Mockup-based fair-faced concrete digital evaluation method and system

Abstract

The invention provides a digital evaluation method and system for fair-faced concrete based on Mockup, and belongs to the field of building construction. According to the technical scheme, the method comprises the steps of collecting a surface image of a bare concrete Mock sample plate, extracting a reference Lab color value and a reference texture feature from the collected sample plate image, and constructing an evaluation reference model; collecting a surface image of the to-be-evaluated bare concrete member, and generating image data; performing defect area identification on the image data, extracting an effective image area, and extracting a Lab color value to be evaluated and a texture feature to be evaluated from the effective image area; generating a comparison result based on the evaluation reference model; and based on the comparison result, generating evaluation output information. The method has the beneficial effects that the color and texture double-feature model is constructed, a defect identification and shielding mechanism is introduced, and a structured comparison algorithm and an output system are adopted, so that the automation of the whole process from data acquisition, feature extraction, defect avoidance to intelligent comparison and evaluation output is realized.

Description

Technical Field

[0001] The present invention relates to the field of building construction, and in particular to a Mockup-based digital evaluation method and system for plain concrete. Background Art

[0002] With the continuous improvement of urban building quality, exposed concrete, due to its natural texture and minimalist aesthetic, has become one of the important materials for modern building facade decoration. To ensure consistency and quality control in large-scale construction, the construction industry has widely introduced a "mockup model first" mechanism. By first producing and accepting standard models, subsequent component construction is guided by this mechanism. However, in actual application, how to compare and evaluate the surface consistency between mockup models and actual components in a quantifiable and traceable manner remains a key issue in project quality control. Traditional evaluation methods mainly rely on manual visual inspection, which is easily affected by observation angle, lighting environment, and individual experience. It is highly subjective and difficult to achieve stable and reliable quality judgment in large-scale construction. Therefore, digital evaluation methods that integrate computer vision and image recognition algorithms have gradually become a research hotspot.

[0003] However, existing digital image processing methods are mostly limited to color distribution analysis or single texture quantification, and a dedicated technical framework for the evaluation of bare concrete surfaces has not yet been formed. On the one hand, most existing technologies directly calculate color or texture features based on the entire image, lacking a mechanism for identifying and shielding surface defect areas, resulting in abnormal areas interfering with the overall evaluation accuracy; on the other hand, evaluation indicators often use simple color difference thresholds or rough texture matching algorithms, which cannot truly reflect the microscopic differences of the bare concrete surface at the visual perception level. In addition, most current systems lack a method for organizing results with structured output, making it difficult to achieve standardized use in construction supervision, defect tracing, and quality archiving. Therefore, how to construct a digital evaluation method that combines the dual dimensions of color and texture, has a defect tolerance mechanism, and can output structured results has become a technical bottleneck that needs to be broken through. Summary of the Invention

[0004] The purpose of the present invention is to provide a Mockup-based digital assessment method and system for bare concrete, which automates the entire process from data acquisition, feature extraction, defect avoidance to intelligent comparison and assessment output by constructing a dual-feature model of color and texture, introducing a defect recognition and shielding mechanism, and adopting a structured comparison algorithm and output system.

[0005] The present invention is achieved by the following measures: A digital evaluation method based on bare concrete mockup, characterized by comprising: Collect surface images of bare concrete mockup samples, extract benchmark Lab color values ​​and benchmark texture features from the collected sample images, and build an evaluation benchmark model; Collect surface images of the bare concrete component to be evaluated and generate image data; Perform defect area recognition on the image data, generate a mask image, use the mask image to mask the defect area to extract the valid image area, and extract the Lab color value to be evaluated and the texture feature to be evaluated from the valid image area; Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated to generate a comparison result.

[0006] The specific features of the present invention also include: Collect surface images of bare concrete mockup samples, and extract benchmark Lab color values ​​and benchmark texture features from the collected sample images, including: The surface image of the bare concrete mockup sample is acquired through an image acquisition device, and the surface image is preprocessed. After the preprocessing is completed, the surface image is converted into the CIE Lab color space to extract the benchmark Lab color value. At the same time, the benchmark texture features of the image are extracted through the gray-level co-occurrence matrix, and finally the benchmark Lab color value and benchmark texture features are obtained for constructing the evaluation benchmark model.

[0007] Constructing the evaluation benchmark model includes: constructing corresponding color feature datasets and texture feature datasets based on the benchmark Lab color values ​​and benchmark texture features extracted from the sample image; The color feature dataset and texture feature dataset are statistically processed and standardized respectively to obtain color description vectors and texture description vectors. The color description vectors and texture description vectors are combined in a fixed format to construct a feature matrix for evaluating the benchmark model. The construction of the evaluation benchmark model is completed when the preset trigger conditions are met.

[0008] The surface image of the bare concrete component to be evaluated is collected to generate image data, including: obtaining an image sequence covering the surface of the bare concrete component through an image acquisition device, and performing format conversion, lighting adjustment, and noise suppression operations on the image sequence after collection to obtain structured image data.

[0009] Performing defect area recognition on the image data to generate a mask image, and using the mask image to mask the defect area to extract the effective image area includes: performing a pixel-level defect recognition operation on the image data to generate a mask image representing the defect area and the non-defect area; According to the information marked by the mask image, the corresponding defective area in the image data is masked, the pixel content corresponding to the non-defective area in the image data is retained, and the effective image area is extracted.

[0010] Extracting the to-be-evaluated Lab color value and the to-be-evaluated texture feature from the effective image region includes: performing image preprocessing, image color space conversion, and texture feature calculation operations on the image frame in the effective image region to obtain the to-be-evaluated Lab color value and the to-be-evaluated texture feature.

[0011] Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated. The comparison results include: The color description vector and texture description vector in the evaluation benchmark model are called and calculated respectively with the Lab color value to be evaluated and the texture feature to be evaluated extracted from the effective image area; when performing texture feature comparison, the texture feature to be evaluated and the benchmark texture feature are matched based on the preset similarity calculation method, and the texture similarity score is output; when performing color value comparison, the standard color difference formula is used to compare the Lab color value to be evaluated and the benchmark Lab color value, and the color difference value is output; the texture similarity score and the color difference value are unified and the comparison result is output; Based on the comparison results, generating evaluation output information related to the to-be-evaluated plain concrete component includes: performing a classification judgment operation based on the comparison results, and generating structured evaluation output information according to preset evaluation rules.

[0012] A digital evaluation system based on bare concrete mockup, characterized by comprising: Evaluation benchmark model construction module: This module collects surface images of bare concrete mockup samples, extracts benchmark Lab color values ​​and benchmark texture features from the collected sample images, and constructs an evaluation benchmark model. Image acquisition module for evaluation: collects surface images of the bare concrete components to be evaluated and generates image data; Image preprocessing and effective area extraction module: This module performs defect area identification on the image data, generates a mask image, and uses the mask image to mask the defect area to extract the effective image area. The Lab color value and texture features to be evaluated are extracted from the effective image area. Texture and color comparison calculation module: Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated to generate the comparison result; Evaluation output generation module: Based on the comparison results, it generates evaluation output information related to the bare concrete component to be evaluated.

[0013] The beneficial effects of the present invention are as follows: by constructing a color and texture dual feature model, introducing a defect recognition and shielding mechanism, and adopting a structured comparison algorithm and output system, the present invention realizes the automation of the entire process from data collection, feature extraction, defect avoidance to intelligent comparison and evaluation output; Compared with existing traditional methods that mainly rely on manual visual inspection or single-dimensional comparison, the present invention not only significantly improves the objectivity and consistency of the evaluation, but also enhances the quantifiability, repeatability and traceability of the comparison process. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is an overall flow chart of a digital evaluation method based on bare concrete mockup provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to clearly illustrate the technical features of this solution, this solution is described below through specific implementation methods.

[0016] Example 1 See also Figure 1 A digital evaluation method based on bare concrete mockup is characterized by comprising: Step S1, collecting a surface image of a bare concrete mockup sample, and extracting a reference Lab color value and a reference texture feature from the collected sample image, including: The surface image of the bare concrete mockup sample is acquired through an image acquisition device, and the surface image is preprocessed. After the preprocessing is completed, the surface image is converted into the CIE Lab color space to extract the benchmark Lab color value. At the same time, the benchmark texture features of the image are extracted through the gray-level co-occurrence matrix, and finally the benchmark Lab color value and benchmark texture features are obtained for constructing the evaluation benchmark model.

[0017] The image acquisition device is used to obtain the surface image of the bare concrete mockup sample. It preferably uses a high-resolution industrial camera with an industrial-grade image sensor, combined with an industrial lens and a constant-current-driven LED array light source to provide a stable and uniform imaging environment. At the same time, to ensure the consistency of the captured image, the image acquisition device is fixedly installed through a structural bracket or track platform, the sampling angle and sampling distance remain constant, and the triggering timing is controlled by the acquisition control module.

[0018] The acquisition control module is used to control the start time of image acquisition. It can be implemented based on an embedded system, such as a Jetson Nano, Raspberry Pi, or STM32 microcontroller. This module monitors the exposure stability, brightness mean, and sampling alignment status of the image sensor in real time. If multiple consecutive frames of images meet a preset trigger threshold (such as the brightness standard deviation is less than a set value, the image center of gravity drift is less than 1 pixel, etc.), the acquisition instruction is automatically sent to ensure that the sampled images are comparable and representative. The acquired image is input into the image preprocessing module, which performs image denoising (using Gaussian filtering), white balance correction, and illumination normalization in sequence to enhance the image's color restoration and boundary information expression. After the preprocessing is completed, the color space conversion module is called to perform color space mapping on the image, converting the image from the RGB color space to the CIE Lab color space. The conversion process can be based on the CIE standard color management process, using the intermediate XYZ space as a transition, and combining the D65 reference white point for nonlinear transformation to extract the three components of L (brightness), a (red-green axis), and b (yellow-blue axis) of each pixel in the image, thereby constructing the baseline Lab color value distribution of the entire image.

[0019] In addition, to extract benchmark texture features, the preprocessed image is first converted into a grayscale image, and then a local sliding window (e.g., 7×7 pixels) is set on the image to construct a gray-level co-occurrence matrix (GLCM). Within each window, the co-occurrence frequencies of different grayscale pairs are counted according to fixed spatial directions (0°, 45°, 90°, 135°), and four texture statistical indices are calculated, including contrast, energy, entropy, and correlation. These indices reflect the differences and structures of the image in spatial grayscale changes. Finally, the indicator set is standardized and aggregated into a unified texture description vector according to the window movement strategy, which is the benchmark texture feature.

[0020] These benchmark Lab color values ​​and texture features serve as foundational data for the subsequent construction of the benchmark model, providing an objective quantitative basis for the digital similarity assessment between the mockup template and the actual component. This combined approach exhibits strong stability and engineering adaptability, effectively improving the reliability and consistency of bare concrete components during the digital comparison process.

[0021] Constructing the evaluation benchmark model includes: constructing corresponding color feature datasets and texture feature datasets based on the benchmark Lab color values ​​and benchmark texture features extracted from the sample image; The color feature dataset and texture feature dataset are statistically processed and standardized respectively to obtain color description vectors and texture description vectors. The color description vectors and texture description vectors are combined in a fixed format to construct a feature matrix for evaluating the benchmark model. The construction of the evaluation benchmark model is completed when the preset trigger conditions are met.

[0022] The evaluation benchmark model is constructed based on benchmark Lab color values ​​and benchmark texture features extracted from bare concrete mockup sample images. A color benchmark dataset and a texture benchmark dataset are constructed, respectively. To improve the model's stability and representativeness, a cluster analysis is performed on the Lab values ​​of each image pixel in the color benchmark dataset, preferably using a K-means or Mean Shift clustering algorithm to extract the primary color distribution areas and form a clustered color feature center. This clustering process not only eliminates background interference but also more accurately captures the dominant color areas of the bare concrete surface. Furthermore, statistical parameters such as color mean, color standard deviation, and primary color distribution density are calculated for the clustered samples and combined to construct a color description vector, which reflects the typical color distribution characteristics of the mockup sample in Lab space.

[0023] To construct the texture benchmark dataset, we used the aforementioned texture statistics extracted from the gray-level co-occurrence matrix as the original feature dimensions, including image contrast, entropy, energy, and correlation. To enhance the model's sensitivity to surface structural differences between samples, we introduced feature selection strategies (such as variance screening or mutual information sorting) to eliminate feature components with poor stability or weak correlation. Furthermore, we used the Z-score normalization method to unify the numerical scales of different statistical quantities. Ultimately, we generated a stable and quantifiable texture description vector.

[0024] The color and texture description vectors are concatenated according to fixed dimensionality rules to construct a feature matrix, which serves as the core parameter structure of the benchmark evaluation model. This dimensional combination ensures that color and texture information are equally weighted in the model and can be evaluated independently, avoiding biased judgments caused by information imbalance. The feature matrix is ​​highly structured and reconfigurable, making it suitable for subsequent similarity calculations and color comparisons.

[0025] The generation of the evaluation benchmark model is automatically completed by the model training control module. This module implements modeling control through dynamic monitoring of the system's internal metrological parameters. For example, when the standard deviation of the color distribution drops to a stable threshold and the correlation between texture features remains within an acceptable range, the module determines that the feature extraction process has reached convergence and automatically triggers the model construction process. This mechanism effectively avoids human intervention and model instability, ensuring that the generated model is highly representative and reusable in engineering applications.

[0026] Step S2, collecting surface images of the bare concrete component to be evaluated, and generating image data includes: obtaining an image sequence covering the surface of the bare concrete component through an image acquisition device, and performing format conversion, lighting adjustment, and noise suppression operations on the image sequence after acquisition to obtain structured image data.

[0027] Image data collection for the bare concrete components to be assessed is performed using a mobile image acquisition device. This device integrates a high-resolution industrial camera, a uniform-brightness LED array light source, and a three-dimensional attitude control system to accurately image the surfaces of complex components in various engineering scenarios. During the image acquisition process, the acquisition device, guided by a control program, performs multi-angle scanning and sampling. This involves periodically adjusting the viewing angle and shooting angle along a set path to ensure that the surface texture and color details of the component are fully captured. After each image acquisition, the system automatically records the acquisition time (timestamp) and spatial position information (which can be obtained through an IMU or visual odometry) and binds it to the corresponding image frame to form a time-series image sequence, providing the necessary foundation for subsequent regional positioning and data alignment operations. The captured raw image frames are then input into the image preprocessing module. This module first converts different image formats (such as RAW and JPEG) into a structured intermediate format (such as TIFF or PNG uncompressed format) to ensure data consistency. Secondly, due to the problem of uneven lighting that is common in actual construction sites, an illumination normalization operation is introduced. Based on the Retinex theory, the global brightness of each image frame is adjusted to eliminate reflection interference caused by lighting changes. In addition, to address possible edge blur or dust contamination on the surface of concrete components, an edge enhancement algorithm (such as Laplacian enhancement) is used to highlight the structural boundaries. At the same time, a combination of median filtering and bilateral filtering algorithms is used to suppress high-frequency noise, improving overall image clarity and the robustness of subsequent processing. After preprocessing is completed, the image data is organized into structured data by the data generation module. The structured format includes not only the processed image frame data itself, but also metadata such as the spatial position information corresponding to the image, acquisition parameters (such as exposure time, focal length setting), shooting time and component number. This organizational form can support subsequent targeted operations such as defect area identification and effective area extraction, ensuring that the analysis is based on images with consistent temporal and spatial distribution, effectively improving recognition accuracy and evaluation stability.

[0028] Step S3, performing defect area recognition on the image data, generating a mask image, and using the mask image to shield the defect area to extract the valid image area includes: performing a pixel-level defect recognition operation on the image data, generating a mask image representing the defect area and the non-defect area; First, for each frame of the original image data, input image Perform semantic pixel classification processing and call the trained image segmentation model (such as DeepLabV3+ or U-Net) for each pixel in the image The corresponding image content is used to predict semantic labels to determine whether it belongs to the defect area; The segmentation model is based on convolutional neural networks, integrates multi-scale receptive field structures, and combines context information to perform multi-level extraction of input image features. Its classification output is a logical tensor. , represents the predicted probability that the pixel is a defect; The segmentation model adopts an encoding-decoding structure. DeepLabV3+ introduces atrous convolution (AtrousConvolution) and atrous spatial pyramid pooling (ASPP) to capture multi-scale contextual features, which is suitable for large-area continuous texture analysis; U-Net adopts a symmetrical structure and skip connection mechanism, with good detail restoration capabilities, and is particularly suitable for identifying fine-grained defect features on the surface of plain concrete. The model training data comes from plain concrete image samples collected at the construction site. The image content covers multiple typical defect types such as cracks, honeycombs, pockmarks, pollution, and color spots. All training images are annotated with pixel-level defects by an expert team to construct a high-quality semantic segmentation dataset.

[0029] The training adopts supervised learning method, and the loss function uses Class-Balanced Cross Entropy Loss to ensure that the model does not ignore rare defect types when the sample distribution is uneven; After training, the model generates a set of probability values ​​for each pixel in the input image. , classified as the predicted category with the maximum probability Finally, all defect classes are assigned a value of 0 in the mask image, and non-defect classes are assigned a value of 1 to generate a binary mask image for masking processing. . Based on this probability map , by setting the classification threshold , perform binarization operation to obtain the mask map ,Right now:

[0030] in , when the pixel probability is less than the threshold When the value is greater than or equal to the threshold, it is judged as a non-defective area (retained); when the value is greater than or equal to the threshold, it is judged as a defective area (shielded).

[0031] Then, using the generated mask map The original image frame is processed by pixel-by-pixel logical operations Perform masking operations to extract non-defective areas and generate valid image areas , which is defined as:

[0032] in Represents the pixel value matrix retained after masking. The unmasked pixels retain their original values ​​in the original image, and the masked areas are set to zero or background fill values. Furthermore, in order to monitor the image quality and decide whether to perform image recapture, the system performs defect ratio calculation on the mask image, defect area ratio The definition is as follows:

[0033] Where D is the total number of defective pixels, is the total number of pixels in the image, and the system is based on the defect area ratio R and the set threshold For comparison:

[0034] in, It is the Boolean logic state of whether to start image resampling. This is the system's preset recapture threshold. Image recapture specifically involves moving the image acquisition position, optimizing the shooting angle, or adjusting the image resolution setting to obtain new image data with less defect interference, thereby ensuring the accuracy of subsequent evaluation and analysis and the reliability of the data foundation. when When the image quality is too poor (e.g. 40%), it indicates that there are too many defects in the current image, which may affect the objectivity and accuracy of the assessment. The system automatically triggers the image recapture module, which can control the image acquisition device to perform one or a combination of the following operations: changing the acquisition angle, moving the camera position, adjusting the focal length or exposure parameters, enabling high-resolution acquisition mode, etc., to generate clearer new image data with less defect interference for subsequent processing.

[0035] In summary, this step achieves structured, high-quality extraction of the effective evaluation area through pixel-level defect recognition of the semantic segmentation model and mask extraction of the mask mechanism, while clearly eliminating image interference factors. It establishes a stable and reliable image basic data source for subsequent color value and texture feature calculations, and enhances the repeatability and engineering applicability of the system evaluation results.

[0036] Extracting the to-be-evaluated Lab color value and the to-be-evaluated texture feature from the effective image region includes: performing image preprocessing, image color space conversion, and texture feature calculation operations on the image frame in the effective image region to obtain the to-be-evaluated Lab color value and the to-be-evaluated texture feature.

[0037] After defect masking and effective area extraction are completed, the effective image area enters the image analysis module for further data extraction processing: The image analysis module first performs image preprocessing operations, including the following steps: Brightness equalization processing: Based on the adaptive histogram equalization (CLAHE) method, the image brightness distribution is adjusted to reduce the image color difference caused by local uneven illumination; Color correction processing: A mapping correction method based on a color reference card is used to correct color shift and eliminate color deviation caused by changes in lighting conditions during shooting; Edge enhancement processing: Use the Laplace or Sobel operator to improve the visual features of the image edge to highlight the boundary information of the surface structure of the bare concrete component.

[0038] After preprocessing, the image data undergoes color space conversion from RGB to CIE Lab. This conversion is achieved through nonlinear mapping defined by the CIE standard, resulting in a color representation that better aligns with human visual perception. The L (lightness channel), a (green-red axis), and b (blue-yellow axis) channel values ​​for each pixel are extracted and statistically calculated to form the Lab color value to be evaluated, serving as the basic parameter for objectively quantifying the color consistency of the concrete surface.

[0039] At the same time, the image analysis module performs texture feature calculation operations on the image: First, the preprocessed color image is converted into a grayscale image using a weighted method. The grayscale image is then used to construct a gray-level co-occurrence matrix (GLCM). This matrix describes the grayscale relationship between image pixels using specific distances (e.g., 1 pixel) and directions (e.g., 0°, 45°, 90°, and 135°). Then, four texture statistical indices, namely energy, contrast, entropy and correlation, are calculated based on GLCM to quantify the texture feature distribution of the image surface. The obtained statistical indices are integrated into the texture feature vector to be evaluated.

[0040] Finally, the Lab color values ​​to be evaluated and the texture feature vectors to be evaluated are combined into a complete structured data set according to a preset unified structural format, and clearly associated with the valid image area number and the identification of the bare concrete component to be evaluated. This standardized organizational storage mode can effectively ensure the traceability and repeatable analysis of the data, and provide an objective and standardized data input source for subsequent texture similarity calculations and color comparison operations, thereby improving the overall accuracy and robustness of the digital evaluation process for bare concrete construction quality, and completely solving the fundamental technical problems of traditional manual evaluation methods, which are highly subjective, poorly repeatable, and susceptible to human interference.

[0041] Step S4: Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the Lab color value to be evaluated in the effective image area is calculated with the benchmark Lab color value of the sample image to generate a comparison result. The color description vector and texture description vector in the evaluation benchmark model are called and calculated respectively with the Lab color value to be evaluated and the texture feature to be evaluated extracted from the effective image area; when performing texture feature comparison, the texture feature to be evaluated and the benchmark texture feature are matched based on the preset similarity calculation method, and the texture similarity score is output; when performing color value comparison, the standard color difference formula is used to compare the Lab color value to be evaluated and the benchmark Lab color value, and the color difference value is output; the texture similarity score and the color difference value are unified and the comparison result is output; In the comparison phase, first, based on the component identification of the component to be evaluated, the corresponding benchmark Lab color value and benchmark texture feature vector are called from the feature matrix of the evaluation benchmark model as a reference; The specific calculation process of texture feature comparison is as follows: define the texture feature vector to be evaluated as , define the reference texture feature vector as ,The vector represents four standardized statistical indicators of contrast, entropy, energy and correlation.

[0042] The Euclidean distance method is used to calculate the difference between the two, and the expression is:

[0043] After obtaining the Euclidean distance of texture features, it is further converted into a standardized texture similarity score through Gaussian mapping. The calculation formula is:

[0044] in, is the texture feature vector of the area to be evaluated, is the texture feature vector of the reference area, 、 、 、 are the contrast, entropy, energy and correlation index values ​​of the area to be evaluated, 、 、 、 are the index values ​​corresponding to the benchmark area, is the Euclidean distance between texture feature vectors, is the distance sensitivity adjustment coefficient, obtained through field experimental calibration, Indicates texture similarity, with a value range of 0 to 1. The closer the value is to 1, the closer the texture of the area to be evaluated is to the reference area, reflecting a better surface quality. The specific steps for comparing and calculating the color values ​​are as follows: Define the Lab color value of the image area to be evaluated as a vector , define the Lab color value corresponding to the reference area as a vector , calculate the difference of each channel in turn:

[0045] Then, considering the difference in local color difference sensitivity of the concrete surface, a nonlinear local perception adjustment function is introduced:

[0046] Then define the interaction term between color channels to reflect the mutual influence of the hue shift of the a and b channels:

[0047] Then determine the weighted correction item of each channel color:

[0048] The final improved color difference evaluation value formula is:

[0049] in, 、 、 They are the Lab values ​​of the brightness, red and green channels, and yellow and blue channels of the area to be evaluated. 、 、 is the Lab value corresponding to the reference area, 、 、 are the channel differences corresponding to the two regions, 、 、 Represent the nonlinear local perception adjustment function of L, a, and b channels respectively, 、 、 are the local sensitivity parameters of L, a, and b channels respectively, obtained through data statistics. is the color interaction term between a and b channels, It is a stabilizing factor to prevent the denominator from approaching zero and causing calculation instability. 、 、 is the weighted correction value of channel color difference, 、 、 The normalized weights of each channel are respectively satisfied , is the channel interaction effect adjustment coefficient, is the final color difference evaluation value. The smaller the value, the closer the color of the area to be evaluated is to the reference area. After the above calculation is completed, the texture similarity score and color difference evaluation value Unified integration into a structured comparison result data format:

[0050] in, Identifies the component to be evaluated. Image regions are numbered and the integrated data is archived by binding component identification and image region number to ensure data traceability, accuracy, and consistency; The entire comparison process solves the technical problems of strong subjectivity and poor repeatability in existing concrete surface quality evaluation through the innovative application of structured data integration and quantitative calculation methods, significantly improving the accuracy and objectivity of automated evaluation of fair-faced concrete construction quality, making it suitable for a wide range of engineering practice applications.

[0051] Step S5, based on the comparison results, generating evaluation output information related to the to-be-evaluated plain concrete component, includes: performing a classification judgment operation based on the comparison results, and generating structured evaluation output information according to preset evaluation rules.

[0052] In the evaluation output, the texture similarity score is obtained Color difference evaluation value On this basis, a structured and standardized evaluation output information generation mechanism is further proposed to effectively guide the actual plain concrete construction and quality control; Specifically, the present invention designs a set of clear classification judgment and evaluation rules to achieve the precise transformation process from quantitative comparison results to actionable and implementable evaluation output information at the construction site; The specific implementation process of generating the evaluation output information is as follows: First, the present invention establishes preset evaluation rules for texture similarity scores and color difference evaluation values. These rules are calibrated based on quality standards for bare concrete construction and a large amount of historical data from field tests, demonstrating clear objectivity and applicability. The texture similarity score The classification threshold can be set to two critical values ​​of 0.90 and 0.75; The specific classification rules are defined as follows: when the texture similarity is ≥0.90, it means that the texture quality is very close to the sample and is "excellent"; when the texture similarity score is between 0.75 and 0.90, it means that the texture quality is acceptable and is "good"; when the texture similarity score is lower than 0.75, it means that the texture quality deviates greatly and is "poor", requiring treatment or rework; At the same time, the color difference evaluation value The classification thresholds can be set to two key values, 1.0 and 3.0, with specific definitions as follows: when the color difference evaluation value is less than or equal to 1.0, it indicates that there is no obvious visual color difference, which is "excellent"; when the color difference evaluation value is between 1.0 and 3.0, it indicates that there is a slight but acceptable color difference, which is "good"; when the color difference evaluation value is greater than or equal to 3.0, the color difference is obviously unacceptable, which is "poor", and measures must be taken to correct or rework. Secondly, when performing the classification judgment operation, the system automatically reads the texture similarity score and color difference evaluation value recorded in the comparison result according to the above-mentioned threshold rules, and determines the specific quality category to which the texture and color difference belong respectively. Then, according to the pre-defined preset mapping table, it automatically calls the evaluation suggestion information that strictly corresponds to the category to which it belongs. The evaluation suggestions are prepared in advance in clear and concise language, including but not limited to: "Passed the acceptance and no further processing is required", "It is recommended to make minor corrections to the site, such as local polishing or cleaning", "Obviously does not meet the standards and must be reworked or the construction plan adjusted", and other specific suggestions, so that the construction site technicians can directly understand and effectively implement them. Finally, the specific evaluation suggestions called are bound to the component identification, texture classification results, and color classification results of the component to be evaluated to form the final evaluation output information in a structured standard format.

[0053] Finally, the specific evaluation recommendation items called are bound to the component identification, texture classification results, and color classification results of the component to be evaluated to form final evaluation output information in a structured standard format; The final structured evaluation output information includes: ① the component identification of the component to be evaluated (uniquely identifying the bare concrete component to be evaluated), ② the classification result of texture similarity (excellent, good, or poor), ③ the classification result of color difference (excellent, good, or poor), and ④ specific evaluation recommendations that strictly correspond to the above classification results (such as acceptance, on-site correction, or rework), to ensure that construction site personnel can intuitively and clearly obtain detailed, targeted, and executable quality control instructions.

[0054] Example 2 A digital evaluation system based on bare concrete mockup, characterized by comprising: Evaluation benchmark model construction module: This module collects surface images of bare concrete mockup samples, extracts benchmark Lab color values ​​and benchmark texture features from the collected sample images, and constructs an evaluation benchmark model. Image acquisition module for evaluation: collects surface images of the bare concrete components to be evaluated and generates image data; Image preprocessing and effective area extraction module: This module performs defect area identification on the image data, generates a mask image, and uses the mask image to mask the defect area to extract the effective image area. The Lab color value and texture features to be evaluated are extracted from the effective image area. Texture and color comparison calculation module: Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated to generate the comparison result; Evaluation output generation module: Based on the comparison results, it generates evaluation output information related to the bare concrete component to be evaluated.

[0055] Technical features not described in the present invention can be achieved through or by adopting existing technologies and will not be described in detail here. Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention should also fall within the scope of protection of the present invention.

Claims

1. A digital evaluation method based on bare concrete mockup, characterized in that: include: Collect surface images of bare concrete mockup samples, extract benchmark Lab color values ​​and benchmark texture features from the collected sample images, and build an evaluation benchmark model; Collect surface images of the bare concrete component to be evaluated and generate image data; Perform defect area recognition on the image data, generate a mask image, and use the mask image to mask the defect area to extract the valid image area, and extract the Lab color value to be evaluated and the texture feature to be evaluated from the valid image area; Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated to generate a comparison result; Based on the comparison results, evaluation output information related to the bare concrete component to be evaluated is generated.

2. The digital evaluation method based on bare concrete mockup according to claim 1 is characterized in that: Collect surface images of bare concrete mockup samples, and extract benchmark Lab color values ​​and benchmark texture features from the collected sample images, including: The surface image of the bare concrete mockup sample is acquired through an image acquisition device, and the surface image is preprocessed. After the preprocessing is completed, the surface image is converted into the CIE Lab color space to extract the benchmark Lab color value. At the same time, the benchmark texture features of the image are extracted through the gray-level co-occurrence matrix, and finally the benchmark Lab color value and benchmark texture features are obtained for constructing the evaluation benchmark model.

3. The digital evaluation method based on bare concrete mockup according to claim 2 is characterized in that: Constructing the evaluation benchmark model includes: constructing corresponding color feature datasets and texture feature datasets based on the benchmark Lab color values ​​and benchmark texture features extracted from the sample image; The color feature dataset and texture feature dataset are statistically processed and standardized respectively to obtain color description vectors and texture description vectors. The color description vectors and texture description vectors are combined in a fixed format to construct a feature matrix for evaluating the benchmark model. The construction of the evaluation benchmark model is completed when the preset trigger conditions are met.

4. The digital evaluation method based on bare concrete mockup according to claim 3 is characterized in that: The surface image of the bare concrete component to be evaluated is collected to generate image data, including: obtaining an image sequence covering the surface of the bare concrete component through an image acquisition device, and performing format conversion, lighting adjustment, and noise suppression operations on the image sequence after collection to obtain structured image data.

5. The digital evaluation method based on bare concrete mockup according to claim 4 is characterized in that: Performing defect area recognition on the image data to generate a mask image, and using the mask image to mask the defect area to extract the valid image area includes: performing a pixel-level defect recognition operation on the image data to generate a mask image representing the defect area and the non-defect area; According to the information marked by the mask image, the corresponding defective area in the image data is masked, the pixel content corresponding to the non-defective area in the image data is retained, and the effective image area is extracted.

6. The digital evaluation method based on bare concrete mockup according to claim 5 is characterized in that: Extracting the to-be-evaluated Lab color value and the to-be-evaluated texture feature from the effective image region includes: performing image preprocessing, image color space conversion, and texture feature calculation operations on the image frame in the effective image region to obtain the to-be-evaluated Lab color value and the to-be-evaluated texture feature.

7. The digital evaluation method based on bare concrete mockup according to claim 6 is characterized in that: Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated. The comparison results include: The color description vector and texture description vector in the evaluation benchmark model are called to perform calculations with the Lab color value to be evaluated and the texture features to be evaluated extracted from the effective image area respectively; when performing texture feature comparison, the texture features to be evaluated and the benchmark texture features are matched based on the preset similarity calculation method, and the texture similarity score is output; when performing color value comparison, the standard color difference formula is used to compare the Lab color value to be evaluated and the benchmark Lab color value, and the color difference value is output; the texture similarity score and the color difference value are unified and organized, and the comparison result is output.

8. The digital evaluation method based on bare concrete mockup according to claim 7 is characterized in that: Based on the comparison results, generating evaluation output information related to the to-be-evaluated plain concrete component includes: performing a classification judgment operation based on the comparison results, and generating structured evaluation output information according to preset evaluation rules.

9. A digital evaluation system based on bare concrete mockup using the method according to any one of claims 1 to 8, characterized in that: include: Evaluation benchmark model construction module: This module collects surface images of bare concrete mockup samples, extracts benchmark Lab color values ​​and benchmark texture features from the collected sample images, and constructs an evaluation benchmark model. Image acquisition module for evaluation: collects surface images of the bare concrete components to be evaluated and generates image data; Image preprocessing and effective area extraction module: This module performs defect area identification on the image data, generates a mask image, and uses the mask image to mask the defect area to extract the effective image area. The Lab color value and texture features to be evaluated are extracted from the effective image area. Texture and color comparison calculation module: Based on the evaluation benchmark model, the texture features to be evaluated in the effective image area are compared with the benchmark texture features of the sample image for similarity, and the color difference between the Lab color value to be evaluated in the effective image area and the benchmark Lab color value of the sample image is calculated to generate the comparison result; Evaluation output generation module: Based on the comparison results, it generates evaluation output information related to the bare concrete component to be evaluated.

Images