollutant icons that match the level of difficulty of the game level; Add a label reflecting the level of difficulty to the new pollutant icons.

[0055] In the game system, it will track and record the game progress of the user in real time, that is, the current game level. Each level has a preset difficulty level. Generally, the higher the level, the greater the difficulty. For example, the game can include multiple difficulty levels such as easy, medium, and hard, corresponding to different level ranges respectively. By determining the user's current level, it is possible to understand the current operation level and knowledge mastery degree of the user, so as to prepare for generating new pollutant icons that match them subsequently, ensuring that the challenge and playability of the game are in line with the actual situation of the user.

[0056] According to the difficulty level of the current level, a certain number and type of new pollutant icons are randomly selected from a pre-designed and stored set of multiple new pollutant icons. In levels with lower difficulty, fewer new pollutant icons with relatively simple types and more obvious characteristics will be generated. For example, only two types of microplastics and antibiotics are included, with 3 icons of each type. The icons have large differences in appearance and are easy to distinguish. In levels with higher difficulty, a larger number of new pollutant icons with richer types and relatively complex characteristics will be generated. For example, four types of persistent organic pollutants, endocrine disruptors, antibiotics and microplastics are included at the same time, with 5 icons of each type, and the color, shape and other characteristics of the icons will have certain similarities, increasing the difficulty of classification, so that users can constantly challenge themselves in the process of gradually increasing difficulty and improve their knowledge of new pollutant classification.

[0057] In addition, in order to enable users to intuitively understand the level difficulty corresponding to the current new pollutant icon, and to facilitate the system's management and subsequent processing of the icon, a special label information will be added to the generated new pollutant icon. The label stores the level difficulty level reflected by the icon, such as text labels such as "easy", "medium", and "difficult", or digital labels (such as 1 for easy, 2 for medium, and 3 for difficult). During the game, the system can adjust the evaluation criteria, prompt information, and other content for user operations based on the label. At the same time, users can also reasonably arrange their own operation strategies and learning focuses based on the label, which helps users better adapt to game challenges at different difficulties.

[0058] In this embodiment, as the level is upgraded, the number, type and feature complexity of the new pollutant icons gradually increase, which can continuously maintain the fun and challenge of the game, avoid the game process being too monotonous and boring, stimulate the user's interest in constantly challenging higher difficulty levels, and allow the user to gradually improve the mastery of new pollutant classification knowledge during the game. At the same time, the corresponding icon is generated according to the user's current level difficulty, providing users with a personalized learning experience, so that users of different levels can play and learn at a difficulty level that suits them, which helps users better understand and master the new pollutant classification knowledge and improve learning effects. In addition, by adding level difficulty labels to icons, users can clearly understand their current learning stage and the goals they need to achieve, as well as their mastery of new pollutant classification knowledge of different difficulty levels.

[0059] In one embodiment, after determining that the classification of the first new pollutant icon by the user meets the requirements, the method further includes: When it is determined that the classification of at least one new pollutant icon does not meet the requirements, the new pollutant icon whose classification does not meet the requirements is used as the second new pollutant icon; For each second new pollutant icon, determine the key information of the new pollutant corresponding to the second new pollutant icon, and search for the popular science video of the new pollutant corresponding to the second new pollutant icon from the video library according to the key information; After screening the popular science videos, obtain the candidate popular science videos corresponding to each second new pollutant icon; After cropping and re - splicing all the candidate popular science videos, obtain the target popular science video; Display the target popular science video in the target game scene.

[0060] During the game, when it is determined that the user's classification operation of a certain new pollutant icon is incorrect (for example, the user wrongly puts the micro - plastic icon into the virtual collection device of persistent organic pollutants), then the new pollutant icon that does not meet the requirements will be marked as the second new pollutant icon.

[0061] According to the type of the misclassified new pollutant icon (such as micro - plastics, persistent organic pollutants, etc.), extract the key information of the pollutant (such as pollution sources, hazard characteristics, treatment measures, etc.). Then, use the key information to search for relevant popular science videos in the pre - established video library. For example, for micro - plastics, popular science videos about the sources of micro - plastics (such as the decomposition of plastic products) and hazards (such as marine organisms ingesting them) will be searched.

[0062] To ensure the quality and relevance of the popular science videos, the searched popular science videos will also be screened. The screening criteria include the authority of the video (such as whether it is produced by a professional environmental protection agency), clarity, duration, etc. Through screening, several candidate popular science videos that meet the requirements will be selected for each second new pollutant icon.

[0063] Then, according to the educational goals and user needs, crop the screened candidate popular science videos, remove irrelevant content (such as advertisements, lengthy introductions, etc.), and re - splice the key information segments about the same new pollutant in different videos. For example, splice the segments of the sources, hazards, and treatment measures of micro - plastics into a complete popular science video, so as to generate the target popular science video for this new pollutant.

[0064] Finally, embed the generated target popular science video into the game scene for users to watch, so as to help users more intuitively understand the relevant knowledge of new pollutants through vivid video content, correct the previous classification errors, and enhance users' environmental awareness at the same time.

[0065] In this embodiment, through the intuitive display of popular science videos, users can more clearly understand the relevant knowledge of new pollutants. Especially for those pollutants with incorrect classifications, they can specifically learn their characteristics and hazards, thereby improving the accuracy of classification and environmental protection awareness. At the same time, by embedding the popular science videos into the game scenarios, users actively watch the videos during the game process. This interactive learning method can effectively improve the enthusiasm and effect of learning, and avoid the dullness of traditional education methods. In addition, the system can timely detect the classification errors of users and provide correct knowledge through popular science videos, helping users continuously improve and enhance during the learning process, and avoiding the accumulation of incorrect knowledge.

[0066] Preferably, determining whether the first virtual collection device can collect the first new pollutant icon includes: Determine all the characteristic information of the new pollutant icons that the first virtual collection device can collect, and obtain a plurality of first characteristic information; Obtain all the characteristic information of the first new pollutant icon, and obtain a plurality of second characteristic information; Respectively perform vectorization processing on the plurality of first characteristic information and the plurality of second characteristic information to obtain a plurality of first characteristic vectors and a plurality of second characteristic vectors; Perform clustering processing on the plurality of first characteristic vectors and the plurality of second characteristic vectors respectively, and screen to obtain a first target characteristic vector and a second target characteristic vector. The clustering processing includes: Randomly select a vector from all the characteristic vectors as the initial clustering center, calculate the Euclidean distance between all the characteristic vectors and the initial clustering center, allocate all the characteristic vectors to different clusters according to the Euclidean distance, and calculate the average vector of the plurality of characteristic vectors as the new clustering center. Replace the initial clustering center with the new clustering center, and return and repeat the step of calculating the Euclidean distance between all the characteristic vectors and the initial clustering center until the change of the new clustering center is less than a preset threshold, and use the new clustering center as the first target characteristic vector or the second target characteristic vector; Calculate the cosine distance between the first target characteristic vector and the second target characteristic vector. When the cosine distance is greater than the preset cosine distance, it is determined that the first virtual collection device can collect the first new pollutant icon.

[0067] In the game system, each virtual collection device is preset with specific types of new pollutant icons that it can collect. The virtual collection device can identify specific feature information, such as shape, color, texture, etc. For the first virtual collection device, all the feature information that it can match is extracted, and this feature information is called the first feature information. For example, if the first virtual collection device is used to collect persistent organic pollutant icons, then its first feature information may include characteristic descriptions such as the common color of the pollutant icon (such as light yellow), shape (such as granular), texture (such as smooth or slightly cracked), etc.

[0068] For the first new pollutant icon dragged by the user into the collection device, all its feature information is also extracted, and this feature information is called the second feature information. For example, if the icon dragged by the user is a microplastic icon, the system will extract its feature information, such as the light blue color commonly presented by the microplastic icon, irregular granular shape, and possible fine texture on the surface, etc.

[0069] Feature vectorization processing is an operation that converts feature information into numerical vectors for mathematical operations and analysis. The system will convert the first feature information and the second feature information into vector forms respectively. For example, features such as color, shape, and texture are converted into numerical vectors through specific encoding methods (such as using RGB values for color, edge feature vectors for shape, and texture pattern vectors for texture). In this way, the feature information of the first virtual collection device generates multiple first feature vectors, and the first new pollutant icon generates multiple second feature vectors.

[0070] Among them, clustering processing is used to group similar data points into the same cluster. In this process, the system will perform clustering on the first feature vectors and the second feature vectors respectively, and extract the most representative feature vectors (i.e., target feature vectors). The specific process is as follows: Randomly select a vector from all the feature vectors as the initial clustering center, calculate the Euclidean distance between all the feature vectors and the initial clustering center, and assign the feature vectors to different clusters according to the distance. The smaller the Euclidean distance, the more similar the feature vectors are. Calculate the average vector of all the feature vectors in each cluster and use it as the new clustering center. Replace the initial clustering center with the new clustering center, and repeat the above steps until the change in the new clustering center is less than the preset threshold (i.e., the clustering center is stable), and take the final clustering center as the target feature vector. In this way, the feature vectors of the first virtual collection device obtain the first target feature vector after clustering processing, and the feature vectors of the first new pollutant icon obtain the second target feature vector after clustering processing.

[0071] Among them, the cosine distance is used to measure the similarity between two vectors. The closer the value is to 1, the higher the similarity. The system calculates the cosine distance between the first target feature vector and the second target feature vector. If the cosine distance is greater than a preset threshold (for example, 0.85), it indicates that these two target feature vectors are highly similar, that is, the first new pollutant icon matches the features of the first virtual collection device, and it is determined that the first virtual collection device can collect this icon; if the cosine distance is less than or equal to the preset threshold, it is determined that this icon cannot be collected.

[0072] In this embodiment, through the vectorization and clustering processing of feature information, the system can accurately identify and match the features of the icon and the collection device, so as to more efficiently and accurately judge whether the classification is correct.

[0073] In one embodiment, after determining whether the first virtual collection device can collect the first new pollutant icon, it further includes: If not, determine the pollution type and pollution damage value of the new pollutant corresponding to the first new pollutant icon; According to the position of the first new pollutant icon in the target game scene, determine the rendering center point of the pollution special effect; Select a matching special effect object according to the pollution type of the new pollutant corresponding to the first new pollutant icon, and set the basic attributes of the special effect object; Dynamically adjust the state of the special effect object according to the pollution damage value, and use the rendering system of the game engine to draw the dynamically adjusted special effect object in a semi-transparent form to the rendering center point in the target game scene.

[0074] If the system determines that the first virtual collection device cannot collect the first new pollutant icon (for example, the user drags the new pollutant icon to the wrong virtual collection device), it is necessary to obtain detailed information about this new pollutant icon for subsequent processing. Specifically, it is necessary to determine the type of this new pollutant (such as microplastics, persistent organic pollutants, etc.) and its corresponding pollution damage value (calculated through a preset evaluation model). The pollution damage value is used to reflect the potential harm degree of this pollutant to the environment or organisms.

[0075] In the game scene, each new pollutant icon has a specific position coordinate. The system will determine a corresponding rendering center point of the "pollution special effect" according to the position of this icon. The rendering center point is the starting point of the special effect animation or visual effect, usually located near the position of the new pollutant icon or at the icon center, and is used to generate subsequent visual effects (such as pollution diffusion, color change, etc.).

[0076] Select a special effect object that matches the type of pollution of the emerging pollutant. For example, for microplastics, the "ripple diffusion" special effect can be selected; for persistent organic pollutants, the "smoky" special effect can be selected. The basic attributes of the special effect object include color, shape, size, diffusion speed, etc., and these basic attributes can be set according to the type and characteristics of the pollutant. For example, the special effect object of microplastics can be light blue translucent particles with a slow diffusion speed; while the special effect object of persistent organic pollutants may be dark yellow smoke with a slightly faster diffusion speed.

[0077] The pollution damage value is not only used to represent the harm degree of the pollutant, but also can be intuitively reflected by dynamically adjusting the state of the special effect object. For example, if the pollution damage value is high (such as 8 or 9, with a full score of 10), the color of the special effect object will be more vivid, the diffusion speed will be faster, and the range will be larger; if the pollution damage value is low (such as 3 or 4), the color of the special effect object will be lighter, the diffusion speed will be slower, and the range will be smaller. This dynamic adjustment allows users to intuitively feel the harm degree of different pollutants.

[0078] Finally, using the rendering system of the game engine, draw the dynamically adjusted special effect object to the rendering center point in the target game scene. The special effect object is usually presented in a semi-transparent form, so that it will not completely block other elements in the game scene, and can attract the user's attention through visual special effects, reminding the user of the harmfulness of the emerging pollutant represented by the current emerging pollutant icon.

[0079] This embodiment can visually display the type and harm degree of emerging pollutants through visual special effects. Users can more intuitively understand the possible consequences of misclassification, and thus more deeply recognize the danger of emerging pollutants. This dynamic visual feedback can not only remind users of the errors in the current operation, but also help users vividly understand the harm degree of different pollutants through the dynamic changes of the special effects, enhancing the educational effect. In addition, the use of semi-transparent special effects does not interfere with the normal progress of the game, and at the same time adds visual effects to the game, improving the overall game experience and immersion.

[0080] Refer to Figure 4 , in the embodiment of the present application, a classification device for emerging pollutants in a game scene is further provided, and the device includes: A generation module 100, configured to randomly generate multiple types of emerging pollutant icons and multiple virtual collection devices when the user enters the target game scene; A judgment module 200, configured to, in response to a drag operation in which the user drags a first emerging pollutant icon selected from the target game scene to a first virtual collection device, judge whether the first virtual collection device can collect the first emerging pollutant icon; A determination module 300, configured to, if so, determine that the user's classification of the first emerging pollutant icon meets the requirements; A determination module 400 is configured to determine that the user has completed the classification task of new pollutant icons when it is determined that the classification of all new pollutant icons meets the requirements.

[0081] As described above, it can be understood that each component of the classification device for new pollutants in a game scenario proposed in this application can implement the functions of any of the classification methods for new pollutants in a game scenario described above, and the specific structure will not be elaborated.

[0082] Referring to Figure 5 , this application embodiment also provides a popular science device, whose internal structure can be as Figure 5 shown. The popular science device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor designed for this popular science device is used to provide computing and control capabilities. The memory of this popular science device includes a storage medium and an internal memory. The storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the storage medium. The database of this popular science device is used to store relevant data of the classification method for new pollutants in a game scenario. The network interface of this popular science device is used to communicate with an external popular science device through a network connection. The computer program, when executed by the processor, can implement the classification method for new pollutants in a game scenario.

[0083] This application embodiment also provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the classification method for new pollutants in a game scenario.

[0084] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or further includes elements inherent to this process, device, article or method. Without more limitations, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, device, article or method including this element.

[0085] The above are only the preferred embodiments of this application, and do not limit the patent scope of this application. Any equivalent structure or equivalent process transformation made using the specification and drawings of this application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of this application.

Claims

1. A classification method for new pollutants in a game scenario, characterized in that, Including: When the user enters the target game scene, a variety of new pollutant icons and a variety of virtual collection devices are randomly generated; In response to the user's drag operation of dragging the first new pollutant icon selected from the target game scene to the first virtual collection device, determine whether the first virtual collection device can collect the first new pollutant icon; If so, determine that the user's classification of the first new pollutant icon meets the requirements; When it is determined that the classification of all new pollutant icons meets the requirements, determine that the user has completed the classification task of the new pollutant icons.

2. The method according to claim 1, wherein The random generation of a variety of new pollutant icons includes: Determine all types of new pollutants, and obtain the real images of each type of new pollutant; Perform digital processing on the real image to obtain a standard real image; According to the standard real image, extract the characteristic elements of each type of new pollutant, and generate virtual icons containing the corresponding characteristic elements for each type of new pollutant, obtaining a variety of new pollutant icons, where the characteristic elements are the elements that best represent the characteristics of the new pollutant in the standard real image.

3. The method according to claim 2, wherein The determination of all types of new pollutants includes: Obtain the attribute information of all new pollutants; Extract the attribute name and pollution damage value from the attribute information of each new pollutant respectively; Use a pre-trained neural network model to analyze the attribute name and pollution damage value of each new pollutant to obtain the type of each new pollutant.

4. The method according to claim 1, characterized in that, The random generation of a variety of new pollutant icons includes: Obtain the real image of each type of new pollutant. For the new pollutant icon of each type of new pollutant, perform edge detection on the real image to obtain an edge point sequence. In the edge point sequence, randomly select a continuous segment of edge points, calculate the angle change rate between every three consecutive edge points, and divide the edges into high-curvature edges and low-curvature edges according to the magnitude of the angle change rate; Depict the concave and convex shapes of the high-curvature edges according to the spline curve interpolation method, connect the low-curvature edges with a smooth curve, and connect the contours of all edge parts to obtain a new pollutant contour; Determine the color interval that best represents the new pollutant in the real image according to the color histogram analysis method, obtain the main color and color distribution, and extract the texture features of the new pollutant surface from the real image; Render the corresponding new pollutant contour according to the main color, color distribution and texture features of each type of new pollutant to obtain a variety of new pollutant icons.

5. The method according to claim 1, wherein The random generation of a variety of new pollutant icons includes: Determine the current game level of the user, and randomly generate a variety of new pollutant icons that match the level difficulty of the game level; Add a label reflecting the level difficulty to the new pollutant icon.

6. The method according to claim 1, wherein After determining that the user's classification of the first new pollutant icon meets the requirements, it further includes: When it is determined that the classification of at least one new pollutant icon does not meet the requirements, use the new pollutant icon whose classification does not meet the requirements as the second new pollutant icon; For each second new pollutant icon, determine the key information of the new pollutant corresponding to the second new pollutant icon, and search for the popular science video of the new pollutant corresponding to the second new pollutant icon from the video library according to the key information; After screening the popular science videos, obtain the candidate popular science videos corresponding to each second icon of the new pollutant; After cropping and re - splicing all the candidate popular science videos, obtain the target popular science video; Display the target popular science video in the target game scene.

7. The method according to claim 1, characterized in that The judgment of whether the first virtual collection device can collect the first new pollutant icon includes: Determine all the characteristic information of the new pollutant icons that the first virtual collection device can collect to obtain a plurality of first characteristic information; Obtain all the characteristic information of the first new pollutant icon to obtain a plurality of second characteristic information; Respectively perform vectorization processing on the plurality of first characteristic information and the plurality of second characteristic information to obtain a plurality of first characteristic vectors and a plurality of second characteristic vectors; Perform clustering processing on the plurality of first characteristic vectors and the plurality of second characteristic vectors respectively, and screen to obtain a first target characteristic vector and a second target characteristic vector. The clustering processing includes: Randomly select a vector from all the characteristic vectors as the initial clustering center, calculate the Euclidean distance between all the characteristic vectors and the initial clustering center, assign all the characteristic vectors to different clusters according to the Euclidean distance, and calculate the average vector of the plurality of characteristic vectors as the new clustering center. Replace the initial clustering center with the new clustering center, and return and repeat the step of calculating the Euclidean distance between all the characteristic vectors and the initial clustering center until the change of the new clustering center is less than the preset threshold, and use the new clustering center as the first target characteristic vector or the second target characteristic vector; Calculate the cosine distance between the first target characteristic vector and the second target characteristic vector. When the cosine distance is greater than the preset cosine distance, it is determined that the first virtual collection device can collect the first new pollutant icon.

8. The method according to claim 1, wherein After the judgment of whether the first virtual collection device can collect the first new pollutant icon, it further includes: If not, determine the pollution type and pollution damage value of the new pollutant corresponding to the first new pollutant icon; According to the position of the first new pollutant icon in the target game scene, determine the rendering center point of the pollution special effect; Select a matching special effect object according to the pollution type of the new pollutant corresponding to the first new pollutant icon, and set the basic attributes of the special effect object; Dynamically adjust the state of the special effect object according to the pollution damage value, and use the rendering system of the game engine to draw the dynamically adjusted special effect object in a semi - transparent form to the rendering center point in the target game scene.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a method for classifying new pollutants in a game scene according to any one of claims 1 to 8.

10. A popular science device, characterized in that, It includes: A processor; A memory; Wherein, the memory stores a computer program, and when the processor executes the computer program, it implements a method for classifying new pollutants in a game scene according to any one of claims 1 to 8.

Images