Urban Land Cover Classification System Based on Multispectral Remote Sensing Technology

Abstract

This invention discloses an urban land cover classification system based on multispectral remote sensing technology, belonging to the field of spectral spatial data calculation technology. This system acquires land surface temperature data of the target area and sub-regions and calculates their rate of change differences. First, based on a comparison of the absolute value of the difference with a preset threshold, a first-level correction is performed on the initial contribution using a material thermal tuning coefficient. Then, based on the correlation coefficient between the temperature changes of the marked sub-region and adjacent regions, a second-level spatial coupling correction is performed, ultimately generating an accurate urban land surface thermal contribution report. This invention realizes a transformation from static identification to dynamic quantitative assessment of urban thermal environment contributions, significantly improving the accuracy, precision, and practical value of the assessment results. It solves the technical problem of insufficient accuracy in the spatial correlation dynamic assessment and adjustment of urban land surface classification thermal contributions.

Description

Technical Field

[0001] This invention relates to the field of spectral spatial data calculation technology, and more specifically, this application relates to an urban land cover classification system based on multispectral remote sensing technology. Background Technology

[0002] With urbanization, the heterogeneity of land surface materials significantly impacts the urban thermal environment, making accurate quantification of the thermal contribution of different materials crucial for urban planning and thermal management. Multispectral remote sensing technology, due to its advantages of large-scale and periodic observation, has become a core tool for land cover classification. However, existing technologies suffer from the following key shortcomings in dynamically assessing land surface thermal contributions, making it difficult to meet the demands for refined data processing:

[0003] Existing methods mostly output static classification maps based on spectral texture features, failing to incorporate temperature temporal changes into the classification criteria. This makes it impossible to explain dynamic physical mechanisms such as the differences in thermal contribution of similar land features under different environments, resulting in assessment results that are disconnected from actual thermal effects. Furthermore, treating impermeable green spaces and other broad categories as homogeneous objects fails to distinguish the differences in heat capacity and thermal inertia among different materials within the same type of land feature, making it impossible to accurately locate key heat source materials and lacking targeted data support for remediation measures. Analyzing independently on a pixel or object basis fails to consider the spatial coupling effect of heat transfer between adjacent areas through radiation and airflow, leading to one-sided thermal contribution assessment results that cannot reflect the region's share of responsibility in the overall thermal environment.

[0004] In summary, existing technologies have significant shortcomings in the dynamic quantification, material-level subdivision, spatial correlation analysis, and multi-scenario adaptability of urban surface heat contribution. There is an urgent need for a classification method that combines multi-time series temperature change rate and spatial correlation correction to achieve a breakthrough from static identification to dynamic and accurate assessment. There is also a technical problem of insufficient accuracy in the spatial correlation dynamic assessment and adjustment of urban surface heat contribution classification. Summary of the Invention

[0005] To address the aforementioned technical problems, a multispectral remote sensing-based urban land cover classification system is provided. This technical solution resolves the issues raised in the background section.

[0006] In a first aspect, embodiments of this application provide an urban land cover classification system based on multispectral remote sensing technology, characterized by comprising: an acquisition module for acquiring the original surface temperature values ​​and initial contribution of a target area and sub-regions, wherein the initial contribution is used to characterize the degree of thermal contribution of the original surface temperature values ​​to land cover classification, and the sub-regions are obtained by segmenting the target area; a calculation module for calculating a first temperature change rate based on the original surface temperature values ​​of the target area and a second temperature change rate based on the original surface temperature values ​​of the sub-regions, and calculating the absolute value of the difference between the first temperature change rate and the second temperature change rate; a first judgment and correction module for judging whether the absolute value of the difference is greater than a preset first threshold, and if so, adjusting the initial contribution of the corresponding sub-region based on the absolute value of the difference to obtain the first contribution and marking the corresponding sub-region; a second judgment and correction module for calculating the difference between the second temperature change rate of the marked sub-region and its adjacent sub-regions, processing to obtain the temperature change correlation coefficient of the marked sub-region, and when the temperature change correlation coefficient is lower than a preset second threshold, calculating and correcting the first contribution based on the temperature change correlation coefficient; and an output module for constructing and outputting an urban land cover classification thermal contribution report based on the corrected first contribution.

[0007] Secondly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned urban land cover classification system based on multispectral remote sensing technology.

[0008] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0009] 1. By calculating the difference in temperature change rate between the target area and sub-regions, the method upgrades from static land cover identification to dynamic thermal behavior assessment. By evaluating the thermal inertia characteristics of different materials and correcting their contribution accordingly, the method highlights the characteristics of dynamic thermal contribution and overcomes the limitation of traditional methods that only provide static classification results.

[0010] 2. The two-stage series correction mechanism significantly improves the accuracy of the assessment. The first-stage correction is based on the anomaly of the temperature change rate for initial adjustment, and the second-stage correction introduces spatial correlation analysis for secondary optimization, so that the assessment results take into account both its own characteristics and the influence of the surrounding environment, making them more comprehensive and reliable.

[0011] 3. The final generated heat contribution report can directly serve application decision-making, accurately locate key heat source areas, quantify the thermal environment responsibility of different areas, and provide intuitive and scientific decision-making basis for urban planning and thermal environment management, effectively shortening the path from data to application. Attached Figure Description

[0012] Figure 1A schematic diagram of the structure of an urban land cover classification system based on multispectral remote sensing technology provided in this application embodiment;

[0013] Figure 2 A schematic diagram of the logical flow of an urban land cover classification system based on multispectral remote sensing technology provided in an embodiment of this application. Detailed Implementation

[0014] This application provides an urban land cover classification system based on multispectral remote sensing technology, which solves the technical problem in the prior art where the accuracy of spatial correlation dynamic assessment and adjustment of the thermal contribution of urban land cover classification is insufficient.

[0015] In existing technologies, the assessment of urban surface thermal contribution mainly relies on static classification methods, which struggle to capture dynamic temperature changes and spatial correlation effects. Traditional techniques classify surface types based on spectral characteristics and then calculate thermal contributions using fixed weights, failing to consider the differences in thermal behavior of similar materials under different environments. For example, impermeable surfaces of the same type may exhibit significantly different temperature change patterns due to structural differences, but existing methods cannot identify such differences. Furthermore, heat transfer between adjacent areas is not included in the assessment system, causing the calculated thermal contribution results to deviate from the actual thermal environment distribution.

[0016] To address the aforementioned issues, a dynamic assessment method is needed that integrates temporal temperature variations with spatial correlation analysis. The inventors discovered that surface heat contribution depends not only on instantaneous temperature but also on the rate of temperature change and heat exchange with surrounding areas. By introducing differences in the rate of temperature change as a trigger condition, regions exhibiting abnormal thermal behavior can be identified; combining this with temperature correlation analysis of adjacent regions can correct errors introduced by isolated assessments. This dynamic adjustment mechanism effectively overcomes the limitations of static classification and spatially isolated assessments.

[0017] Therefore, this application proposes a systematic solution including data acquisition, rate of change calculation, contribution correction, and correlation analysis. The system first acquires surface temperature data and initial thermal contribution of the target area and its sub-regions, and calculates the differences in temperature change rates at different levels. When the difference exceeds a threshold, contribution adjustment is triggered, followed by secondary correction through temperature correlation analysis of adjacent areas, ultimately generating a thermal contribution report that includes spatial correlation effects.

[0018] The initial contribution refers to the basic thermal impact weight of the land surface unit obtained after preprocessing multispectral remote sensing data, which can be determined using regression analysis of spectral indices and land surface temperature. The first temperature change rate is obtained by differential calculation of continuous time-series temperature data of the target area, reflecting the overall thermal environment evolution trend. The absolute value of the difference serves as the benchmark for trigger threshold comparison, and the temperature difference change rate between adjacent time phases can be calculated using the sliding window method. The temperature change correlation coefficient is calculated using the Pearson correlation coefficient to determine the synchronicity of temperature change trends in adjacent sub-regions, used to identify anomalous heat source areas. The correction amplitude mapping relationship is established based on the heat conduction equation, converting the correlation strength into a contribution adjustment amount.

[0019] Specifically, the system extracts the land surface temperature field from multi-temporal remote sensing data and calculates the difference between the temperature change rate of each sub-region and the overall region. When the temperature change of a certain sub-region deviates significantly from the overall trend, an initial adjustment of its thermal contribution is automatically triggered. Subsequently, the correlation between the temperature changes of this region and the surrounding regions is analyzed. If the correlation is weak, it is determined to be an independent heat source, and its contribution weight needs to be further reduced. Finally, the spatially correlated and corrected contribution data is combined with the geographic information system to generate a thermal contribution distribution map.

[0020] Compared with existing technologies, this scheme innovatively combines the temporal dimension of temperature change rate with the spatial dimension of thermal correlation analysis. Traditional methods calculate thermal contribution based on only single-phase data, while this scheme captures dynamic thermal effects by continuously monitoring temperature change trends. Existing technologies assess the contribution of each region in isolation, while this scheme introduces a correlation correction mechanism between adjacent regions, effectively eliminating assessment bias caused by heat transfer.

[0021] Through the above technical solutions, this application achieves a transformation in urban surface heat contribution assessment from static to dynamic, and from isolated to correlated approaches. The dynamic adjustment mechanism can accurately identify sudden heat sources, and spatial correlation correction can eliminate inter-regional thermal interference errors, providing more accurate data support for urban heat island management. The multi-level threshold triggering and dual correction strategy ensure that the assessment results reflect both local thermal anomalies and conform to the overall thermal environment evolution patterns.

[0022] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0023] like Figure 1The diagram shown is a schematic representation of the structure of an urban land cover classification system based on multispectral remote sensing technology provided in this application embodiment. It includes: an acquisition module for acquiring the original surface temperature values ​​and initial contribution of the target area and sub-regions, wherein the initial contribution characterizes the degree of thermal contribution of the original surface temperature values ​​to land cover classification; a calculation module for calculating a first temperature change rate based on the original surface temperature values ​​of the target area and a second temperature change rate based on the original surface temperature values ​​of the sub-regions, and calculating the absolute value of the difference between the first and second temperature change rates; a first judgment and correction module for judging whether the absolute value of the difference is greater than a preset first threshold; if so, adjusting the initial contribution of the corresponding sub-region based on the absolute value of the difference to obtain the first contribution and marking the corresponding sub-region; a second judgment and correction module for calculating the difference between the second temperature change rates of the marked sub-region and its adjacent sub-regions, processing to obtain the temperature change correlation coefficient of the marked sub-region, and calculating and correcting the first contribution based on the temperature change correlation coefficient when the temperature change correlation coefficient is lower than a preset second threshold; and an output module for constructing and outputting an urban land cover classification thermal contribution report based on the corrected first contribution.

[0024] Furthermore, the initial contribution of the corresponding sub-region is adjusted based on the absolute value of the difference, specifically including: calculating the difference between the absolute value of the difference and a preset first threshold; multiplying the difference by the material thermal tuning coefficient to obtain a first correction value; adding the first correction value to the initial contribution to obtain a first contribution; wherein, the first contribution is used to characterize the thermal contribution assessment result after preliminary correction.

[0025] In this embodiment, the initial contribution of the corresponding sub-region is adjusted based on the absolute value of the difference. The specific correction constraint formula is as follows:

[0026] ;in, The first contribution score represents the preliminary corrected result of the region's thermal contribution, which is used as input for subsequent modules.

[0027] The initial contribution level is set based on historical data or expert experience. The technical effect is to provide a corrective benchmark.

[0028] The absolute value of the difference indicates the magnitude of the deviation between the second target region and the first target region in terms of the rate of temperature change, used to measure the difference between local and global behavior. Technical effectiveness is a key indicator for triggering corrections; a larger value indicates a more pronounced anomaly.

[0029] The preset first threshold is used to determine whether the absolute value of the difference is significant. It is set as a percentage of the absolute value of the first temperature change rate based on historical data or expert experience.

[0030] Indicates the first rate of temperature change. This represents the second rate of temperature change.

[0031] The thermal tuning coefficient is a coefficient that reflects the material's sensitivity to temperature changes. It is determined by the material type and proportion and is used to adjust the correction range.

[0032] Specifically, when the difference in temperature change rate between the target area and the sub-region exceeds a preset threshold, the portion exceeding the threshold is first calculated. For example, if the absolute difference is 5% and the threshold is 3%, the excess is 2%. This excess is then multiplied by the thermal tuning coefficient of the corresponding sub-region. For instance, if the sub-region is primarily composed of asphalt and its thermal tuning coefficient is 0.8, the correction value is 2% × 0.8 = 1.6%. Finally, the correction value is added to the initial contribution. For example, if the initial contribution is 15%, the corrected contribution becomes 16.6%. This process, by introducing the sensitivity of material properties to temperature changes, achieves dynamic adjustment of the thermal contribution assessment.

[0033] Existing technologies only correct the contribution of materials based on a fixed percentage of temperature differences, without considering the differences in thermal response between different materials. For example, even if concrete and vegetated areas have the same rate of temperature change, their actual thermal contributions may differ due to differences in the thermal inertia of their materials. This solution incorporates material properties into the correction process through a material thermal tuning coefficient, enabling the correction value to reflect the true degree of influence of different materials on temperature changes.

[0034] Furthermore, adjusting the initial contribution of the corresponding sub-region based on the absolute value of the difference also includes: when the original value before taking the absolute value of the difference corresponding to the marked sub-region is less than zero, the difference between the absolute value of the difference and the preset first threshold is calculated and then a proportion analysis is performed with the preset first threshold; a difference analysis is performed between the predefined constant and the proportion analysis result, and then multiplied with the material thermal tuning coefficient to obtain the second correction value; the second correction value is multiplied with the initial contribution to obtain the first contribution.

[0035] In this embodiment, when At that time, the specific correction constraint formula remains unchanged, and it should be noted that... and None of them are zero, when At that time, the specific correction constraint formula was modified as follows:

[0036] ;in, As the first contributor, The initial contribution level, with a value between 0 and 1. The absolute value of the difference. The first threshold is preset. Indicates the first rate of temperature change. This represents the second rate of temperature change. The thermal tuning coefficient of the material.

[0037] When the first temperature change rate is greater than the second temperature change rate, it indicates that the temperature fluctuation in the target region is more severe than that in the corresponding sub-region. In this case, the first contribution should be the result obtained by adjusting the initial contribution downward.

[0038] Specifically, when the temperature change rate of a sub-region is lower than the overall level of the target region, it indicates that the region may contain materials with high thermal inertia or impeded heat exchange. In this case, a correction factor is generated by calculating the relative percentage of the absolute difference between the difference and the threshold, combined with the material's thermal tuning coefficient. For example, when the temperature change rate of a concrete region is 15% lower than the overall rate and the threshold is set to 10%, the percentage difference is (15%~10%) / 10%=0.5. A difference analysis is performed between this percentage and a predefined constant (e.g., 1-0.5=0.5), and then multiplied by the tuning coefficient of the concrete material in that region (0.8) to obtain a correction factor of 0.4. Finally, this correction factor is multiplied by the initial contribution value to achieve a non-linear adjustment of the thermal contribution value, accurately reflecting the suppressive effect of material thermal inertia on temperature changes.

[0039] Existing technologies only make linear corrections based on the absolute difference in the rate of temperature change, without considering the impact of the direction of temperature change on the assessment of thermal contribution.

[0040] This application effectively solves the problem of misjudgment of material thermal contribution caused by ignoring the direction of temperature change in existing technologies. By establishing a correction mechanism sensitive to the difference in direction, it can accurately distinguish the dynamic thermal response differences between thermally inertial materials and low-heat-capacity materials, avoiding the misidentification of thermal buffering effects as low-heat-contribution behavior. For example, when evaluating large-area water bodies, it can accurately capture their bidirectional thermal regulation effect of daytime heat absorption and nighttime heat release, providing reliable data support for the selection of materials for urban heat island mitigation measures.

[0041] Furthermore, the specific process for obtaining the material thermal tuning coefficient is as follows: Remote sensing image classification and spectral analysis are performed on the raw urban surface remote sensing data to obtain the area of ​​the marked sub-region and the area of ​​different urban surface materials within the marked sub-region. These area regions are then numbered. A ratio analysis is performed between the area of ​​the same urban surface material and the area of ​​the marked sub-region, followed by coupling analysis with the corresponding standard thermal tuning coefficient. The coupling analysis results are then summed according to the area region numbers to obtain the material thermal tuning coefficient for different urban surface materials. The material thermal tuning coefficient reflects the comprehensive sensitivity level of different urban surface materials to temperature changes within the marked sub-region.

[0042] In this embodiment, the specific formula for calculating the material thermal tuning coefficient is as follows:

[0043] ;in, Indicates the type of urban surface material. This indicates the initial value of the urban surface material type number. , This indicates the total number of urban surface material type numbers. The marked sub-region contains multiple areas with the same urban surface material type, which are then numbered. This indicates the area zone number under the urban surface material type. , This indicates the total number of area zone numbers under each urban surface material type.

[0044] Indicates the first Material thermal tuning coefficient for urban surface materials.

[0045] It is the first The first city surface material The area ratio under each area region is used to represent the coverage of a certain urban surface material within the marked sub-region, and is obtained through remote sensing image classification and spectral analysis.

[0046] The area of ​​the corresponding marked sub-region is obtained through remote sensing image classification and spectral analysis.

[0047] It is the first The first city surface material The standard thermal tuning coefficient for a given area is extracted from an urban land cover classification database. It is a preset constant used to characterize the inherent sensitivity parameters of the material, obtained through laboratory measurements or calibration with historical data and then normalized. For example, the K_m of asphalt may be higher, while that of vegetation may be lower. The technical effect is to provide a standard value.

[0048] according to Substituting these into the specific correction constraint formula, we can further obtain... That is, the first sub-region to be marked. The first contribution of the surface material of each city is reflected in the material thermal tuning coefficient, which reflects the comprehensive sensitivity of different surface materials of different cities to temperature changes in the marked sub-region.

[0049] After acquiring multispectral remote sensing data, a supervised classification algorithm is first used to delineate the distribution areas of different materials such as building concrete, asphalt pavement, and green vegetation within the marked sub-regions, and the area of ​​the dispersed patches for each material is calculated. For example, in a sub-region with a total area of ​​1000 square meters, the total area of ​​concrete patches is 300 square meters, and the total area of ​​asphalt patches is 500 square meters. Next, the ratio of each material patch area to the total area of ​​the sub-region is calculated, resulting in a concrete proportion of 30% and an asphalt proportion of 50%. Then, each material proportion is multiplied by its corresponding standard thermal tuning coefficient. For example, the standard coefficient for concrete is 0.8, and the standard coefficient for asphalt is 1.2, so the coupling value for concrete is 0.3 × 0.8 = 0.24, and the coupling value for asphalt is 0.5 × 1.2 = 0.6. Finally, the coupling values ​​of each material are summed to obtain the material thermal tuning coefficient for the sub-region as 0.24 + 0.6 = 0.84. This value comprehensively reflects the thermal sensitivity characteristics and spatial distribution features of different materials within the region.

[0050] Existing technologies treat impermeable surfaces as homogeneous, while this solution, through material-level subdivision and spatial distribution coupling calculation, can distinguish the differences in thermal response between similar impermeable surfaces such as concrete structures and asphalt pavements. Existing technologies use fixed thermal contribution coefficients, leading to assessment biases. This solution, by dynamically coupling area proportion and standard coefficients, accurately reflects the comprehensive impact of different material combinations on temperature changes, providing a data foundation for identifying key heat source materials. For example, when a high tuning coefficient is detected in a sub-region, areas with excessively high asphalt content can be quickly located.

[0051] Furthermore, the temperature change correlation coefficient of the marked sub-region is obtained through processing, specifically including: obtaining the second temperature change rate of different urban surface materials under the marked sub-region and its adjacent sub-regions; calculating the mean and standard deviation of the second temperature change rate under the marked sub-region and its adjacent sub-regions based on the second temperature change rate through mean and standard deviation calculations; calculating the temperature change correlation coefficient between the marked sub-region and its adjacent sub-regions based on the mean and standard deviation of the second temperature change rate through the Pearson correlation coefficient; the temperature change correlation coefficient is used to quantify the synchronicity level of temperature changes between the marked sub-region and its adjacent sub-regions.

[0052] In this embodiment, the temperature change correlation coefficient is the correlation coefficient of the second temperature change rate between the marked sub-region and its adjacent sub-regions, used to quantify the synchronicity of temperature changes between regions, and is calculated using the Pearson correlation coefficient formula.

[0053] ;in, Indicates the marked sub-region, This indicates the adjacent sub-regions of the marked sub-region. Indicates the sub-region under the marker. The second temperature change rate of different urban surface materials Indicates the next adjacent subregion The second temperature change rate of surface materials in different cities.

[0054] Indicates the sub-region under the marker. The average of the second temperature change rate of surface materials in different cities Indicates the next adjacent subregion The average of the second temperature change rate of surface materials in different cities.

[0055] Indicates the sub-region under the marker. The standard deviation of the second temperature change rate of different urban surface materials Indicates the next adjacent subregion The standard deviation of the second temperature change rate of different urban surface materials.

[0056] Number the adjacent sub-regions of the marked sub-region. , This represents the initial value of the adjacent sub-region number. This represents the total number of adjacent sub-regions, for example, Indicates the sub-region marked and the first The correlation coefficient of temperature change between adjacent sub-regions.

[0057] When calculating the correlation coefficient of temperature changes, the temperature change rate data of the marked sub-region and its adjacent sub-regions under different materials are first extracted. The mean and standard deviation of each region are calculated using statistical methods to eliminate noise interference from random fluctuations. Subsequently, a standardized data sequence is constructed based on the mean and standard deviation, and the correlation degree between the marked sub-region and each adjacent sub-region is calculated using the Pearson correlation coefficient formula. This process effectively captures the synergy or difference in temperature changes between adjacent regions, providing a quantitative basis for subsequent judgment on whether contribution correction is necessary.

[0058] Existing technologies only analyze the thermal contribution of a single sub-region independently, without considering the coupling effect of heat transfer between adjacent regions on temperature changes, which leads to the evaluation results deviating from the spatial correlation characteristics of the actual thermal environment.

[0059] This application can accurately identify the thermal interactions between the marked sub-region and the surrounding region, avoiding evaluation bias caused by isolated analysis. By quantifying the statistical characteristics and spatial correlation of temperature changes, it provides a reliable correlation strength index for contribution correction, thereby improving the spatial consistency and physical rationality of the thermal contribution report.

[0060] Furthermore, the first contribution is corrected based on the temperature change correlation coefficient. Specifically, this includes: jointly analyzing the first contribution of the marked sub-region with its neighboring sub-regions, the first temperature change correlation coupling factor, and the contribution difference correction factor to obtain the corrected first contribution components of the marked sub-region with different neighboring sub-regions; summing and averaging the corrected first contribution components of all neighboring sub-regions to obtain the corrected first contribution of the marked sub-region; the first temperature change correlation coupling factor is obtained by coupling the temperature change correlation coefficient between the marked sub-region and its neighboring sub-regions with a predefined constant; the contribution difference correction factor is obtained by analyzing the difference between the predefined constant and the correction factor; the correction factor is obtained by coupling the difference between the first contribution of the marked sub-region and the first contribution of its neighboring sub-regions with the heat exchange attenuation coefficient of the marked sub-region and its neighboring sub-regions; the heat exchange attenuation coefficient is calculated using a predefined characteristic attenuation distance and distance ratio.

[0061] In this embodiment, the formula for calculating the corrected first contribution based on the temperature change correlation coefficient is as follows:

[0062] ;

[0063] ;

[0064] in, The sub-region marked and the first The corrected first contribution component of the association between adjacent sub-regions The corrected first contribution of the labeled sub-region.

[0065] The first contribution of the labeled sub-region.

[0066] The sub-region marked and the first The heat exchange attenuation coefficient of adjacent sub-regions.

[0067] Indicates the sub-region marked and the first The correlation coefficient of temperature change between adjacent sub-regions.

[0068] For the first The first contribution of each adjacent sub-region.

[0069] Specifically, when calculating the corrected first contribution, a first temperature change correlation coupling factor is first obtained based on the product of the temperature change correlation coefficient and a predefined constant. This factor can amplify or reduce the correlation effect of temperature changes in adjacent regions. Simultaneously, a contribution difference correction factor is calculated using the difference between the predefined constant and the correction factor. This correction factor is determined by the product of the contribution difference between adjacent regions and the heat exchange attenuation coefficient, thus incorporating the spatial heat exchange effect into the correction process. The above factors are coupled with the first contribution of adjacent regions to obtain the correction component of each adjacent region on the current region. Finally, the corrected contribution of the current region is obtained by averaging all components. For example, the heat exchange attenuation coefficient can be based on an exponential function to simulate the attenuation characteristics of heat with distance, ensuring stronger temperature correlation in nearby regions.

[0070] Existing technologies analyze the thermal contribution of a single region in isolation, neglecting the heat transfer and coupling effects between adjacent regions. This proposed solution, however, introduces a heat exchange attenuation coefficient and a temperature change-related coupling factor to dynamically adjust the weighting of the influence of adjacent regions on the current region's thermal contribution, thus more accurately reflecting the spatial transfer and distribution patterns of heat. For example, existing technologies fail to consider heat exchange attenuation effects, leading to incorrect estimates of the thermal contribution of distant regions. This proposed solution, through the coupling calculation of distance ratios and natural constants, effectively corrects such errors.

[0071] Furthermore, the calculation of the corrected first contribution based on the temperature change correlation coefficient also includes: when the original value before taking the absolute value of the difference corresponding to the marked sub-region is less than zero, a difference analysis is performed on the predefined constant and the temperature change correlation coefficient between the marked sub-region and its adjacent sub-regions to obtain the second temperature change correlation coupling factor; the first contribution, the second temperature change correlation coupling factor, and the contribution difference correction factor of the marked sub-region and its adjacent sub-regions are coupled and analyzed together to obtain the corrected first contribution components of the marked sub-region and different adjacent sub-regions; the corrected first contribution components of all adjacent sub-regions are summed and averaged to obtain the corrected first contribution of the marked sub-region.

[0072] In this embodiment, when At that time, the formula used to derive the first contribution remains unchanged. It should be noted that... and None of them are zero, when At that time, the formula for specifically correcting the first contribution level was modified as follows:

[0073] ;

[0074] ;

[0075] The contribution difference correction factor is a correction parameter generated based on the heat exchange attenuation coefficient and contribution difference between the marked sub-region and its neighboring sub-regions. Specifically, it can be achieved by multiplying the contribution difference with the heat exchange attenuation coefficient, and is used to quantify the impact of heat transfer between neighboring regions on the contribution correction.

[0076] The heat exchange attenuation coefficient is a parameter that reflects the degree of heat transfer attenuation between a marked sub-region and its adjacent sub-regions with distance. Specifically, it can be calculated by raising the negative distance ratio power of the natural constant, where the distance ratio is the ratio of the distance between the center points of the two regions to the predefined characteristic attenuation distance, and is used to characterize the spatial attenuation effect of heat transfer.

[0077] Specifically, when the original value of the absolute difference is negative, it indicates that the temperature change rate of the marked sub-region is lower than that of the adjacent regions. In this case, a second temperature change correlation coupling factor needs to be introduced to adjust the correction weight. First, the difference between the temperature change correlation coefficient and a predefined constant is calculated to generate the second temperature change correlation coupling factor. This factor increases the correction magnitude when the correlation coefficient is low. Next, the contribution difference correction factor is calculated by combining the difference in the first contribution of the marked sub-region and the adjacent regions and the heat exchange attenuation coefficient. The above factor is multiplied by the second temperature change correlation coupling factor and weighted and coupled with the first contribution of the adjacent regions to obtain the correction component corresponding to each adjacent region. Finally, the average value of the correction components of all adjacent regions is taken as the first contribution of the marked sub-region after correction. This process dynamically adjusts the correction weight under negative difference values ​​and introduces the heat exchange attenuation coefficient to quantify the spatial heat transfer effect, making the correction result more consistent with the spatial correlation characteristics of the actual thermal environment.

[0078] Existing technologies only use a fixed threshold to adjust the contribution when the absolute value of the difference is negative, without considering the influence of the temperature change correlation coefficient and the heat exchange attenuation in adjacent areas on the correction process, resulting in the correction results failing to accurately reflect the spatial correlation of heat transfer.

[0079] By introducing a second temperature change correlation coupling factor and a heat exchange attenuation coefficient, dynamic adaptation of the correction weights is achieved in negative difference scenarios, and the physical law of heat transfer attenuation with distance is quantified. By coupling the temperature change correlation coefficient, the heat exchange attenuation coefficient and the contribution difference, the corrected first contribution can more accurately characterize the responsibility ratio of the marked sub-region in the overall thermal environment, providing reliable data support for the location of key heat sources in urban heat island effect governance.

[0080] Furthermore, the specific calculation process of the heat exchange attenuation coefficient is as follows: the heat exchange attenuation coefficient between the marked sub-region and its adjacent sub-region is calculated by raising the negative distance ratio of the predefined characteristic attenuation distance to the natural constant; the distance ratio is obtained by analyzing the ratio of the distance between the center points of the marked sub-region and its adjacent sub-region to the predefined characteristic attenuation distance.

[0081] In this embodiment, the formula for calculating the heat exchange attenuation coefficient is:

[0082] This formula is based on an exponential decay model, where, Represents the natural constant; This is the basic attenuation constant; it reflects the maximum heat exchange capacity under specific conditions and is related to the properties of the medium and environmental conditions. This parameter can be obtained through experimental measurement or fitting of historical data, and can be adjusted to adapt to environmental changes under different seasons and climatic conditions.

[0083] The sub-region marked and the first The distance between the center points of adjacent sub-regions; this distance is calculated using geographic coordinates and the Euclidean distance formula: ,in This represents the coordinates of the center points of the two regions mentioned above. This parameter is a key input for attenuation calculation and directly affects the calculated heat exchange coefficient.

[0084] The predefined characteristic attenuation distance is a scale parameter of the heat exchange attenuation model, representing the characteristic length of thermal effect attenuation. This parameter can be determined by fitting observational data, and is typically set within the range of 50–200 meters in urban environments.

[0085] According to the above text That is, the first sub-region marked Substituting the first contribution value of each city's surface material into the formula for correcting the first contribution value, we obtain... , for the marked sub-region and the first The first adjacent sub-region associated with the first The corrected first contribution component under the surface material of each city is then obtained. , is the first of the marked sub-regions The first contribution of the corrected urban surface material.

[0086] When calculating the heat exchange attenuation coefficient, the geographic center coordinates of the marked sub-region and its adjacent sub-regions are first obtained, and the Euclidean distance between them is calculated as the actual spatial distance. The actual spatial distance is divided by a predefined characteristic attenuation distance to obtain the distance ratio, which reflects the proportional relationship between the heat transfer path and the typical effective range of the material. Subsequently, using the negative power of the distance ratio (the natural constant) as the core of the calculation, an exponential function is used to simulate the attenuation characteristics of heat during spatial transfer. For example, when the distance ratio is 1, it indicates that the actual distance equals the characteristic attenuation distance. At this time, the heat exchange attenuation coefficient is the negative first power of the natural constant, corresponding to a heat transfer efficiency of approximately 36.8%, which conforms to the basic laws of the thermodynamic diffusion model.

[0087] Existing technologies typically use the reciprocal of distance or a fixed threshold to determine the spatial correlation strength, without considering the nonlinear effect of the thermal conductivity of different materials on heat attenuation.

[0088] This application addresses the problem of neglecting spatial attenuation effects in existing heat contribution assessments and optimizes the quantification method of heat exchange impact between adjacent areas. By dynamically adjusting the characteristic attenuation distance and constructing an exponential attenuation model, it can accurately reflect the differences in the effects of different surface materials on heat transfer, making the heat contribution correction results more consistent with the spatial coupling characteristics of the real thermal environment and providing reliable data support for urban heat island effect management.

[0089] Furthermore, the system also includes a third judgment module: after obtaining the corrected first contribution value, it records the current time period and obtains the temperature change correlation coefficient between the marked sub-region and its adjacent external sub-region within the corresponding time period of the continuous predefined monitoring time period; it judges whether the temperature change correlation coefficient is within a preset third threshold; if it is within the preset third threshold, it keeps the corrected first contribution value of the marked sub-region unchanged; if it is not within the preset third threshold, it immediately updates the obtained corrected first contribution value until the number of updates reaches the upper limit. When the number of updates reaches the upper limit, it rolls back the corrected first contribution value of the marked sub-region to the corresponding first contribution value and issues an early warning.

[0090] In this embodiment, by establishing time-series markers and spatial correlation coefficient analysis, combined with iterative correction and backoff early warning mechanisms, the data accuracy and reliability of the temperature monitoring system are significantly improved. Its beneficial effects include: achieving cross-cycle data consistency alignment and scientifically quantifying regional thermal correlations; enhancing system stability and anti-interference capabilities through adaptive iterative updates and fault-tolerant processing; and possessing multi-scenario scalability, providing an efficient and reliable closed-loop solution for thermal environment management.

[0091] Furthermore, the urban surface classification thermal contribution report specifically includes: outputting the urban surface classification thermal contribution report based on the corrected first contribution value of the marked sub-regions; the content of the urban surface classification thermal contribution report includes the marked sub-regions and corresponding thermal contribution distribution maps for different predefined time periods; the thermal contribution distribution maps are assigned corresponding predefined labels based on different corrected first contribution values.

[0092] like Figure 2 The diagram shown is a logical flow diagram of an urban land cover classification system based on multispectral remote sensing technology provided in an embodiment of this application. It determines whether the absolute value of the difference is greater than a preset first threshold; otherwise, it maintains the initial contribution level, constructs and outputs an urban land cover classification thermal contribution report based on the initial contribution level. When the temperature change correlation coefficient is higher than or equal to a preset second threshold, it maintains the first contribution level, constructs and outputs an urban land cover classification thermal contribution report based on the first contribution level.

[0093] In this embodiment, after completing the spatial correlation correction of the thermal contribution of sub-regions, the system integrates the corrected data from different time periods to generate a report. For example, the first contribution of each sub-region is recorded during the high-temperature period in summer and the low-temperature period in winter, and the dynamic characteristics of the heat source material changing with the seasons are revealed through time-series comparison. The thermal contribution distribution map matches the corrected contribution with spatial coordinates through a geographic information system and uses gradient color rendering technology to achieve visualization output. For the same sub-region, if the difference in its thermal contribution in multiple time periods exceeds a set range, the cross-time period correlation analysis module is triggered to generate a material thermal inertia assessment sub-report as supplementary data.

[0094] The generation of the heat contribution distribution map can be combined with the overlay display of remote sensing imagery. For example, the corrected first contribution layer can be fused with high-resolution satellite imagery using transparency to help users identify the surface material type corresponding to high contribution areas. The threshold settings for predefined labels can be dynamically adjusted based on historical data statistics. For example, the percentile method can be used to define the top 20% contribution areas as high heat source areas.

[0095] Existing technologies only output static classification results in their thermal contribution reports, failing to reflect spatiotemporal dimension correction data, thus making it impossible to track the dynamic evolution of heat sources. This solution, however, enables urban planning departments to quickly locate material distribution areas with consistently high thermal contributions by providing visualized output and dynamic tracking capabilities for thermal contribution assessment results.

[0096] This application also provides a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the steps of an urban land cover classification system based on multispectral remote sensing technology.

[0097] The execution of a computer program by a processor refers to the process of calling the instruction set in the storage medium through the central processing unit. Specifically, this can be achieved using a multi-threaded parallel computing framework, enabling the system to process multispectral remote sensing data in real time and perform operations such as temperature change rate calculation, contribution correction, and thermal contribution report generation.

[0098] Specifically, after the computer program is loaded into memory, the processor parses the instructions and drives computing resources to execute the following processes: acquiring surface temperature data of the target area and sub-regions, calculating the absolute value of the temperature change rate difference; triggering a contribution correction mechanism based on the difference threshold, adjusting the initial contribution through the material thermal tuning coefficient; performing secondary correction based on the correlation coefficient of temperature changes in adjacent sub-regions, and finally constructing a thermal contribution distribution map. During the processing, the program achieves dynamic correction and spatial correlation analysis by calling predefined mapping relationships and coupling analysis algorithms.

[0099] Compared to existing technologies, traditional storage media only store static classification models and lack dynamic correction logic, making it unable to support the coupled calculation of temperature change rate and spatial correlation. This solution controls the processor through program instructions to perform operations such as absolute difference value judgment, material thermal tuning coefficient coupling, and correlation coefficient correction, enabling the storage medium to not only carry data but also become the core carrier driving dynamic thermal contribution assessment.

[0100] Through the above technical solution, this application enables the urban surface heat contribution assessment system to have cross-platform deployment capabilities, and ensures the consistency of data processing logic under different hardware environments through standardized program instructions. The program has embedded temperature change rate difference analysis rules and correction algorithms in the storage medium to avoid calculation errors caused by manual intervention. At the same time, it supports the batch generation and storage of heat contribution distribution maps for multiple time periods, providing traceable decision-making basis for urban planning.

[0101] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0103] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0104] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0105] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0106] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A city land cover classification system based on multispectral remote sensing technology, characterized in that, The system includes: Acquisition module: used to acquire the original surface temperature value and initial contribution of the target area and sub-regions. The initial contribution is used to characterize the degree of contribution of the original surface temperature value to the surface classification heat. The sub-regions are obtained by segmenting the target area. Calculation module: used to calculate the first temperature change rate based on the original surface temperature value of the target area, and to calculate the second temperature change rate based on the original surface temperature value of the sub-area, and to calculate the absolute value of the difference between the first temperature change rate and the second temperature change rate; First judgment and correction module: used to judge whether the absolute value of the difference is greater than a preset first threshold. If so, the initial contribution of the corresponding sub-region is adjusted based on the absolute value of the difference to obtain the first contribution and mark the corresponding sub-region. The second judgment and correction module is used to calculate the difference in the second temperature change rate between the marked sub-region and its adjacent sub-regions, process and obtain the temperature change correlation coefficient of the marked sub-region, and when the temperature change correlation coefficient is lower than the preset second threshold, calculate and correct the first contribution based on the temperature change correlation coefficient. The output module is used to construct and output a city surface classification thermal contribution report based on the corrected first contribution. The calculation of the corrected first contribution based on the temperature change correlation coefficient specifically includes: Indicates the first rate of temperature change. This represents the second rate of temperature change; when When the first contribution, first temperature change correlation coupling factor, and contribution difference correction factor of the marked sub-region are coupled together with the adjacent sub-regions for analysis, the corrected first contribution components of the marked sub-region and different adjacent sub-regions are obtained. The corrected first contribution components of all the adjacent sub-regions are summed and averaged to obtain the corrected first contribution of the marked sub-region; The first temperature change correlation coupling factor is obtained by coupling the temperature change correlation coefficient between the marked sub-region and its neighboring sub-regions with a predefined constant; The contribution difference correction factor is obtained through difference analysis between a predefined constant and a correction factor; The correction factor is obtained through a coupled analysis of the difference between the first contribution of the marked sub-region and the first contribution of the adjacent sub-region and the heat exchange attenuation coefficient of the marked sub-region and its adjacent sub-region. The heat exchange attenuation coefficient is calculated using a predefined characteristic attenuation distance and distance ratio; The calculation of the corrected first contribution based on the temperature change correlation coefficient also includes: Indicates the first rate of temperature change. This represents the second rate of temperature change; when When the temperature change correlation coefficient between the predefined constant and the labeled sub-region and its adjacent sub-region is analyzed, a second temperature change correlation coupling factor is obtained. The first contribution degree, second temperature change correlation coupling factor, and contribution degree difference correction factor of the marked sub-region and its neighboring sub-regions are coupled and analyzed together to obtain the corrected first contribution degree components of the marked sub-region and its different neighboring sub-regions. The corrected first contribution components of all the adjacent sub-regions are summed and averaged to obtain the corrected first contribution of the marked sub-region; The specific calculation process for the heat exchange attenuation coefficient is as follows: The distance ratio is obtained by analyzing the ratio of the distance between the center points of the marked sub-region and its two neighboring sub-regions to the predefined feature decay distance. The heat exchange attenuation coefficient between the marked sub-region and its neighboring sub-regions is calculated by raising the ratio of the predefined characteristic attenuation distance to the negative distance of the natural constant.

2. The urban land cover classification system based on multispectral remote sensing technology according to claim 1, characterized in that, The specific steps to obtain the first contribution score and mark the corresponding sub-region are as follows: Obtain the thermal tuning coefficient of the material; Indicates the first rate of temperature change. This represents the second rate of temperature change; when When the difference is calculated, the result of the difference between the absolute value of the difference and the preset first threshold is multiplied by the material thermal tuning coefficient to obtain the first correction value; The first correction value is added to the initial contribution value to obtain the first contribution value, and the corresponding sub-region is marked.

3. The urban land cover classification system based on multispectral remote sensing technology according to claim 2, characterized in that, Obtaining the first contribution score and marking the corresponding sub-region also includes: Indicates the first rate of temperature change. This represents the second rate of temperature change; when When the difference is calculated, the result of the difference between the absolute value of the difference and the preset first threshold is compared with the preset first threshold for a ratio analysis. The predefined constant is compared with the ratio analysis results by performing a difference analysis, and then multiplied by the material thermal tuning coefficient to obtain the second correction value; Multiply the second correction value by the initial contribution to obtain the first contribution and mark the corresponding sub-region.

4. The urban land cover classification system based on multispectral remote sensing technology according to claim 2 or 3, characterized in that, The specific process for obtaining the thermal tuning coefficient of the material is as follows: Remote sensing image classification and spectral analysis were performed on raw urban surface remote sensing data to obtain the area of ​​marked sub-regions and the area of ​​different urban surface materials under the marked sub-regions. Number the area regions to obtain area region numbers; After analyzing the proportion of the area of ​​the same urban surface material to the area of ​​the marked sub-region, coupling analysis is performed with the corresponding standard thermal tuning coefficient to obtain the coupling analysis results. The coupling analysis results are then summed based on the area region numbering to obtain the material thermal tuning coefficients for different urban surface materials.

5. The urban land cover classification system based on multispectral remote sensing technology according to claim 1, characterized in that, The processing yields the temperature change correlation coefficients for the marked sub-regions, specifically including: Obtain the second temperature change rate of different urban surface materials under the marked sub-region and its adjacent sub-regions; Based on the second temperature change rate, the mean and standard deviation of the second temperature change rate under the marked sub-region and its adjacent sub-regions are calculated by mean and standard deviation respectively. The correlation coefficient between the temperature change of the marked sub-region and its neighboring sub-regions is calculated using the Pearson correlation coefficient based on the mean of the second temperature change rate and the standard deviation of the second temperature change rate.

6. The urban land cover classification system based on multispectral remote sensing technology according to claim 1, characterized in that, The system also includes a third judgment module: The third judgment module is used to record the current time period after obtaining the corrected first contribution, and to obtain the temperature change correlation coefficient between the marked sub-region and its adjacent sub-region and the external adjacent sub-region within the corresponding time period in the continuous predefined monitoring time period. Determine whether the temperature change correlation coefficient is within a preset third threshold. If it is within the preset third threshold, the corrected first contribution of the marked sub-region remains unchanged; If it is not within the preset third threshold, the corrected first contribution value is immediately updated until the number of updates reaches the upper limit. When the number of updates reaches the upper limit, the corrected first contribution value of the marked sub-region is rolled back to the corresponding first contribution value and an early warning is issued.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the system as described in any one of claims 1-6.

Images