City block building color harmony degree evaluation method and device

Abstract

The invention relates to the technical field of crossing of city planning, color evaluation and computer vision, in particular to a city block building color harmony evaluation method and device, and the method comprises the steps: collecting and preprocessing data, obtaining block building color data, spatial position data and planning constraint data, converting the data into a unified color space, and removing noise; feature extraction: calculating single building color features, adjacent building color correlation features and block overall color distribution features; constructing a multi-dimensional evaluation index system including color suitability, adjacent coordination degree, overall balance degree and planning integrating degree; determining an index weight based on an analytic hierarchy process-entropy weight method; according to the method, objective and quantitative evaluation of the color harmony degree of the building is realized, the evaluation efficiency and accuracy are improved, a scientific basis is provided for urban block color planning and reconstruction, and the technical problems that traditional evaluation depends on subjective judgment and lacks a unified standard are solved.

Description

Technical Field

[0001] This invention relates to the interdisciplinary fields of urban planning, color assessment, and computer vision, and in particular to a method and apparatus for assessing the color harmony of buildings in urban blocks. Background Technology

[0002] Architectural color, as the visual soul of a city's landscape, is a core carrier for showcasing the regional characteristics, cultural heritage, and international image of the Hainan Free Trade Port. With the accelerated construction of the Free Trade Port, the pace of urban development and renewal in core cities such as Haikou and Sanya has significantly accelerated. Traditional architectural complexes, such as old arcade streets and Li and Miao villages, coexist with newly built commercial buildings, cultural and tourism facilities, and cross-border trade platforms, forming a spatial pattern of diverse architectural forms. According to the "Hainan Free Trade Port Territorial Spatial Planning Regulations," the color of building facades must be approved by the natural resources and planning authorities and cannot be changed without authorization. Those who change it without authorization will face fines ranging from 10,000 to 50,000 yuan. This regulation highlights the rigid status of color control in the spatial governance of the Free Trade Port. Meanwhile, the "Implementation Opinions on Further Strengthening Urban Planning, Construction, and Governance" clearly states that architectural colors should be determined in conjunction with the natural environment and local landscape elements to fully reflect regional, ethnic, and contemporary characteristics. Haikou City, in its historical and cultural city protection plan, emphasizes that newly constructed buildings around historical buildings must be coordinated with them in terms of color and other dimensions.

[0003] Against this backdrop, the control of architectural color in the Hainan Free Trade Port faces multiple practical challenges: First, the dynamic planning characteristics of the Free Trade Port's "flexible development zone" mean that there is a lack of quantitative standards for the color connection between new and existing buildings. This has led to problems such as clashes between high-saturation commercial buildings and low-brightness traditional arcade buildings, and a disconnect between modern coastal architecture and Li and Miao ethnic totem color elements, thus disrupting the overall "coconut breeze and sea charm" aesthetic. Second, existing assessment methods are ill-suited to the diverse needs of the Free Trade Port's various street types. From the international color positioning of the Boao Lecheng International Medical Tourism Pilot Zone to the technological color guidance around the Wenchang Space Launch Center, and the cultural heritage of the Danzhou Dongpo Cultural Street, the color requirements of different functional blocks vary significantly. However, the traditional manual subjective assessment model lacks a unified quantitative system, resulting in poor repeatability and insufficient persuasiveness of the assessment results. Thirdly, the development scope of the Free Trade Port blocks is wide and the pace is fast. The traditional on-site manual measurement method is inefficient and cannot meet the rapid assessment needs of contiguous development areas such as Haikou Jiangdong New Area and Sanya Yazhou Bay Science and Technology City. Moreover, it cannot accurately locate areas with color violations to support law enforcement and supervision. Fourthly, existing assessments mostly focus on the color compliance of individual buildings, ignoring the color relationship between adjacent buildings and the overall balance of the block. Furthermore, there is a lack of color harmony standards that match the "blue-green intertwined" ecological pattern of the Free Trade Port, making it difficult to form targeted optimization solutions.

[0004] Therefore, in response to the special planning requirements and style positioning of the Hainan Free Trade Port, there is an urgent need to develop an objective, quantitative, and efficient method and device for assessing the color harmony of urban blocks and buildings. This would address the technical pain point that traditional assessments are incompatible with the development needs of the Free Trade Port, provide scientific support for the protection of historical blocks, the development of new areas, and the rectification of illegal colors, and help create a visual image for the Free Trade Port that combines international flair with local flavor. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for evaluating the color harmony of buildings in urban blocks, which realizes the objective and quantitative evaluation of building color harmony, improves the evaluation efficiency and accuracy, provides a scientific basis for urban block color planning and renovation, and solves the technical problems of traditional evaluation relying on subjective judgment and lacking unified standards.

[0006] To achieve the above objectives, the present invention provides a method and apparatus for evaluating the color harmony of buildings in urban blocks, comprising the following steps: Data Acquisition and Preprocessing: Collect building color data, building spatial location data, and street color planning constraint data of the target city blocks; convert the building color data to the CIELab color space, remove image noise through median filtering, and extract color sample sets of individual buildings based on image segmentation technology; Feature extraction: Single building color features: Calculate the mean hue angle, mean lightness, mean saturation, and color distribution variance for each building color sample set; Adjacent building color association features: Based on the building spatial location data, determine adjacent building pairs within a distance threshold, and calculate the hue angle difference, lightness difference, saturation difference, and color similarity for each pair of adjacent buildings; Overall street color distribution features: Statistically analyze the hue interval distribution frequency, lightness interval distribution frequency, and saturation interval distribution frequency of building colors within the street, and calculate the color diversity entropy value and the proportion of the dominant color tone; Evaluation index calculation: c. Construct a multi-dimensional evaluation index system, including: Color suitability index: Based on the street color planning constraint data, calculate the degree of fit between the color of a single building and the main color of the plan; Adjacent coordination index: Based on the color difference value and color similarity of adjacent buildings, calculate the overall adjacent coordination of the street using a weighted average method; Overall balance index: Combine color diversity entropy value, main color proportion, and color distribution variance to construct a quantitative formula for balance measurement; Planning fit index: Statistically calculate the proportion of buildings that meet the street color planning constraints and the degree of color deviation. Weight determination: Subjective weights of indicators are determined using the analytic hierarchy process (AHP), objective weights are determined using the entropy weight method, and a comprehensive weight is obtained based on linear weighted fusion. Harmony assessment: After normalizing each assessment indicator, the results are substituted into a fuzzy comprehensive evaluation model to obtain a preliminary harmony score. The preliminary score is then corrected using a trained random forest model to output the final harmony level and key influencing factors. Output results: Generate an evaluation report, including harmony score, level classification, location of disharmonious areas, and suggestions for color optimization and adjustment.

[0007] The building color data is obtained by combining drone aerial images, high-definition ground images, and building completion color archives. The resolution of the aerial images is no less than 0.1m, and the ground images cover the main facades of all buildings in the block.

[0008] The threshold for the distance between adjacent buildings is determined by the street network density, building volume ratio and planning specifications, and the value range is 5-30m. When the street building density is greater than 0.8, the threshold is 5-15m; when the building density is less than 0.3, the threshold is 20-30m.

[0009] The color suitability index is calculated as follows: The hue angle range of the main color scheme is [H1,H2], the brightness range is [L1,L2], and the saturation range is [S1,S2]. The hue angle of a single building color is H, the brightness is L, and the saturation is S. If H∈[H1,H2] and L∈[L1,L2] and S∈[S1,S2], the suitability score is 1; otherwise, the score is 1-[|H-(H1+H2) / 2| / (180)+|L-(L1+L2) / 2| / 100+|S-(S1+S2) / 2| / 100] / 3.

[0010] The quantitative formula for the overall balance index is: Balance = 0.4 × (1 - |Entropy value - 0.6| / 0.6) + 0.3 × Main color proportion + 0.3 × (1 - Color distribution variance / Maximum variance), where the entropy value ranges from 0 to 1, the main color proportion is the proportion of buildings in the block that meet the main color requirements, and the maximum variance is the preset upper limit of the color distribution variance.

[0011] The training process of the random forest model includes: collecting building color data and corresponding expert ratings for different cities and different types of blocks as training samples, with a sample size of no less than 500 sets; dividing the samples into training set and validation set in a 7:3 ratio, with the training set used for model parameter optimization and the validation set used for model accuracy verification; and stopping training when the model prediction accuracy is higher than 90%.

[0012] A device for evaluating the color harmony of buildings in urban blocks, applied to the aforementioned method for evaluating the color harmony of buildings in urban blocks. The method and apparatus for evaluating the color harmony of urban street buildings of the present invention have the following beneficial effects: 1. Achieve quantitative and objective evaluation: Construct a multi-dimensional evaluation index system, combine subjective and objective weights, and achieve harmonious quantitative scoring through fuzzy comprehensive evaluation and machine learning models to avoid the limitations of human subjective judgment and make the evaluation results more convincing.

[0013] 2. Improve the comprehensiveness and accuracy of the assessment: Simultaneously consider the color suitability of individual buildings, the coordination of adjacent buildings, the overall balance of the block and the fit with the planning, covering the core influencing factors of block color harmony; adopt multi-source data collection and advanced algorithms (such as Mask R-CNN, random forest) to improve data quality and assessment accuracy.

[0014] 3. Improve assessment efficiency and practicality: By using drone aerial photography and automated processing technology, the workload of manual labor is greatly reduced, enabling rapid assessment of large-area blocks; the output assessment report includes specific location of discordant areas and optimization suggestions, which can directly guide urban block color planning and renovation projects.

[0015] It has good scalability and adaptability: it supports the assessment needs of different types of blocks (such as historical blocks, commercial blocks, and residential blocks), and can be adapted to different scenarios by adjusting parameters (such as distance thresholds and weight coefficients); the device module design is modular, which facilitates later upgrades and functional expansion. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0017] Figure 1 This is a flowchart of the urban street building color harmony evaluation method according to the first embodiment of the present invention. Detailed Implementation

[0018] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the present invention.

[0019] The first embodiment of this application is as follows: Please see Figure 1 ,in, Figure 1 This is a flowchart of the urban street building color harmony assessment method according to the first embodiment of the present invention. The present invention provides an urban street building color harmony assessment method and device, which realizes the objective and quantitative assessment of building color harmony, improves assessment efficiency and accuracy, provides a scientific basis for urban street color planning and renovation, and solves the technical problems of traditional assessment relying on subjective judgment and lacking unified standards.

[0020] This method comprises six core steps: data acquisition and preprocessing, feature extraction, evaluation index calculation, weight determination, harmony evaluation, and result output, as detailed below: Step 1: Data Acquisition and Preprocessing The project collects building color data, building spatial location data, and street color planning constraint data for the target street. Building color data is obtained through a combination of drone aerial photography, high-definition ground photography, and building completion archives to ensure data comprehensiveness and accuracy. Building spatial location data is obtained through a GIS system and includes building coordinates, floor area, and floor information. Street color planning constraint data includes the planned main color tone, prohibited colors, and color saturation limits.

[0021] The collected building color data was converted to the CIELab color space, which is more in line with the characteristics of human vision and facilitates the calculation of color differences. Median filtering was used to remove noise points in the images to avoid interfering with the extraction of color parameters. The MaskR-CNN image segmentation algorithm was used to extract color sample sets for individual buildings based on building outline features. The number of color samples for each building was no less than 50, covering different areas of the main facade of the building.

[0022] Step 2: Feature Extraction Individual building color characteristics: For each building's color sample set, calculate the mean hue angle (H), mean lightness (L), mean saturation (S), and color distribution variance (σ). The hue angle ranges from 0 to 360°, the lightness and saturation range from 0 to 100, and the color distribution variance reflects the uniformity of the building's own color.

[0023] Color association characteristics of adjacent buildings: Based on the spatial location data of buildings, a distance threshold (5-30m) is set to filter out the adjacent buildings of each building to form adjacent building pairs; the hue angle difference (ΔH), lightness difference (ΔL), and saturation difference (ΔS) of each pair of adjacent buildings are calculated, and the color similarity (Sim) is calculated based on the Euclidean distance in the CIELab space. The similarity range is 0-1, and the closer it is to 1, the more harmonious the colors are.

[0024] Overall color distribution characteristics of the block: Statistically analyze the distribution frequency of hue intervals (6 intervals such as 0-60°, 60-120°, etc.), lightness intervals (5 intervals such as 0-20, 20-40, etc.), and saturation intervals (5 intervals such as 0-20, 20-40, etc.) of all building colors within the block; calculate the color diversity entropy value (Ent), the larger the entropy value, the richer the color variety; determine the main color tone of the block (the color corresponding to the hue interval with the highest distribution frequency), and calculate the proportion of the main color tone (the ratio of the number of buildings that meet the main color tone requirements to the total number of buildings).

[0025] Step 3: Calculation of evaluation indicators A four-dimensional evaluation indicator system is constructed, and the definitions and calculation methods of each indicator are as follows: Color suitability (C1): Assess the degree of compatibility between the color of a single building and the main color scheme of the block plan. The calculation method is as described in claim 4, and the value range is 0-1. The closer to 1, the better the suitability.

[0026] Adjacent Harmony (C2): This assesses the degree of color harmony between adjacent buildings. It is calculated using a weighted average method, with the weight being the reciprocal of the distance between adjacent building pairs (the closer the distance, the greater the weight). The formula is: C2=Σ(Simᵢⱼ×ωᵢⱼ) / Σωᵢⱼ, where Simᵢⱼ is the color similarity of the i-th and j-th adjacent building pairs, and ωᵢⱼ is the corresponding weight, with a value range of 0-1.

[0027] Overall balance (C3): Evaluates the balance between diversity and uniformity of the overall color of the block. The quantitative formula is as described in claim 5, with a value range of 0-1. The closer to 1, the better the balance.

[0028] Planning Compliance (C4): Assess the degree to which the colors of buildings in the block comply with planning constraints. The calculation method is: C4 = 0.6 × Percentage of Compliant Buildings + 0.4 × (1 - Average Color Deviation). The percentage of compliant buildings is the ratio of the number of buildings that comply with planning constraints to the total number of buildings. The average color deviation is the normalized value of the average deviation of all building colors from planning requirements, with a value range of 0-1.

[0029] Step 4: Determine the weights The comprehensive weights are determined by combining the Analytic Hierarchy Process (AHP) with the entropy weight method. Subjective weight (W) s ): Construct a hierarchical model, invite 5-10 experts in urban planning and color design to compare the importance of each indicator pairwise, construct a judgment matrix, and obtain subjective weights after passing the consistency test.

[0030] Objective weight (W) o Based on the indicator values ​​of multiple sets of sample data, the information entropy of each indicator is calculated, and the objective weight is determined according to the entropy value. The smaller the entropy value, the higher the indicator's distinguishability and the greater its weight.

[0031] Overall Weight (W): A linear weighted fusion is used, with the formula W = α × W s +(1-α)×W o α is the subjective weight coefficient, which ranges from 0.3 to 0.7 and is adjusted according to the evaluation scenario. The default value is 0.5.

[0032] Step 5: Harmony Assessment Indicator normalization: The min-max normalization method is used to convert the values ​​of each indicator to the 0-1 range to eliminate the influence of the unit.

[0033] Fuzzy comprehensive evaluation: Construct a fuzzy evaluation matrix, determine the evaluation set (highly harmonious, relatively harmonious, generally harmonious, relatively disharmonious, disharmonious), and calculate the preliminary harmony score (0-100 points) through fuzzy transformation.

[0034] Model Refinement: The initial scores are refined using a trained random forest model. This model takes normalized index values ​​as input and expert-calibrated harmony scores as output. The parameters are optimized through sample training to improve the accuracy of the evaluation.

[0035] The harmony level is divided into five levels based on the final score: 90-100 points is highly harmonious, 80-89 points is relatively harmonious, 70-79 points is generally harmonious, 60-69 points is relatively disharmonious, and below 60 points is disharmonious.

[0036] Step 6: Output Results Generate an evaluation report containing the following: basic information about the neighborhood, data collection instructions, scores for each evaluation indicator, final harmony score and grade, harmony heat map, specific locations and causes of disharmonious areas, and suggestions for optimizing and adjusting building colors (such as adjusting the specific range of hue, brightness, and saturation).

[0037] The urban street building color harmony assessment device consists of 9 functional modules, and the functions of each module are as follows: Data acquisition module: includes a drone aerial photography unit (equipped with an RGB high-definition camera and GPS positioning module, supporting automatic flight path planning), a ground photography unit (equipped with a color calibration card to ensure color accuracy), an archive data interface (connecting to the building completion archive system), and a GIS data interface (acquiring building spatial location data), realizing the collaborative acquisition of multi-source data.

[0038] The preprocessing module is responsible for color space conversion, noise removal, and building color sample extraction. It uses a median filtering algorithm to remove image noise and a Mask R-CNN algorithm to segment building outlines and acquire color samples.

[0039] Feature extraction module: includes a color parameter calculation unit (calculates the color features of a single building), a spatial correlation analysis unit (analyzes the color correlation between adjacent buildings), and a distribution statistics unit (statistics the overall color distribution of the block), outputting three types of feature data.

[0040] Indicator Calculation Module: Calculates four-dimensional evaluation indicator values ​​based on feature data, with built-in indicator calculation models and parameter thresholds.

[0041] Weight determination module: Automates the operation of the analytic hierarchy process, entropy weighting method and weighted fusion algorithm, and outputs the comprehensive weight of each indicator.

[0042] Harmony evaluation module: includes fuzzy comprehensive evaluation unit and random forest correction unit, where the random forest model has been pre-trained and supports online evaluation and model update.

[0043] Results output module: Provides a visual interface to display assessment results in the form of heatmaps, tables, text, etc., supports the export of assessment reports (formats include PDF, Word, Excel) and data interface sharing (supports integration with urban planning management systems). The urban street building color harmony assessment method and apparatus of this embodiment realizes the objective and quantitative assessment of building color harmony, improves assessment efficiency and accuracy, provides a scientific basis for urban street color planning and renovation, and solves the technical problems of traditional assessment relying on subjective judgment and lacking unified standards.

[0044] The above-disclosed embodiments are merely one or more preferred embodiments of this application and should not be construed as limiting the scope of this application. Those skilled in the art can understand that all or part of the processes for implementing the above embodiments and equivalent changes made in accordance with the claims of this application still fall within the scope of this application.

Claims

1. A method for evaluating the color harmony of buildings in urban blocks, characterized in that, Includes the following steps; Data Acquisition and Preprocessing: Collect building color data, building spatial location data, and street color planning constraint data of the target city blocks; convert the building color data to the CIELab color space, remove image noise through median filtering, and extract color sample sets of individual buildings based on image segmentation technology; Feature extraction: Single building color features: Calculate the mean hue angle, mean lightness, mean saturation, and color distribution variance for each building color sample set; Adjacent building color association features: Based on the building spatial location data, determine adjacent building pairs within a distance threshold, and calculate the hue angle difference, lightness difference, saturation difference, and color similarity for each pair of adjacent buildings; Overall street color distribution features: Statistically analyze the hue interval distribution frequency, lightness interval distribution frequency, and saturation interval distribution frequency of building colors within the street, and calculate the color diversity entropy value and the proportion of the dominant color tone; Evaluation index calculation: c. Construct a multi-dimensional evaluation index system, including: Color suitability index: Based on the street color planning constraint data, calculate the degree of fit between the color of a single building and the main color of the plan; Adjacent coordination index: Based on the color difference value and color similarity of adjacent buildings, calculate the overall adjacent coordination of the street using a weighted average method; Overall balance index: Combine color diversity entropy value, main color proportion, and color distribution variance to construct a quantitative formula for balance measurement; Planning fit index: Statistically calculate the proportion of buildings that meet the street color planning constraints and the degree of color deviation. Weight determination: Subjective weights of indicators are determined using the analytic hierarchy process (AHP), objective weights are determined using the entropy weight method, and a comprehensive weight is obtained based on linear weighted fusion. Harmony assessment: After normalizing each assessment indicator, it is substituted into the fuzzy comprehensive evaluation model to obtain a preliminary harmony score. The preliminary score is then corrected using a trained random forest model to output the final harmony level and key influencing factors. Output results: Generate an evaluation report, including harmony score, level classification, location of disharmonious areas, and suggestions for color optimization and adjustment.

2. The method for evaluating the color harmony of urban street buildings as described in claim 1, characterized in that, The building color data is obtained by combining drone aerial images, high-definition ground images, and building completion color archives. The resolution of the aerial images is no less than 0.1m, and the ground images cover the main facades of all buildings in the block.

3. The method for evaluating the color harmony of urban street buildings as described in claim 2, characterized in that, The threshold for the distance between adjacent buildings is determined by the street network density, building volume ratio and planning specifications, and the value ranges from 5 to 30m. When the street building density is greater than 0.8, the threshold is 5 to 15m. When the building density is less than 0.3, the threshold is set at 20-30m.

4. The method for evaluating the color harmony of urban street buildings as described in claim 3, characterized in that, The color suitability index is calculated as follows: Let the hue angle range of the main color scheme be [H1,H2], the brightness range be [L1,L2], and the saturation range be [S1,S2]. Let the hue angle of a single building color be H, the brightness be L, and the saturation be S. If H∈[H1,H2] and L∈[L1,L2] and S∈[S1,S2], then the suitability score is 1; otherwise, the score is 1-[|H-(H1+H2) / 2| / (180)+|L-(L1+L2) / 2| / 100+|S-(S1+S2) / 2| / 100] / 3.

5. The method for evaluating the color harmony of urban street buildings as described in claim 4, characterized in that, The quantitative formula for the overall balance index is: Balance = 0.4 × (1 - |Entropy value - 0.6| / 0.6) + 0.3 × Main color proportion + 0.3 × (1 - Color distribution variance / Maximum variance), where the entropy value ranges from 0 to 1, the main color proportion is the proportion of buildings in the block that meet the main color requirements, and the maximum variance is the preset upper limit of the color distribution variance.

6. The method for evaluating the color harmony of urban street buildings as described in claim 5, characterized in that, The training process of the random forest model includes: collecting building color data and corresponding expert ratings for different cities and different types of blocks as training samples, with a sample size of no less than 500 sets; dividing the samples into training set and validation set in a 7:3 ratio, with the training set used for model parameter optimization and the validation set used for model accuracy verification; and stopping training when the model prediction accuracy is higher than 90%.

7. An urban block building color harmony assessment device, applied to the urban block building color harmony assessment method as described in claim 1.

Images