A method and system for detecting floor wear resistance

Abstract

The application provides a floor wear resistance detection method and system, applied to the floor wear detection technical field, which reflects the wear resistance of the floor coating by obtaining the change amount of porosity before and after the floor is worn. When the porosity after wear is obtained, the method uses multi-directional light to collect the two-dimensional optical image of the surface to be measured, and aligns the image. By analyzing the light and dark change information and shadow shape information of the dark area region under different light directions, when the light and dark change information of the dark area region changes with the change of the light source position, and the shadow shape information presents a dynamic change of rotation or movement, it is judged that the floor is porous. This discrimination mechanism based on the dynamic change of light effectively solves the problem that the two-dimensional optical microscope scheme in the prior art cannot distinguish between real pores and shallow scratches, avoids misjudging scratches caused by a large amount of wear as pores, and significantly improves the accuracy of the porosity calculation after wear.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of floor wear detection, and in particular to a floor wear resistance detection method and system. BACKGROUND

[0002] In modern industrial production, the evaluation of the wear resistance of the ultra-thin functional coating on the surface of new composite floor is a key link to ensure product quality. The traditional wear test method faces challenges in evaluating such ultra-thin coatings, because the coating is easily worn out, resulting in test results that cannot truly reflect the performance of the coating itself. Even if the test parameters are adjusted to adapt to the ultra-thin coating, making the wear degree extremely slight, the traditional macroscopic measurement means will be difficult to accurately quantify due to the small change, for example, microgram-level mass loss or nanometer-level wear depth may be below the resolution of the measuring instrument, resulting in a very low signal-to-noise ratio, making it difficult to stably and accurately quantify the wear degree.

[0003] In the face of macroscopic measurement difficulties, technical personnel turn to the microscopic level and find that even extremely slight wear can significantly change the pore morphology of the coating surface, for example, increasing the average size of the pores or making the profile irregular. This provides a new way to indirectly evaluate wear resistance, that is, by quantifying the change in porosity before and after wear to reflect the wear resistance of the coating.

[0004] However, this detection method based on porosity change also faces challenges in production efficiency in actual industrial applications. In order to meet the demand for rapid detection, the technical team has tried to use three-dimensional profile measurement equipment such as confocal microscopes or white light interferometers, which can effectively distinguish pores and shallow scratches through depth threshold. However, obtaining high-precision three-dimensional data usually takes tens of minutes, which is too high in time cost and not practical. Therefore, in order to meet the production efficiency requirements, the technical team has to return to the two-dimensional optical microscope scheme that can realize second-level imaging.

[0005] In such rapidly acquired two-dimensional optical images, due to the extremely slight wear, the shallow and fine scratches produced exhibit highly similar image characteristics to real pores, for example, both may appear as dark areas, and the size and shape may also have overlaps or similarities that are difficult to distinguish in two-dimensional projection. This makes it difficult for conventional image processing algorithms, such as simple gray threshold segmentation or basic morphological operations, to effectively distinguish these shallow scratches from real pores. As a result, when calculating the porosity, the image analysis algorithm will misjudge a large number of scratches produced by wear as pores, thereby severely overestimating the surface porosity after wear. This inaccurate measurement result makes the method of evaluating wear resistance through porosity change unreliable, thus failing to meet the requirements of industrial quality inspection for data reliability.

[0006] In view of the above problems, the prior art needs to be improved. SUMMARY

[0007] In view of the above problems of the prior art, the present application provides a floor wear resistance detection method and system, aiming to solve the problems in the prior art that the traditional macroscopic measurement method cannot accurately quantify the wear degree when evaluating the wear resistance of the ultra-thin functional coating, and the porosity detection method based on two-dimensional optical images cannot effectively distinguish between real pores and shallow scratches caused by wear, resulting in inaccurate detection results.

[0008] In a first aspect, a floor wear resistance detection method is provided, the method comprising the steps of:

[0009] S1: obtaining the initial porosity of the floor before wear;

[0010] S2: obtaining two-dimensional optical images of the surface to be tested under multiple different light directions after wear, and performing alignment processing on the two-dimensional optical images to obtain aligned images;

[0011] S3: identifying dark area regions in the aligned images, and analyzing the light-dark change information and shadow shape information of the dark area regions under multiple different light directions;

[0012] S4: when the light-dark change information changes with the change of the light source position, and the shadow shape information presents a dynamic change of rotation or movement with the change of the light source position, judging that the corresponding dark area region is a floor pore;

[0013] S5: calculating the porosity of the floor after wear according to the total area of the floor pores;

[0014] S6: comparing the floor porosity after wear with the initial porosity to obtain the change amount of the floor porosity before and after wear, which reflects the wear resistance of the floor coating.

[0015] The floor wear resistance detection method provided by the present application evaluates the wear resistance of the floor coating by analyzing the change of the floor porosity before and after wear, especially at the microscopic level, and uses the image feature change under multiple directions of light to accurately distinguish between real pores and wear scratches, thereby overcoming the limitations of traditional macroscopic measurement methods and the misjudgment problem in two-dimensional optical image analysis, improving the accuracy and reliability of the detection.

[0016] Further, step S1 comprises:

[0017] S11: selecting a plurality of floor samples, and selecting a plurality of detection regions on the surface of each floor sample;

[0018] S12: for each of the detection regions, a two-dimensional optical image is collected under multi-directional light illumination, and information about the light-dark change and shadow shape of the dark region in the two-dimensional optical image under different light directions is analyzed, so as to identify and quantify the porosity of the detection region;

[0019] S13: the average value of the porosities of all the detection regions is calculated to obtain the initial porosity representative of the batch.

[0020] The floor wear resistance detection method provided in the application quantifies the porosities of a plurality of floor samples under multi-region and multi-directional light illumination, and calculates the average value, so as to obtain a more representative and accurate initial porosity, and provide reliable benchmark data for subsequent wear evaluation, and effectively reduce the accidental error of single-point or single-time measurement.

[0021] Further, step S2 comprises:

[0022] S21: the two-dimensional optical images are subjected to noise reduction processing, and one of the two-dimensional optical images after the noise reduction processing is selected as a reference image;

[0023] S22: key feature points are detected in the reference image and the remaining two-dimensional optical images after the noise reduction processing;

[0024] S23: according to the key feature points, error matching points in the key feature points are removed to obtain matching points;

[0025] S24: according to the matching points, a geometric transformation matrix between the reference image and the remaining two-dimensional optical images after the noise reduction processing is calculated;

[0026] S25: the remaining two-dimensional optical images after the noise reduction processing are subjected to geometric correction by using the geometric transformation matrix, so that the remaining two-dimensional optical images after the noise reduction processing are accurately overlapped with the reference image to obtain aligned images.

[0027] The floor wear resistance detection method provided in the application ensures the accurate alignment of images under different light directions by accurately performing noise reduction, key feature point detection, error matching point removal and geometric correction on the two-dimensional optical images under multi-directional light illumination, lays a solid foundation for accurately identifying and analyzing the light-dark change and shadow shape information of the dark region in the subsequent step, and effectively avoids misjudgment caused by misalignment of images.

[0028] Further, step S22 comprises:

[0029] S221: in the reference image and the remaining two-dimensional optical images after the noise reduction processing, the brightness change information of the local region of the image is analyzed;

[0030] S222: identifying, according to the luminance change information, a pixel point with high contrast or significant edge feature;

[0031] S223: taking the pixel point with high contrast or significant edge feature as a key feature point.

[0032] The floor wear resistance detection method provided in the application improves the accuracy and robustness of key feature point detection by analyzing the luminance change information of the local area of the image, identifying a pixel point with high contrast or significant edge feature as a key feature point, and providing a more reliable matching basis for subsequent image alignment.

[0033] Further, in step S3, the identifying the dark area region in the aligned image comprises the steps of:

[0034] S31: performing statistical analysis on the luminance value distribution of each pixel in the aligned image to generate a luminance histogram;

[0035] S32: obtaining the luminance distribution feature of the luminance histogram, and calculating a light-dark segmentation threshold according to the luminance distribution feature;

[0036] S33: dividing the pixel points with luminance values less than or equal to the light-dark segmentation threshold into dark area pixel points;

[0037] S34: performing morphological operation on the dark area pixel points to optimize the boundary of the dark area pixel point forming region to obtain the optimized dark area region.

[0038] Further, in step S3, analyzing the light-dark change information and shadow morphology information of the dark area region under different light directions comprises the steps of:

[0039] S35: for each dark area region, calculating the gray center coordinates in the images collected under different light directions;

[0040] S36: comparing the gray center coordinates in multiple images to obtain the displacement distance and displacement direction of the gray center coordinates, and the shadow morphology information includes the displacement distance and the displacement direction;

[0041] S37: for each pixel point in the dark area region, calculating the gray value in the images collected under different light directions;

[0042] S38: comparing the gray values of the same pixel points in multiple images to obtain the gray value change, and the light-dark change information includes the gray value change.

[0043] Further, step S5 comprises:

[0044] S51: Calculate the area of all dark regions, exclude dark regions with an area less than a preset threshold, and obtain the dark regions for calculation;

[0045] S52: Accumulate the area of the dark regions for calculation to obtain the total area of the floor pores;

[0046] S53: Calculate the floor porosity after wear according to the total area of the floor pores.

[0047] Further, step S6 includes:

[0048] S61: Obtain the floor porosity after wear of a plurality of floor samples, and calculate the average floor porosity of the plurality of floor samples;

[0049] S62: Compare the average floor porosity with the initial porosity to obtain the average porosity change;

[0050] S63: Evaluate the wear resistance of the floor coating according to the average porosity change.

[0051] Further, step S63 includes:

[0052] S631: Compare the average porosity change with a plurality of preset wear resistance level thresholds;

[0053] S632: Determine the wear resistance level of the floor coating according to the comparison result.

[0054] Second aspect, a floor wear resistance detection system for implementing any of the above methods, the system comprises:

[0055] A first acquisition module: acquire the initial porosity before floor wear;

[0056] A second acquisition module: acquire two-dimensional optical images of the surface to be tested under a plurality of different light directions after floor wear, and perform alignment processing on the two-dimensional optical images to obtain aligned images;

[0057] An identification module: identify dark regions in the aligned images, and analyze light-dark change information and shadow shape information of the dark regions under a plurality of different light directions;

[0058] A judgment module: when the light-dark change information changes with the change of the light source position, and the shadow shape information presents dynamic changes such as rotation or movement with the change of the light source position, the corresponding dark region is judged as a floor pore;

[0059] A calculation module: calculate the floor porosity after wear according to the total area of the floor pores;

[0060] The performance detection module: compares the porosity of the floor after wear with the initial porosity, to obtain the change in porosity of the floor before and after wear, which reflects the wear resistance of the floor coating.

[0061] Beneficial effects: The floor wear resistance detection method and system proposed in the present application can effectively distinguish between real pores and shallow scratches caused by wear in two-dimensional optical images by introducing image feature analysis under multi-directional lighting. Specifically, the method reflects the wear resistance of the floor coating by obtaining the change in porosity of the floor before and after wear. When obtaining the porosity after wear, the method uses multi-directional lighting to collect two-dimensional optical images of the surface to be measured, and aligns the images. The key is that by analyzing the light-dark change information and shadow shape information (such as the displacement of gray center coordinates and the change in gray value) of dark areas under different lighting directions, it can accurately determine which dark areas are real pores. When the light-dark change information of the dark area region changes with the change of the light source position, and the shadow shape information presents a dynamic change of rotation or movement, it is judged as a floor pore. This discrimination mechanism based on dynamic changes in lighting effectively solves the problem that the two-dimensional optical microscope scheme in the prior art cannot distinguish between real pores and shallow scratches, avoiding the misjudgment of a large number of scratches caused by wear as pores, thereby significantly improving the accuracy of the calculation of the porosity after wear. Compared with traditional macroscopic measurement methods, the method of the present application can accurately quantify extremely slight wear, overcoming the difficulty of accurately quantifying the change due to the small change. At the same time, compared with the time-consuming three-dimensional profile measurement device, the present application uses a two-dimensional optical microscope scheme combined with an innovative image analysis algorithm to realize imaging and rapid detection in seconds, meeting the efficiency requirements of industrial production. In summary, the method of the present application can provide an accurate, efficient and reliable evaluation scheme for the wear resistance of ultra-thin functional coatings, effectively solving the problems of detection accuracy and efficiency in the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 A flowchart of a floor wear resistance detection method proposed in the present application.

[0063] Figure 2 A structural diagram of a floor wear resistance detection system proposed in the present application.

[0064] Figure 3 A simple schematic diagram of a floor wear resistance detection system proposed in the present application.

[0065] Label explanation: 201, first acquisition module; 202, second acquisition module; 203, identification module; 204, judgment module; 205, calculation module; 206, performance detection module. DETAILED DESCRIPTION

[0066] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. The components of the embodiments of the present application described and indicated in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0067] It should be noted that similar reference numerals and letters refer to similar items throughout the drawings, and therefore, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", and the like are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

[0068] Please refer to Figure 1 A floor wear resistance detection method, the method comprising the steps of:

[0069] S1: obtaining an initial porosity of the floor before wear;

[0070] S2: obtaining two-dimensional optical images of the surface to be tested under a plurality of different light directions after wear of the floor, and performing alignment processing on the two-dimensional optical images to obtain aligned images;

[0071] S3: identifying dark area regions in the aligned images, and analyzing light-dark change information and shadow shape information of the dark area regions under a plurality of different light directions;

[0072] S4: when the light-dark change information changes with the change of the light source position, and the shadow shape information presents dynamic changes such as rotation or movement with the change of the light source position, judging that the corresponding dark area region is a floor porosity;

[0073] S5: calculating a porosity of the floor after wear according to a total area of the floor porosities;

[0074] S6: comparing the porosity of the floor after wear with the initial porosity to obtain a porosity change amount of the floor before and after wear, and the porosity change amount of the floor reflects the wear resistance of the floor coating.

[0075] Specifically, in step S1, the initial porosity before floor wear can be obtained in various ways. For example, the initial porosity can be obtained by detecting the surface of the un-worn floor sample, collecting its two-dimensional optical image, and identifying and quantifying the pores on its surface using image processing techniques. As a preferred embodiment, a plurality of floor samples can be selected, and a plurality of detection regions can be selected on the surface of each sample for porosity measurement, and then the average of these measurement values is calculated as the initial porosity to improve representativeness.

[0076] In step S2, the acquisition of two-dimensional optical images can be achieved by a microscope system equipped with an adjustable light source, which can illuminate the surface to be measured from different angles. The purpose of image alignment is to eliminate the slight displacement or rotation that may exist in the image acquisition process, and to ensure that the images under different light directions can be accurately overlapped, so as to accurately analyze the changes in brightness and shadow of the same area in the subsequent step. The alignment process can use a method based on image feature point matching, for example, several easily identifiable key feature points are selected in multiple images, the relative positional relationship of these key feature points in different images is calculated, and then the geometric transformation parameters between the images are calculated, and the images are corrected using these parameters.

[0077] In step S3, the identification of dark area regions can be achieved by analyzing the brightness histogram of the image, setting a suitable light-dark segmentation threshold, and identifying the pixel points with a brightness value lower than the light-dark segmentation threshold as dark area pixel points. For the analysis of light-dark change information, the gray value changes of the pixel points in the same dark area region in the images under different light directions can be compared. For the analysis of shadow shape information, the displacement and rotation of the geometric center or gray center of the dark area region in the images under different light directions can be tracked.

[0078] In step S4, the core of the present application is to utilize the optical response difference between real pores and shallow scratches under different light conditions. Real pores have a certain depth and three-dimensional structure, and when the light source position changes, the reflection and scattering characteristics inside the pores will change, resulting in changes in the brightness of the image, and the shadow area formed by the pores will rotate or move due to the change in the angle of the light source. In contrast, shallow scratches are usually small in depth, and their optical response is not sensitive to changes in the position of the light source, or their shadow shape changes are not obvious. Therefore, by setting corresponding judgment criteria, real pores and shallow scratches can be effectively distinguished.

[0079] In step S5, the floor porosity after abrasion is calculated according to the total area of the floor pores identified. This generally involves accumulating the pixel areas of all dark region areas identified as pores, and then dividing by the total area of the surface to be measured to obtain the porosity after abrasion. In order to improve the accuracy of the calculation, dark region areas with too small or too large areas can be excluded to avoid misjudging noise or atypical features as pores.

[0080] In step S6, the floor porosity change directly reflects the degree of change in the surface pore structure of the floor coating during the abrasion process, thereby indirectly reflecting the abrasion resistance of the floor coating. The greater the floor porosity change, the more severe the coating abrasion and the poorer the abrasion resistance; on the contrary, the smaller the floor porosity change, the better the abrasion resistance of the coating.

[0081] The scheme of the present application effectively solves the problem that the traditional two-dimensional optical detection method cannot distinguish between real pores and shallow scratches by introducing multi-directional light collection and dynamic optical response analysis. Compared with the time-consuming three-dimensional profile measurement equipment, the method of the present application can realize high-precision detection using rapidly acquired two-dimensional optical images, significantly improving the detection efficiency and meeting the needs of rapid detection in industrial production. Thus, the present application can provide an accurate, efficient and practical solution for evaluating the abrasion resistance of ultra-thin functional coatings, thereby ensuring product quality.

[0082] Further, step S1 comprises:

[0083] S11: selecting a plurality of floor samples and selecting a plurality of detection regions on the surface of each floor sample;

[0084] S12: for each detection region, collecting a two-dimensional optical image using multi-directional light, and analyzing the light-dark change information and shadow shape information of the dark region area in the two-dimensional optical image under a plurality of different light directions, thereby identifying and quantifying the porosity of the detection region;

[0085] S13: calculating the average value of the porosities of all detection regions to obtain an initial porosity representative of the batch.

[0086] In step S11, by selecting a plurality of detection regions, it is ensured that the initial porosity obtained can reflect the overall characteristics of a batch of floor products, rather than only the local features of a single region or a single floor, thereby improving the representativeness of the data.

[0087] In step S12, the analysis process can accurately distinguish between real pores and surface textures or impurities by observing the optical response of dark areas under different lighting conditions, and further quantify the area ratio, ensuring the accuracy of pore identification. When the light and shadow change information and shadow shape information of the dark area region in the two-dimensional optical image change with the change of the light position, it can be determined that the dark area region is a real gap, not a scratch.

[0088] In step S13, by averaging the porosities of multiple detection regions, the influence of single measurement error can be effectively reduced, further improving the accuracy and reliability of the initial porosity.

[0089] The scheme of the present application can ensure that the initial porosity data obtained has high representativeness and accuracy by introducing a multi-sample, multi-region detection strategy when obtaining the initial porosity, and combining image analysis technology under multi-directional lighting. Specifically, multiple floor samples and multiple detection regions are selected to cover the variability of the floor product batch, avoiding measurement bias caused by local unevenness. At the same time, multi-directional lighting is used to collect images and analyze light and shadow change information and shadow shape information, making the identification of pores more accurate and effectively excluding the interference of non-pore features. Thus, by calculating the average value, the measurement data is further smoothed, so that the initial porosity can truly reflect the original state of the floor coating before wear.

[0090] Further, step S2 comprises:

[0091] S21: denoising the two-dimensional optical image, and selecting one of the denoised two-dimensional optical images as a reference image;

[0092] S22: detecting key feature points in the reference image and the remaining denoised two-dimensional optical images;

[0093] S23: according to the key feature points, removing the error matching points in the key feature points to obtain matching points;

[0094] S24: according to the matching points, calculating the geometric transformation matrix between the reference image and the remaining denoised two-dimensional optical images;

[0095] S25: using the geometric transformation matrix to perform geometric correction on the remaining denoised two-dimensional optical images, so that the remaining denoised two-dimensional optical images and the reference image are accurately overlapped to obtain aligned images.

[0096] Specifically, in step S21, during the image acquisition process, due to the limitations of the sensor itself, changes in lighting conditions, environmental interference and other factors, different types of noise such as Gaussian noise, salt and pepper noise, Poisson noise, etc. may be introduced into the image. These noises can reduce the signal-to-noise ratio of the image, blur the image details, and even cause errors in subsequent feature extraction and image recognition. Therefore, the purpose of the noise reduction processing of the two-dimensional optical image in the present application is to eliminate the noise that may be introduced during the image acquisition process, thereby improving the image quality and providing a clear data basis for subsequent feature detection and image alignment. Noise reduction processing can use various image processing algorithms, such as Gaussian filtering.

[0097] wherein the Gaussian filter is a commonly used linear smoothing filter method, and the Gaussian filter is implemented by convolving a two-dimensional Gaussian kernel (or called Gaussian template) with the two-dimensional optical image. The Gaussian kernel is a matrix, and each element in the matrix represents the weight of the corresponding pixel. When the Gaussian kernel slides over the image, it calculates the weighted sum of the pixels within the kernel and takes this sum as the new gray value of the central pixel. The size (e.g. 3x3, 5x5) and the standard deviation (σ) of the Gaussian kernel are two key parameters of the Gaussian filter. The size of the kernel determines the range of the filter, while the standard deviation controls the degree of smoothing. The larger the standard deviation, the flatter the Gaussian function curve, the stronger the filtering effect, and the more blurred the two-dimensional optical image; on the contrary, the smaller the standard deviation, the weaker the filtering effect, and the more details of the two-dimensional optical image are retained. The present application uses Gaussian filtering to complete noise reduction processing, which can better preserve the edge details of the two-dimensional optical image.

[0098] After the noise reduction processing is completed, one image is selected from the processed multiple two-dimensional optical images as a reference image, which will serve as the reference for aligning all other images. Generally, the image with the best image quality, the most abundant features or located in the middle of the sequence can be selected as the reference image to optimize the alignment effect.

[0099] Further, in step S22, the key feature points refer to local image features that are unique, repeatable and have a certain robustness to changes in lighting, scale, rotation, etc. These feature points can be corner points, edge points or regions with rich texture. The method of detecting key feature points can use the SIFT (Scale Invariant Feature Transform) algorithm.

[0100] Among them, SIFT algorithm is a classic feature detection algorithm, which can detect local features of images in different scale spaces and generate descriptors that are invariant to scale and rotation. The core idea is to simulate images at different scales by building a Gaussian difference pyramid, and find extreme points as key points in these scale spaces. Then, the key points are assigned directions and 128-dimensional descriptors are generated, which can effectively capture the local image information around the key points. SIFT features are widely used in image matching, target recognition and other fields due to their robustness to image transformation.

[0101] Therefore, the skilled person in the art can establish the correspondence between the two images by detecting and matching these key feature points between the reference image and the image to be aligned. These matching points are then used to calculate the geometric transformation matrix between the images, thereby achieving accurate alignment of the images. For example, two-dimensional optical images collected under multi-directional lighting may have subtle shifts or deformations in image content due to changes in light source angle, and the stable feature points detected by the SIFT algorithm can effectively overcome these changes, ensuring that images under different lighting directions can be accurately overlapped, providing accurate image basis for subsequent dark area recognition and analysis.

[0102] In step S23, when matching between the preliminarily detected key feature points, a large number of false matches may be generated due to factors such as image noise, repetitive texture or changes in viewing angle. In order to ensure the accuracy of alignment, these matching points need to be screened. The method of eliminating false matching points can use the RANSAC (Random Sample Consensus) algorithm, which is widely used to estimate the correct model parameters from a data set containing a large number of outliers. It can estimate model parameters from a data set containing a large number of outliers, thereby identifying and removing false matching points that do not conform to the geometric transformation model, and retaining correct matching points.

[0103] Specifically, the implementation process of the skilled person in the art for eliminating false matching points using the RANSAC algorithm is as follows:

[0104] First, randomly select a minimum sample set: randomly select a minimum number of matching point pairs (e.g. for affine transformation, at least 3 matching point pairs are usually required; for perspective transformation, at least 4 matching point pairs are usually required) from all the preliminary matching points. This minimum number of matching point pairs is assumed to be "inliers", i.e. they are correct matches.

[0105] Second, estimate the model parameters: use these randomly selected matching point pairs to calculate the affine transformation model, which can solve the parameters of the transformation matrix through the coordinates of these matching point pairs.

[0106] Third, verify the model and count inliers: using this estimated affine transformation model, transform all the remaining matched points. For each matched point, calculate the distance (e.g., Euclidean distance) between its corresponding points before and after the transformation. If this distance is less than a pre-defined error threshold, consider the matched point consistent with the current model and mark it as an "inlier"; otherwise, mark it as an "outlier". Count the number of inliers supported by the current model.

[0107] Finally, iterate and select the best model: repeat the above process of random selection, model estimation, and inlier counting multiple times. After all iterations are completed, select the model that supports the most inliers as the final radiometric transformation model. All matched points consistent with this best model are considered correct matched points, while those inconsistent are identified and removed as false matched points.

[0108] Thus, in step S24, the geometric transformation matrix describes the spatial mapping relationship between images, such as translation, rotation, scaling, or perspective transformation. By using the accurate matched point pairs obtained in step S23, the least squares method can be used to calculate the geometric transformation matrix that can accurately transform one image to the coordinate system of another image. The least squares method is a commonly used optimization algorithm that can be used to estimate the geometric transformation matrix. The basic idea is to find a geometric transformation matrix that minimizes the sum of the squared distances between all matched point pairs after the transformation.

[0109] Specifically, by using the point coordinates (x_s, y_s) in a set of known reference images and the corresponding point coordinates (x_t, y_t) in the remaining denoised two-dimensional optical images, assuming the transformation model is a radiometric transformation model, a system of over-determined linear equations can be constructed, and then the least squares method can be used to solve the system of equations to obtain the parameters of the geometric transformation matrix. The geometric transformation model can minimize the sum of the squared distances between all matched point pairs after the transformation.

[0110] Finally, in step S25, the geometric correction process is to apply the calculated geometric transformation matrix to the two-dimensional optical images to be corrected, and through pixel interpolation and other techniques, each pixel in the image is mapped from its original position to the corresponding position in the reference image. In this way, the spatial misalignment between images collected under different lighting directions can be eliminated, and all images can be accurately overlapped in space, providing a unified coordinate system for subsequent image analysis.

[0111] The scheme of the present application ensures the purity of image data by performing noise reduction processing on the original two-dimensional optical images, laying a foundation for subsequent processing. Subsequently, the spatial correspondence between images is effectively captured by detecting and accurately matching key feature points between images. Further, the reliability of the matching is greatly improved by eliminating false matching points, avoiding alignment deviation caused by false matching. Based on these reliable matching points, the transformation matrix describing the geometric relationship between images can be accurately calculated. Finally, the geometric correction of the images is performed using the geometric transformation matrix, so that all images are accurately overlapped in space. Through this series of logically rigorous steps, the scheme of the present application can overcome the possible slight displacement, rotation or deformation of images collected under different light directions, ensuring the high consistency of multiple images in space.

[0112] Further, step S22 comprises:

[0113] S221: Analyzing the brightness change information of the local area of the image in the reference image and the remaining two-dimensional optical images after noise reduction processing;

[0114] S222: Identifying the pixel points with high contrast or significant edge features according to the brightness change information;

[0115] S223: Taking the pixel points with high contrast or significant edge features as key feature points.

[0116] Wherein, analyzing the brightness change information of the local area of the image refers to investigating each pixel point and its neighborhood in the reference image and the remaining two-dimensional optical images after noise reduction processing, and calculating the change degree of its brightness value relative to the surrounding pixels. For example, a gradient operator (such as Sobel operator) can be used to calculate the gradient amplitude and direction of the image to quantify the degree of brightness change.

[0117] Specifically, the gradient of the image represents the rate of change of image brightness in space. The gradient amplitude reflects the intensity of brightness change, i.e. the "strength" of the edge; the gradient direction indicates the direction of brightness change, i.e. the "trend" of the edge.

[0118] The Sobel operator is widely used in image processing for tasks such as edge detection, feature extraction, image segmentation, and can effectively extract key information reflecting the outline and texture of objects from images. Those skilled in the art use the Sobel operator to approximate the gradient of an image in the horizontal and vertical directions by performing convolution operations between the convolution kernel (usually a 3x3 matrix) and the image (including the reference image and the two-dimensional optical image after noise reduction processing). For example, the Sobel operator contains two convolution kernels, one for detecting edges in the horizontal direction (Gx) and the other for detecting edges in the vertical direction (Gy). By convolving the image with these two kernels, the gradient components of each pixel in the horizontal and vertical directions can be obtained. Then, the magnitude (usually sqrt(Gx^2 + Gy^2)) and direction (usually arctan(Gy / Gx)) of the gradient can be calculated based on these two components. The larger the gradient magnitude calculated by the Sobel operator, the more intense the brightness change at that pixel, and the more likely it is an edge of the image.

[0119] Further, identifying the pixel points with high contrast or significant edge features based on the brightness change information means that based on a pre-set threshold, those pixel points with the most significant changes are selected from the brightness change information. High-contrast pixel points are usually located at the junction of light and dark in the image, while pixel points with significant edge features constitute the outline of the image. These points have good stability under image transformation (such as rotation, scaling), and are not easily changed in their relative positions by small disturbances. For example, in image processing, in order to identify the key features of an image, the brightness change of the local region of the image needs to be analyzed. The brightness change information is usually obtained by calculating the gradient of the image, which reflects the degree of change in pixel brightness in space. After obtaining the brightness change information, a pre-set threshold can be set to select the pixel points. For example, a gradient magnitude threshold can be set, and only when the gradient magnitude of a pixel point exceeds this threshold, the pixel point is considered to be a point with significant brightness change.

[0120] Thus, the identified pixel points with high contrast or significant edge features are used as key feature points. These key feature points are the basis for subsequent image alignment processing, and they can provide sufficient information to accurately calculate the geometric transformation relationship between images.

[0121] The scheme of the present application can effectively capture the structural features with distinguishability in the image by analyzing the brightness change information of the local region of the image. It is because the edges and high-contrast regions in the image are relatively stable under different illumination and viewing angles, and therefore, the pixel points identified based on these regions as key feature points can provide reliable reference for subsequent image alignment. When the image is translated, rotated or scaled, the position and relative relationship of these key feature points change regularly, making the calculation of the geometric transformation matrix more accurate. Thus, by identifying these pixel points with high contrast or significant edge features, it can be ensured that the selected feature points have high matchability between different images, thereby laying a foundation for accurate image alignment.

[0122] Further, in step S3, identifying the dark region in the aligned image comprises the steps of:

[0123] S31: statistically analyzing the brightness value distribution of each pixel in the aligned image to generate a brightness histogram;

[0124] S32: obtaining the brightness distribution feature of the brightness histogram, and calculating a light-dark segmentation threshold according to the brightness distribution feature;

[0125] S33: dividing the pixel points with brightness values less than or equal to the light-dark segmentation threshold into dark region pixel points;

[0126] S34: performing morphological operation on the dark region pixel points to optimize the boundary of the dark region pixel points to form region, to obtain the optimized dark region.

[0127] In step S31, the brightness histogram can intuitively reflect the pixel number distribution of different brightness levels in the image, thereby providing basic data for subsequent image segmentation. Specifically, each pixel in the aligned image can be traversed to obtain its gray value or brightness value, and these values are counted and mapped to the preset brightness interval to finally form the brightness histogram.

[0128] Further, in step S32, the brightness distribution feature includes but is not limited to the peak value, valley value, distribution range, skewness and kurtosis of the brightness histogram. In image processing, in order to effectively separate the light region and the dark region, a suitable threshold is usually needed to divide the image pixels into foreground (light region) and background (dark region), for example, the Otsu (Otsu) algorithm can be used to automatically determine the best threshold.

[0129] Otsu algorithm is a global threshold-based image segmentation method, its core idea is to find a best threshold, so that the threshold can divide the pixels in the image into two classes (foreground and background), and the variance between the two classes is maximized. Specifically, Otsu algorithm will traverse all possible gray levels as the threshold, for each candidate threshold, it will divide the pixels of the image into two groups: one group is the pixels whose gray value is less than or equal to the candidate threshold, the other group is the pixels whose gray value is greater than the candidate threshold. Then, the algorithm calculates the intra-class variance and inter-class variance of the two groups. The goal of Otsu algorithm is to find a threshold that maximizes the inter-class variance, when the inter-class variance is maximized, it means that the difference between the foreground and the background is most significant, so that the best segmentation effect can be achieved. This method is particularly effective for images with bimodal histogram, because it can effectively separate the two main gray peaks.

[0130] In step S33, the dark region pixels preliminarily constitute the dark region area in the image, which usually corresponds to the pores, depressions or other low reflectivity areas of the floor surface.

[0131] In step S34, in order to further improve the recognition accuracy and morphological integrity of the dark region area, morphological operations are performed on the dark region pixels divided above. Morphological operations include but are not limited to erosion, dilation, opening operation or closing operation, which can effectively remove noise points in the image, fill small breaks, smooth the region boundary, and connect adjacent dark region pixels.

[0132] Specifically, morphological operations are a series of nonlinear operations based on shape analysis in image processing, mainly applied to binary images, but can also be extended to grayscale images. These operations define a structure element (or kernel), slide on the image, and change the pixel value according to the interaction between the structure element and the local region of the image.

[0133] In the scheme of the present application, the erosion operation can shrink the foreground object in the aligned image or remove the pixels at the edge of the image. When the structure element slides on the aligned image, only when the structure element is completely contained in the foreground object, the center pixel will be retained. This operation can effectively remove isolated noise points in the aligned image, because these noise points are usually small in size and cannot completely contain the structure element, so they are eliminated after erosion.

[0134] Dilation operation is opposite to erosion, it can expand the foreground object in the aligned image or fill the holes in the image. When the structure element slides on the image, as long as the structure element has any overlap with the foreground object, the center pixel will be set to foreground. This operation can be used to fill small breaks in the image or connect adjacent dark region pixels, so that the originally separated regions are connected into a whole.

[0135] An opening operation is a combination of an erosion operation followed by a dilation operation. It first erodes the aligned image to remove small noise points and thin connections, then dilates to restore the main shape of the objects. The main purpose of opening is to smooth the object contours, remove small protrusions, and break thin connections.

[0136] A closing operation is a combination of a dilation operation followed by an erosion operation. It first dilates the image to fill small holes and breaks, then erodes to restore the main shape of the objects. The main purpose of closing is to smooth the object contours, fill small holes, and connect adjacent broken areas.

[0137] In summary, morphological operations can effectively remove noise points in the image, fill small breaks, smooth region boundaries, and connect adjacent dark pixels, thus optimizing the morphology of dark regions in the image, making it more accurately reflect the actual pore structure.

[0138] Through these operations, the boundaries of the regions formed by dark pixels can be optimized to better match the actual pore morphology, resulting in an optimized dark region that provides more accurate input for subsequent pore analysis.

[0139] Further, in step S3, analyzing the light-dark change information and shadow morphology information of the dark region under multiple different illumination directions includes steps:

[0140] S35: For each dark region, calculate the gray center coordinates in images collected under different illumination directions;

[0141] S36: Compare the gray center coordinates in multiple images to obtain the displacement distance and direction of the gray center coordinates, and the shadow morphology information includes the displacement distance and direction;

[0142] S37: For each pixel in the dark region, calculate the gray value in the image collected under different illumination directions;

[0143] S38: Compare the gray values of the same pixel in multiple images to obtain the gray value change, and the light-dark change information includes the gray value change.

[0144] In step S35, for each dark region identified, the gray center coordinates in the images collected under different illumination directions need to be calculated. The gray center coordinates refer to the weighted average position of the pixel gray values in the image region, which can reflect the brightness distribution center of the region. Its calculation can be obtained by weighted average of the gray values and corresponding coordinates of all pixels in the dark region.

[0145] Specifically, assume that a dark region is composed of a series of pixel points, each with its specific coordinates (x, y) and gray value I(x', y'). The gray center coordinates (Xc, Yc) of this dark region can be calculated by the following formulas:

[0146] Xc = Σ [x * I(x’, y’)] / Σ I(x’, y’)

[0147] Yc = Σ [y * I(x’, y’)] / Σ I(x’, y’).

[0148] Where Σ represents the summation of all pixels in the dark region. In this formula, the coordinates (x, y) of each pixel are multiplied by its gray value I(x', y') as a weighting factor, and then all weighted coordinate values are added up and divided by the total sum of gray values of all pixels in the region. This weighted average approach makes the pixels with higher brightness have a greater impact on the center position, thus more accurately reflecting the brightness distribution center of the region.

[0149] In step S36, by comparing the gray center coordinates of the same dark region obtained under multiple different lighting directions, the displacement distance and direction of these gray center coordinates can be obtained. These displacement information collectively constitutes the shadow morphology information. For example, when the light source position changes, the shadow of the real pore will move or rotate, causing the gray center coordinates to displace. By quantifying this displacement, the dynamic change characteristics of the shadow can be effectively captured.

[0150] In addition, in step S37, for each pixel point in the dark region, the gray value in the image collected under different lighting directions needs to be calculated. The gray value reflects the brightness information of the pixel.

[0151] In step S38, by comparing the gray values of the same pixel points in multiple images, the gray value change amount can be obtained. These gray value change amounts constitute the light-dark change information. For example, when the light source position changes, the pixel points inside the real pore will have significant changes in gray value due to different light blocking and reflection characteristics. By analyzing the change amount of these gray values, it can be determined whether the region is a real pore, because the surface defects (such as scratches, stains) of non-pore usually do not show similar light-dark change rules under different lighting.

[0152] The scheme of the present application can quantify the displacement distance and direction of the shadow by calculating and comparing the gray center coordinates of the dark area region under different light directions, thereby capturing the dynamic morphological changes of the shadow. At the same time, by comparing the gray values of the pixel points in the dark area region, the light and dark change information can be obtained. These information are collectively used to determine whether the dark area region is a real floor pore. The shadow inside the real floor pore will move or rotate obviously under different light directions, and its brightness value will also change accordingly, while other surface defects (such as stains, scratches) usually do not have such regular light and shadow change characteristics. Thus, the real pores and non-pore dark areas can be more accurately distinguished.

[0153] Further, step S5 comprises:

[0154] S51: Calculate the area of all dark area regions, exclude dark area regions with an area less than a preset threshold, and obtain dark area regions for calculation;

[0155] S52: Sum the areas of the dark area regions for calculation to obtain the total area of the floor pores;

[0156] S53: Calculate the floor porosity after wear according to the total area of the floor pores.

[0157] In step S51, calculating the area of all dark area regions means quantifying the area of each independent dark area region judged as a floor pore in step S4. The purpose of excluding dark area regions with an area less than a preset threshold is to filter out small noise points, artifacts or small defects that are not pores that may be generated during image acquisition or processing. These small regions may not be representative due to their small size, or not be real floor pores.

[0158] The preset threshold can be determined by experiment to ensure that only dark area regions that meet the characteristics of real pores are included in subsequent calculations.

[0159] In step S52, the areas of the dark area regions determined for calculation are summed to obtain the total area of all real pores on the floor surface being tested. This total area is a key indicator of the degree of porosity of the floor surface.

[0160] In step S53, the floor porosity can be defined as the ratio of the total area of the floor pores to the total area of the surface being tested. For example, the porosity in percentage form can be obtained by dividing the total area of the floor pores by the total area of the surface being tested and multiplying by 100%. This porosity value directly reflects the surface integrity of the floor coating after wear.

[0161] By the above technical solution, the accuracy and reliability of the calculation of the porosity of the floor after wear can be effectively improved. The improvement makes the calculated floor porosity more truly reflect the actual wear condition of the floor coating, thereby providing a more accurate data basis for subsequent evaluation of the wear resistance of the floor. This helps to avoid errors introduced by noise or irrelevant features, making the wear resistance judgment more objective.

[0162] Further, step S6 comprises:

[0163] S61: Obtain the floor porosity after wear of a plurality of floor samples, and calculate the average floor porosity of the plurality of floor samples;

[0164] S62: Compare the average floor porosity with the initial porosity to obtain the average porosity change amount;

[0165] S63: Evaluate the wear resistance of the floor coating according to the average porosity change amount.

[0166] In step S61, the floor porosity after wear of each sample is calculated. Then, these individually calculated floor porosities are aggregated, and the average floor porosity of the plurality of floor samples is obtained by calculating the average value. This average value can more comprehensively reflect the overall porosity condition of a batch of floors after wear.

[0167] In step S62, by subtracting the average floor porosity after wear from the initial porosity, an average porosity change amount reflecting the overall wear degree can be obtained.

[0168] In step S63, the calculated average porosity change amount is used as a core index to quantify and evaluate the ability of the floor coating to resist wear. For example, the greater the average porosity change amount, the worse the wear resistance of the floor coating; on the contrary, the smaller the average porosity change amount, the better the wear resistance.

[0169] The scheme of the present application effectively solves the problem of the randomness and insufficient representation of the test results of a single floor sample by introducing the way of testing a plurality of floor samples and calculating the average porosity change amount. It is because of the statistical analysis of a plurality of samples that the obtained average porosity change amount can more accurately and stably reflect the real wear resistance of the floor coating of the entire batch. This evaluation method based on the average value of multiple samples can effectively reduce the influence of local defects or test errors, thereby improving the reliability and accuracy of the evaluation results.

[0170] Further, step S63 comprises:

[0171] S631: Compare the average porosity change amount with a plurality of preset wear resistance grade thresholds;

[0172] S632: Determine the wear resistance grade of the floor coating according to the comparison result.

[0173] The preset multiple wear resistance threshold values refer to numerical limits for dividing different wear resistance grades, which are preset according to industry standards, product specifications or empirical data. These threshold values can be a series of discrete numerical points, for example, 0.01, 0.03, 0.05, etc., corresponding to different wear resistance grade intervals. The calculated average porosity change is compared with these preset threshold values to determine the specific interval in which the change falls.

[0174] Further, according to the comparison result, determining the wear resistance grade of the floor coating refers to classifying the average porosity change into the corresponding wear resistance grade according to the threshold interval it falls into. For example, if the average porosity change is less than the first threshold value, it can be determined as "excellent" grade; if it is between the first and second threshold values, it can be determined as "good" grade, and so on. In this way, the wear resistance of the floor coating can be quantified and graded.

[0175] The scheme of the present application realizes the standardized and quantitative evaluation of the wear resistance of the floor coating by introducing preset wear resistance threshold values and comparing the average porosity change with these threshold values. This threshold value comparison mechanism makes the abstract "evaluation" process specific and operable, and can map the continuous porosity change to discrete and easy-to-understand grade classification. In this way, it can provide a clear basis for quality control, performance verification and market classification of floor products.

[0176] Please refer to Figure 2 , Figure 3 A floor wear resistance detection system for implementing any of the above methods, the system comprising:

[0177] The first acquisition module 201 is configured to acquire the initial porosity of the floor before wear. Specifically, this module can be configured to perform step S1 in the above method, that is, by selecting a plurality of floor samples and selecting a plurality of detection areas on the surface of each floor sample, collecting two-dimensional optical images in multiple directions, and analyzing the light-dark change information and shadow shape information of the dark area in the two-dimensional optical images in multiple directions, the porosity of the detection area is identified and quantified, and finally the average value of the porosity of all detection areas is calculated to obtain the initial porosity representative of the batch.

[0178] The second acquisition module 202 is configured to acquire two-dimensional optical images of the surface to be measured under different light directions after the floor is worn, and perform alignment processing on the two-dimensional optical images to obtain an aligned image. Specifically, the module can be configured to perform step S2 in the above method, that is, to perform noise reduction processing on the two-dimensional optical images, select one image as a reference image, detect key feature points between the reference image and the remaining two-dimensional optical images after noise reduction processing, remove error matching points according to the key feature points to obtain matching points, calculate a geometric transformation matrix according to the matching points, and finally perform geometric correction on the remaining two-dimensional optical images after noise reduction processing by using the geometric transformation matrix to make them accurately overlap with the reference image to obtain the aligned image.

[0179] The recognition module 203 is configured to recognize dark area regions in the aligned image, and analyze light-dark change information and shadow morphology information of the dark area regions under different light directions. Specifically, the module can be configured to perform step S3 in the above method, that is, to statistically analyze the brightness value distribution of each pixel in the aligned image, generate a brightness histogram, obtain brightness distribution characteristics of the brightness histogram, and calculate a light-dark segmentation threshold, divide pixel points with a brightness value less than or equal to the light-dark segmentation threshold into dark area pixel points, and perform morphological operations on the dark area pixel points to optimize the boundary of the dark area pixel point formed region to obtain an optimized dark area region. At the same time, the module is also responsible for calculating the gray center coordinates of each dark area region in images collected under different light directions, comparing the gray center coordinates in multiple images to obtain displacement distances and displacement directions as shadow morphology information, and calculating the gray values of pixel points in each dark area region in images collected under different light directions, comparing the gray values of the same pixel points in multiple images to obtain gray value changes as light-dark change information.

[0180] The judgment module 204 is configured to judge that a corresponding dark area region is a floor pore when the light-dark change information changes with the change of the light source position, and the shadow morphology information presents dynamic changes such as rotation or movement with the change of the light source position. The module is configured to perform step S4 in the above method by comprehensively analyzing the light-dark change information and the shadow morphology information to accurately distinguish real pores from surface textures or scratches.

[0181] The calculation module 205 is configured to calculate the floor porosity after wear according to the total area of the floor pores. Specifically, the module can be configured to perform step S5 in the above method, that is, to calculate the areas of all dark area regions, exclude dark area regions with an area less than a preset threshold to obtain dark area regions for calculation, accumulate the areas of the dark area regions for calculation to obtain the total area of the floor pores, and calculate the floor porosity after wear according to the total area of the floor pores.

[0182] The performance detection module 206 is configured to compare the floor porosity after wear with the initial porosity to obtain a floor porosity change before and after wear, and the floor porosity change reflects the wear resistance of the floor coating. The module is configured to perform step S6 in the above method, that is, to obtain the floor porosity after wear of a plurality of floor samples, calculate the average floor porosity of the plurality of floor samples, compare the average floor porosity with the initial porosity to obtain an average porosity change, and evaluate the wear resistance of the floor coating according to the average porosity change, for example, compare it with a plurality of preset wear resistance grade thresholds to determine the wear resistance grade of the floor coating.

[0183] The scheme of the present application modularizes each step of the above floor wear resistance detection method and gives it a specific hardware or software implementation carrier, thereby realizing the automatic and standardized execution of the method. The modules work together to ensure the accuracy of data acquisition, the robustness of image processing, the accuracy of pore identification, and the reliability of the final wear resistance evaluation. Specifically, the first acquisition module 201 and the second acquisition module 202 serve as data input terminals, responsible for providing key image data of the floor surface before and after wear; the identification module 203 and the judgment module 204 serve as core processing units, accurately extracting and identifying floor pores from complex images through image analysis algorithms; the calculation module 205 quantifies the identified pores into comparable porosity data; finally, the performance detection module 206 reflects the wear resistance of the floor coating by comparing the porosity change before and after wear. This modular design makes the entire detection process efficient and repeatable, effectively solving the automation and efficiency problems that the method may face in actual application.

[0184] In this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

[0185] The above only describes the embodiments of the present application and does not limit the protection scope of the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of detecting the wear resistance of a floor, characterized by, The method comprises steps of: S1: obtaining initial porosity of floor before wear; S2: obtaining two-dimensional optical images of the surface to be tested under multiple different light directions after wear of the floor, and performing alignment processing on the two-dimensional optical images to obtain aligned images; S3: identifying dark area regions in the aligned images, and analyzing light and dark change information and shadow shape information of the dark area regions under multiple different light directions; Step S3 comprises: S35: for each dark area region, calculating gray center coordinates of the dark area region in images collected under different light directions; S36: comparing the gray center coordinates in multiple images to obtain displacement distance and displacement direction of the gray center coordinates, and the shadow shape information comprises the displacement distance and the displacement direction; S37: for each pixel point in each dark area region, calculating gray values of the pixel point in images collected under different light directions; S38: comparing the gray values of the same pixel point in multiple images to obtain gray value change amount, and the light and dark change information comprises the gray value change amount; S4: when the light and dark change information changes with change of light source position, and the shadow shape information presents dynamic changes of rotation or movement with change of light source position, judging that the corresponding dark area region is a floor porosity; S5: calculating floor porosity after wear according to total area of the floor porosity; 2. The method of claim 1, wherein, S6: comparing the floor porosity after wear with the initial porosity to obtain floor porosity change amount before and after wear, and the floor porosity change amount reflects wear resistance of a floor coating. Step S1 comprises: S11: selecting multiple floor samples, and selecting multiple detection regions on the surface of each floor sample; S12: for each detection region, collecting two-dimensional optical images by using multi-directional light, and analyzing light and dark change information and shadow shape information of dark area regions in the two-dimensional optical images under multiple different light directions, so as to identify and quantify porosity of the detection region; 3. The method of claim 1, wherein, S13: calculating average value of porosity of all detection regions to obtain the initial porosity with batch representation. Step S2 comprises: S21: performing noise reduction processing on the two-dimensional optical images, and selecting one of the two-dimensional optical images after the noise reduction processing as a reference image; S22: detecting key feature points in the reference image and the remaining two-dimensional optical images after the noise reduction processing; S23: according to the key feature points, removing error matching points in the key feature points to obtain matching points; S24: according to the matching points, calculating a geometric transformation matrix between the reference image and the remaining two-dimensional optical images after the noise reduction processing; 4. The method of claim 3, wherein the floor wear resistance is determined by the following equation: Floor Wear Resistance = 1 - (1 - Floor Wear Resistance of the reference floor) x (1 - Floor Wear Resistance of the test floor). S25: using the geometric transformation matrix to perform geometric correction on the remaining two-dimensional optical images after the noise reduction processing, so that the remaining two-dimensional optical images after the noise reduction processing are accurately overlapped with the reference image to obtain aligned images. Step S22 comprises: S221: analyzing brightness change information of local regions of images in the reference image and the remaining two-dimensional optical images after the noise reduction processing; S222: identifying, according to the luminance change information, a pixel point with high contrast or a significant edge feature; S223: taking the pixel point with high contrast or a significant edge feature as a key feature point.

5. The method of claim 3, wherein the step of determining the wear resistance of the floor is performed by a method comprising: In step S3, the identifying the dark area region in the aligned image comprises steps of: S31: performing statistical analysis on the luminance value distribution of each pixel in the aligned image to generate a luminance histogram; S32: obtaining a luminance distribution feature of the luminance histogram, and calculating a light-dark segmentation threshold according to the luminance distribution feature; S33: dividing a pixel point with a luminance value less than or equal to the light-dark segmentation threshold into a dark area pixel point; S34: performing a morphological operation on the dark area pixel point to optimize the boundary of the dark area pixel point forming region to obtain an optimized dark area region.

6. The method of claim 1, wherein, Step S5 comprises: S51: calculating the area of all the dark area regions, excluding the dark area regions with an area less than a preset threshold to obtain the dark area regions for calculation; S52: accumulating the areas of the dark area regions for calculation to obtain the total area of the floor porosity; S53: calculating the floor porosity after wear according to the total area of the floor porosity.

7. The method of claim 2, wherein the step of determining the wear resistance of the floor is performed by a method comprising: Step S6 comprises: ​ S61: obtaining the floor porosities after wear of a plurality of floor samples, and calculating the average floor porosity of the plurality of floor samples; S62: comparing the average floor porosity with the initial porosity to obtain an average porosity change amount; S63: evaluating the wear resistance of the floor coating according to the average porosity change amount.

8. The method of claim 7, wherein the step of determining the wear resistance of the floor is performed by a method comprising: Step S63 comprises: ​ S631: comparing the average porosity change amount with a plurality of preset wear resistance grade thresholds; S632: determining the wear resistance grade of the floor coating according to the comparison result.

9. A floor wear resistance detection system characterized by, The system for implementing the method of any one of claims 1-8 comprises: a first acquisition module: acquiring an initial porosity before wear of the floor; a second acquisition module: acquiring two-dimensional optical images of the surface to be tested under a plurality of different illumination directions after wear of the floor, and performing alignment processing on the two-dimensional optical images to obtain an aligned image; an identification module: identifying a dark area region in the aligned image, and analyzing light-dark change information and shadow shape information of the dark area region under a plurality of different illumination directions; a judgment module: when the light-dark change information changes with the change of the light source position, and the shadow shape information presents a dynamic change of rotation or movement with the change of the light source position, judging that the corresponding dark area region is a floor porosity; a calculation module: calculating a floor porosity after wear according to the total area of the floor porosity; a performance detection module: comparing the floor porosity after wear with the initial porosity to obtain a floor porosity change amount before and after wear, which reflects the wear resistance of the floor coating.

Images