A performance testing method and system for moisture-proof and waterproof furniture materials

Abstract

The present application relates to the technical field of waterproof material performance testing, and specifically to a performance testing method and system for moisture-proof and waterproof furniture materials, the method comprising: obtaining a grayscale surface image; obtaining a grayscale difference sequence at each acquisition moment; and then obtaining a grayscale distribution sequence at each acquisition moment; calculating a color change coefficient at each acquisition moment; obtaining all crack lines and mildew pixel points in the grayscale surface image at each acquisition moment; and then obtaining a mildew distribution coefficient at each acquisition moment; and calculating a texture distribution coefficient at each acquisition moment, and testing the performance of home furnishing materials. The present application reduces the cost of material performance testing and improves performance testing efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of waterproof material performance testing, and in particular to a performance testing method and system for moisture-proof and waterproof furniture materials. Background Art

[0002] In modern home life, moisture-proof and waterproof furniture materials are highly favored for their excellent durability and aesthetics, and the market demand for high-performance moisture-proof and waterproof furniture materials continues to grow. However, ensuring that these materials can maintain their functionality and appearance quality for a long time in various environments has become a major challenge for manufacturers. For this reason, material performance testing has become an indispensable part of the production process, which is directly related to product quality and user satisfaction. With the advancement of technology, the performance testing of moisture-proof and waterproof furniture materials has also tended to be more accurate, efficient and intelligent to adapt to the rapidly changing market demand and the continuous improvement of product standards.

[0003] In the production of moisture-proof and waterproof furniture materials, due to errors caused by production processes and environmental conditions, the performance of moisture-proof and waterproof furniture materials between different batches may not be exactly the same. For different batches of moisture-proof and waterproof furniture materials, performance tests need to be carried out separately, and weather resistance testing is an important part of this. In the prior art, the test of weather resistance is to simulate the natural environment in the laboratory and accelerate the aging of the material, and measure the performance of the material according to the changes before and after the experiment. This method lacks monitoring of the intermediate links. Whether the performance of the material is qualified requires a complete test cycle to determine, which affects the performance testing efficiency of moisture-proof and waterproof furniture materials. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a performance testing method and system for moisture-proof and waterproof furniture materials. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for testing the performance of moisture-proof and waterproof furniture materials, the method comprising the following steps:

[0006] Acquire the grayscale surface image of the furniture material at each acquisition moment during the furniture material image acquisition process;

[0007] Based on the difference between the grayscale values ​​of each grayscale surface image, a grayscale difference sequence at each acquisition moment is obtained; based on the distribution of grayscale values ​​in the grayscale surface image, a grayscale distribution sequence at each acquisition moment is obtained;

[0008] Based on the change trend of the grayscale difference sequence and the difference between the grayscale distribution sequences at each acquisition moment, the color change coefficient at each acquisition moment is obtained;

[0009] Based on the edge features and gray value features in the gray surface image, all crack lines and mildew pixel points in the gray surface image at each acquisition time are obtained; based on the distance between the mildew pixel points and the crack line, the mildew distribution coefficient at each acquisition time is obtained;

[0010] Based on the number of crack lines, the distance between the mildew spot pixel point and the crack line, the mildew spot distribution coefficient and the color change coefficient, the texture distribution coefficient at each acquisition time is obtained;

[0011] The prediction coefficient is obtained based on the texture distribution coefficient to test the performance of home furnishing materials.

[0012] Furthermore, the grayscale difference sequence is obtained by:

[0013] For the grayscale surface image at each acquisition moment, the grayscale mean of the grayscale surface image is calculated; for each acquisition moment, the absolute value of the difference between the grayscale surface image and the grayscale mean of the grayscale surface image at the previous acquisition moment is calculated as the grayscale difference at each moment, and the grayscale differences of all acquisition moments before each acquisition moment are arranged in chronological order to obtain the grayscale difference sequence at each acquisition moment.

[0014] Furthermore, the grayscale distribution sequence is obtained by:

[0015] For each pixel in each grayscale surface image, for a neighborhood of a preset size centered on the pixel, the average of the grayscale value differences between other pixels in the neighborhood and the central pixel is taken as the grayscale distribution value of each pixel;

[0016] For each grayscale surface image, the grayscale distribution values ​​of all pixels in the grayscale surface image are arranged in order from left to right and from top to bottom to construct a grayscale distribution sequence of each grayscale surface image.

[0017] Further, the method for obtaining the color change coefficient is:

[0018] Based on the changing trend of the grayscale difference sequence, the material color change rate at each acquisition moment is obtained;

[0019] For each acquisition moment, the difference between the grayscale distribution sequence and the grayscale distribution sequence at the first acquisition moment is calculated as the grayscale distribution difference at each acquisition moment;

[0020] Based on the material color change rate and grayscale distribution difference, the color change coefficient at each acquisition moment is obtained, and the color change coefficient is positively correlated with the material color change rate and grayscale distribution difference, respectively.

[0021] Further, the method for obtaining the material color change rate is:

[0022] Calculate the first-order difference sequence of the grayscale difference sequence at each acquisition moment, perform curve fitting on the first-order difference sequence to obtain a fitting curve function, and perform straight line fitting to obtain a fitting straight line function;

[0023] For the fitted straight line function and curve function at each acquisition moment, the absolute value of the difference between the fitted straight line function value and the curve function value at the same acquisition moment is calculated as the first absolute value of the difference. For each acquisition moment, the variance of all the first absolute values ​​of the difference is calculated as the material color change rate at each acquisition moment.

[0024] Furthermore, the method for obtaining the crack straight line and the mildew spot pixel points is:

[0025] Annotate the bounding box of each crack in the grayscale surface image at each acquisition moment;

[0026] Use an edge detection algorithm to obtain edge lines in each grayscale surface image, take the longest edge line in each boundary box area as the crack edge line, and use a straight line fitting algorithm to fit each crack edge line to obtain each crack straight line;

[0027] For each pixel in each grayscale surface image, a neighborhood of a preset size centered on each pixel is used to calculate the variance of the grayscale values ​​of all pixels in the neighborhood as the neighborhood grayscale fluctuation value of each pixel;

[0028] The neighborhood grayscale fluctuation values ​​of all pixels in each grayscale surface image are threshold segmented, and the segmentation threshold is output. All pixels greater than the segmentation threshold are marked as mildew pixels.

[0029] Further, the method for obtaining the mildew spot distribution coefficient is:

[0030] For each mildew spot pixel point, the distance between the mildew spot pixel point and each crack straight line is calculated, and the minimum value of all distances to the crack straight line is taken as the crack distance of each mildew spot pixel point;

[0031] Clustering algorithms are used to cluster all the mildew spots pixels to obtain clusters;

[0032] For the grayscale surface image at each acquisition moment, the distance between the cluster centers of any two clusters in the grayscale surface image is calculated as the first distance, and the sum of all the first distances is taken as the mildew spot distribution coefficient.

[0033] Further, the method for obtaining the texture distribution coefficient is:

[0034] For the grayscale surface image at each acquisition moment, the mean of the crack distances of all mildew pixel points in the grayscale surface image is calculated as the first mean; the calculation result of an exponential function with a natural constant as the base and the sum of the number of crack straight lines and the first mean as the exponent is calculated as the first exponent; the product of the first index, the mildew distribution coefficient and the color change coefficient is taken as the texture distribution coefficient at each acquisition moment.

[0035] Furthermore, the method of obtaining a prediction coefficient based on the texture distribution coefficient and testing the performance of the home furnishing material includes:

[0036] Obtain the grain distribution coefficient at the last acquisition moment in the standard weathering test for qualified waterproof wood;

[0037] The maximum value of the texture distribution coefficients of the qualified waterproof wood at the last collection moment for the preset number of times is recorded as the standard texture distribution coefficient;

[0038] For the weather resistance test of each batch of waterproof wood to be tested, the sequence composed of the texture distribution coefficients at each acquisition moment and all previous acquisition moments is used as the texture distribution sequence at each acquisition moment;

[0039] For each acquisition moment, the texture distribution sequence and the preset smoothing coefficient are used as input, and the exponential smoothing method is used to predict the texture distribution coefficient at the last acquisition moment as the prediction coefficient at each acquisition moment;

[0040] When the prediction coefficient is greater than the standard texture distribution coefficient, the performance of the current batch of waterproof wood is unqualified; otherwise, the performance of the current batch of waterproof wood is qualified.

[0041] In a second aspect, an embodiment of the present application also provides a performance testing system for moisture-proof and waterproof furniture materials, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0042] This application has at least the following beneficial effects:

[0043] This application calculates the color change coefficient based on the color change of the wood surface caused by the photodegradation phenomenon of waterproof wood under ultraviolet light, the gray distribution change caused by wood mildew and cracks, the time series change of color and the change time. This coefficient represents the detailed characteristics of the time series change of wood color, and improves the accuracy of judging the performance of waterproof wood according to color change; then, combined with the distribution characteristics of cracks and mildew on the surface of waterproof wood, the texture distribution coefficient is calculated. This coefficient takes into account the causes of the formation of wood cracks and mildew, as well as the positive correlation with the color change coefficient, and further improves the accuracy of judging the performance of waterproof wood according to the color and texture changes of wood; finally, the predicted value of the texture distribution coefficient is calculated according to the texture distribution coefficient at each acquisition moment, so as to judge the performance of waterproof wood. In this way, combined with the time series change characteristics of the color and texture of the surface of waterproof wood, non-destructive testing of waterproof wood is carried out, and the performance of waterproof wood is predicted and judged according to the intermediate characteristic changes, so as to determine whether it is necessary to end the test in advance to reduce unnecessary resource consumption, reduce testing costs, and improve performance testing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 A flowchart of a method for testing the performance of moisture-proof and waterproof furniture materials provided in one embodiment of the present application;

[0046] Figure 2 A diagram showing the relationship between the fitting curve and the straight line position provided for one embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of the performance detection method and system for moisture-proof and waterproof furniture materials proposed in the present application, its specific implementation method, structure, characteristics and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0048] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0049] The following is a detailed description of a specific scheme of a performance testing method and system for moisture-proof and waterproof furniture materials provided by the present application in conjunction with the accompanying drawings.

[0050] See also Figure 1 , which shows a flow chart of a method for testing the performance of moisture-proof and waterproof furniture materials provided by an embodiment of the present application, the method comprising the following steps:

[0051] Step S1, collecting the surface image of the waterproof wood and performing preprocessing.

[0052] Taking waterproof wood as an example of moisture-proof and waterproof furniture materials, a CMOS camera is used to collect surface images of waterproof wood in weather resistance experiments, once every 12 hours.

[0053] In order to reduce the interference of noise, median filtering is used to denoise the waterproof wood surface image. The median filtering technology is a well-known technology, and the specific process is not repeated here. The denoised waterproof wood surface images are recorded as S and S in the order of shooting. i ,i=1,2,…,n; grayscale the denoised waterproof wood surface image to obtain each grayscale surface image, denoted as G i .

[0054] Step S2, calculating the color change coefficient according to the color change of the wood surface caused by the photodegradation phenomenon of the waterproof wood and the temporal change of the color; and calculating the texture distribution coefficient in combination with the distribution characteristics of cracks and mildew spots on the waterproof wood surface.

[0055] Waterproof wood may be damaged in the natural environment, which will affect the performance of waterproof wood. First, when waterproof wood is exposed to ultraviolet rays for a long time, photodegradation will occur, which will change the color of the wood; secondly, waterproof wood will expand or shrink with the changes in temperature and humidity. Long-term expansion and contraction will cause cracks on the surface of waterproof wood; finally, when waterproof wood is in a humid environment for a long time, it will cause mold spots on the wood itself. Waterproof treatment can prevent this from happening, but this depends on the performance of the waterproof treatment. The better the performance, the lower the probability of mold spots. The earlier these changes appear on the waterproof wood and the more serious the changes, the worse the waterproof performance of the waterproof wood.

[0056] Furthermore, the color change coefficient is obtained according to the temporal change of the image grayscale and the difference change of the grayscale of adjacent images.

[0057] For the grayscale surface image at each acquisition moment, the mean of the grayscale values ​​of all pixels in the grayscale surface image is calculated as the grayscale mean of each grayscale surface image; for each acquisition moment, the absolute value of the difference between the grayscale mean of the grayscale surface image and the grayscale surface image at the previous acquisition moment is calculated as the grayscale difference at each moment, and the grayscale differences of all acquisition moments before each acquisition moment are arranged in chronological order to obtain the grayscale difference sequence at each acquisition moment.

[0058] Furthermore, the first-order difference sequence of the grayscale difference sequence at each acquisition moment is calculated, and a curve fitting is performed on the first-order difference sequence to obtain a fitting curve function, and a straight line fitting is performed to obtain a fitting straight line function.

[0059] For each pixel in each grayscale surface image, a 3×3 neighborhood centered on the pixel is taken, and the average of the grayscale differences between the other 8 pixels in the 3×3 neighborhood and the pixel is taken as the grayscale distribution value of each pixel. If a pixel cannot form a complete 3×3 neighborhood around it, it is calculated based on the pixels in the neighborhood without completion.

[0060] For each grayscale surface image, the grayscale distribution values ​​of all pixels in the grayscale surface image are arranged in order from left to right and from top to bottom to construct a grayscale distribution sequence of each grayscale surface image.

[0061] After photodegradation occurs, the color change of waterproof wood is a global change on the wood surface. The better the performance of waterproof wood, the stronger its surface's ability to resist ultraviolet rays and the lower the degree of photodegradation, that is, the color change is not obvious, and photodegradation will not cause continuous color changes. When the sensitive chemical components on the wood surface gradually reach a state of equilibrium due to photodegradation, the remaining components have a strong resistance to photodegradation, and the color change will tend to be stable. That is, during the waterproof wood performance test, the color change curve shows an increasing trend, and the growth rate changes from slow to fast to slow, such as Figure 2 .

[0062] When only photodegradation occurs, the global impact of photodegradation causes all pixels on the waterproof wood surface to experience the same degree of photodegradation, causing the grayscale value of each pixel on the waterproof wood surface to change to the same degree, so that the grayscale distribution of pixels at the same position in different grayscale surface images remains basically unchanged. When the performance of waterproof wood is poor, the wood will expand, shrink, and mold due to moisture and temperature. These phenomena will cause cracks or mildew on the wood surface, causing differences in the grayscale distribution of some pixels. The more pixels with different changes, the more abnormal the wood is and the more serious the abnormality is, that is, the worse the performance of waterproof wood is.

[0063] According to the above analysis, the color change coefficient at each acquisition moment is constructed, and the acquisition method is:

[0064] For the fitted straight line function and curve function at each acquisition moment, the absolute value of the difference between the fitted straight line function value and the curve function value at the same acquisition moment is calculated as the first absolute value of the difference. For each acquisition moment, the variance of all the first absolute values ​​of the difference is calculated as the material color change rate at each acquisition moment.

[0065] For each acquisition moment, the difference between the grayscale distribution sequence and the grayscale distribution sequence of the first acquisition moment is calculated as the grayscale distribution difference of each acquisition moment. In this embodiment, DTW distance is selected as the measure of the difference, and the implementer can select other measures according to the actual situation.

[0066] Based on the material color change rate and the grayscale distribution difference, the color change coefficient at each acquisition moment is obtained, and the color change coefficient is positively correlated with the material color change rate and the grayscale distribution difference, respectively.

[0067] It should be noted that the positive correlation means that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases.

[0068] Specifically, the present application proposes an embodiment, for each acquisition moment, the product of the material color change rate and the grayscale distribution difference is calculated as the color change coefficient at each acquisition moment.

[0069] Furthermore, the present application proposes another embodiment, in which, for each acquisition moment, the sum of the material color change rate and the grayscale distribution difference is calculated as a first sum, and the calculation result of an exponential function with a natural constant as the base and the first sum as the exponent is calculated as the color change coefficient at each acquisition moment.

[0070] It is understandable that when the performance of waterproof wood is better, the photodegradation phenomenon occurs later, so the color change difference between images taken at different times is smaller, that is, the color change speed is slower, so the function difference between the fitting straight line function and the fitting curve function obtained according to the first-order difference sequence at the same acquisition time is smaller, that is, the absolute value of the first difference is smaller, such as Figure 2 As shown in (1) in the figure, the better the waterproof wood performance is, the less water the wood absorbs, so that the expansion, contraction, and mold of the wood will not occur or will affect a smaller area, and thus the grayscale distribution of the image will be less affected, that is, the grayscale distribution difference is smaller, so when the waterproof wood performance is better, the color change coefficient A is smaller. On the contrary, the worse the waterproof wood performance is, the larger the color change coefficient A is.

[0071] exist Figure 2Among them, (1) is the case where the color change is small, that is, the grayscale changes slowly; (2) is the case where the color change is large, that is, the grayscale changes from slow to fast; (3) is the case where the color change is the largest, that is, the grayscale changes from slow to fast to slow; Among them, (1) the function difference between the straight line and the fitting curve at the same point is small and positive, so the variance is small; (2) the function difference between the straight line and the fitting curve at the same point is large and positive, so the variance is large; (3) the function difference between the straight line and the fitting curve at the same point is first positive and then negative, so the variance is large.

[0072] Furthermore, the texture distribution coefficient is calculated according to the relationship between the distribution of mildew spots and cracks in each image.

[0073] The above considerations from the perspective of color change consider the degree of color change of waterproof wood, thus measuring the performance of waterproof wood to a certain extent. However, color change is affected by many factors, and the above considerations can only reflect color changes under ideal conditions. In order to ensure the accuracy of measuring the performance of waterproof wood, further analysis is conducted in combination with the distribution characteristics of texture and edge lines.

[0074] When the performance of waterproof wood is good, it can prevent the wood itself from contacting or absorbing more moisture. When wood contacts or absorbs more moisture, it will expand and contract due to the absorption and evaporation of moisture, and then form cracks on the surface of the wood. The more cracks there are, the worse the performance of the waterproof wood may be. Secondly, when wood is damp or has more moisture, mildew will appear. In the weather resistance experiment, the temperature and humidity of the environment in which the waterproof wood is located are basically the same. Therefore, if mildew only appears near the cracks, it means that the mildew is caused by the large amount of moisture in the cracks; if the mildew is widely distributed on the waterproof wood and the correlation with the crack distribution is weak, it means that there is a problem with the performance of the waterproof wood itself.

[0075] 500 grayscale surface images of waterproof wood containing cracks during the waterproof wood performance test were collected as the sample set. The Labelme image labeling tool was used to annotate the cracks in all grayscale surface images in the sample set with bounding boxes. 80% of the grayscale surface images were randomly selected from the sample set as the training set, and the remaining 20% ​​of the grayscale surface images were used as the test set.

[0076] The grayscale surface image in the training set is taken as input, the YOLO model is adopted, the binary cross-entropy loss (BinaryCross-Entropy Loss) is used as the loss function, the Adam optimization algorithm and the Sigmoid activation function are used, and the coordinates of all bounding boxes in each grayscale surface image are output; the process of YOLO training and testing using the test set is a well-known technology, and the specific process is not repeated here.

[0077] After completing the YOLO model training, the grayscale surface image G i As input to the YOLO model, it outputs the coordinates of all bounding boxes in the image.

[0078] Furthermore, the Canny edge detection algorithm is used to obtain edge lines in each grayscale surface image. Canny edge detection is a well-known technology, and the specific process is not described in detail.

[0079] The longest edge line in each boundary box area is taken as the crack edge line, and each crack edge line is fitted using a straight line fitting algorithm to obtain each crack straight line.

[0080] Furthermore, for each pixel in each grayscale surface image, the variance of the grayscale values ​​of the 9 pixels in the 3×3 neighborhood centered on each pixel is calculated as the neighborhood grayscale fluctuation value of each pixel. If a complete 3×3 neighborhood cannot be formed around a certain pixel, it is calculated based on the pixels in the neighborhood without completion.

[0081] For pixels within the normal wood texture range, the grayscale around them is not completely consistent, but the difference between the grayscales is small; however, mold spots are often gray-black, dark green, and other colors that are very different from the wood itself, and the distribution of mold spots is often discontinuous, resulting in a large difference in grayscale around pixels in the mold spot area.

[0082] According to the above analysis, the neighborhood grayscale fluctuation values ​​of all pixels in each grayscale surface image are used as the input of the Otsu threshold segmentation algorithm, and the segmentation threshold is output. All pixels greater than the segmentation threshold are marked as mildew spots. The Otsu threshold segmentation algorithm is a well-known technology, and the specific process is not repeated here.

[0083] For each mildew spot pixel point, the distance between the mildew spot pixel point and each crack straight line is calculated, and the minimum value of the distance between the mildew spot pixel point and the crack straight line is taken as the crack distance of each mildew spot pixel point.

[0084] Use the k-means clustering algorithm to cluster all the mildew spots pixels to obtain clusters. In this embodiment, the number of clusters is set to 5, and the implementer can set other numbers according to actual conditions. The k-means clustering algorithm is a well-known technology, and the specific process is not repeated here.

[0085] For the grayscale surface image at each acquisition moment, the mean of the crack distances of all mildew pixels in the grayscale surface image is calculated as the first mean; the distance between the cluster centers of any two clusters in the grayscale surface image is calculated as the first distance, and the sum of all the first distances is used as the mildew distribution coefficient.

[0086] Furthermore, for the grayscale surface image at each acquisition moment, the texture distribution coefficient at each acquisition moment is obtained based on the number of crack lines, the first mean, the mildew distribution coefficient and the color change coefficient in the grayscale surface image. The calculation method is: calculate the result of an exponential function with a natural constant as the base and the sum of the number of crack lines and the first mean as the exponent as the first index, and take the product of the first index, the mildew distribution coefficient and the color change coefficient as the texture distribution coefficient at each acquisition moment.

[0087] It should be noted that when the performance of waterproof wood is good, the color change coefficient is small, the number of cracks on the wood is small, and the mildew appears around the cracks, that is, the number of crack lines and the first mean are small; secondly, the distribution range of mildew is small, that is, the mildew distribution is relatively concentrated, and the distance between the cluster centers of any two clusters is small, so the mildew distribution coefficient is small, so the texture distribution coefficient is small. Conversely, the worse the performance of waterproof wood, the larger the texture distribution coefficient.

[0088] Step S3, according to the method of step S2, the texture distribution coefficient at each acquisition moment is calculated, and the performance of the waterproof wood is judged according to the predicted value of the texture distribution coefficient.

[0089] According to the method in step S2, the texture distribution coefficient of the qualified waterproof wood in the standard weather resistance test at the last acquisition moment is obtained.

[0090] Furthermore, the texture distribution coefficients of the waterproof wood that passed γ times at the last collection time are counted, and the maximum value of all the texture distribution coefficients at the last collection time is recorded as the standard texture distribution coefficient. In this embodiment, the value of γ is 10.

[0091] For the weather resistance test of each batch of waterproof wood to be tested, the sequence composed of the texture distribution coefficients at each acquisition moment and all previous acquisition moments is used as the texture distribution sequence at each acquisition moment.

[0092] For each acquisition moment, the texture distribution sequence and the smoothing coefficient α are used as input, and the exponential smoothing method is used to predict the texture distribution coefficient of the last acquisition moment as the prediction coefficient of each acquisition moment. In this embodiment, the smoothing coefficient α is 0.3, and the exponential smoothing method is a well-known technology, and the specific process is not repeated.

[0093] When the prediction coefficient is greater than the standard texture distribution coefficient, it is considered that the performance of this batch of waterproof wood is unqualified; otherwise, the performance of the current batch of waterproof wood is qualified.

[0094] In this way, weather resistance tests can be carried out during the production of moisture-proof and waterproof furniture materials. When the prediction coefficient at a collection moment is greater than the standard texture distribution coefficient, it is determined that the performance of this batch of waterproof wood is unqualified, and the test is stopped directly without waiting for the end, thereby reducing the test time, which is conducive to reducing unnecessary resource and time consumption, reducing detection costs, and improving efficiency.

[0095] Based on the same inventive concept as the above method, an embodiment of the present application also provides a performance testing system for moisture-proof and waterproof furniture materials, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned performance testing methods for moisture-proof and waterproof furniture materials are implemented.

[0096] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] The various embodiments in the present application are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0098] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A performance testing method for moisture-proof and waterproof furniture materials, characterized in that: The method comprises the following steps: Acquire the grayscale surface image of the furniture material at each acquisition moment during the furniture material image acquisition process; Based on the difference between the grayscale values ​​of each grayscale surface image, a grayscale difference sequence at each acquisition moment is obtained; based on the distribution of the grayscale values ​​in the grayscale surface image, a grayscale distribution sequence at each acquisition moment is obtained; Based on the change trend of the grayscale difference sequence and the difference between the grayscale distribution sequences at each acquisition moment, the color change coefficient at each acquisition moment is obtained; Based on the edge features and gray value features in the gray surface image, all crack lines and mildew pixel points in the gray surface image at each acquisition time are obtained; based on the distance between the mildew pixel points and the crack line, the mildew distribution coefficient at each acquisition time is obtained; Based on the number of crack lines, the distance between the mildew spot pixel point and the crack line, the mildew spot distribution coefficient and the color change coefficient, the texture distribution coefficient at each acquisition time is obtained; Obtain prediction coefficients based on texture distribution coefficients to test the performance of home furnishing materials; The method for obtaining the grayscale difference sequence is: For the grayscale surface image at each acquisition moment, the grayscale mean of the grayscale surface image is calculated; for each acquisition moment, the absolute value of the difference between the grayscale surface image and the grayscale mean of the grayscale surface image at the previous acquisition moment is calculated as the grayscale difference at each moment, and the grayscale differences of all acquisition moments before each acquisition moment are arranged in chronological order to obtain the grayscale difference sequence at each acquisition moment; The method for obtaining the grayscale distribution sequence is: For each pixel in each grayscale surface image, for a neighborhood of a preset size centered on the pixel, the average of the grayscale value differences between other pixels in the neighborhood and the central pixel is taken as the grayscale distribution value of each pixel; For each grayscale surface image, the grayscale distribution values ​​of all pixels in the grayscale surface image are arranged in order from left to right and from top to bottom to construct a grayscale distribution sequence of each grayscale surface image; The method for obtaining the color change coefficient is: Based on the changing trend of the grayscale difference sequence, the material color change rate at each acquisition moment is obtained; For each acquisition moment, the difference between the grayscale distribution sequence and the grayscale distribution sequence at the first acquisition moment is calculated as the grayscale distribution difference at each acquisition moment; Based on the material color change rate and grayscale distribution difference, the color change coefficient at each acquisition moment is obtained, and the color change coefficient is positively correlated with the material color change rate and grayscale distribution difference respectively; The method for obtaining the crack straight line and the mildew spot pixel points is: Annotate the bounding box of each crack in the grayscale surface image at each acquisition moment; Use an edge detection algorithm to obtain edge lines in each grayscale surface image, take the longest edge line in each boundary box area as the crack edge line, and use a straight line fitting algorithm to fit each crack edge line to obtain each crack straight line; For each pixel point in each grayscale surface image, a neighborhood of a preset size centered on each pixel point is calculated, and the variance of the grayscale values ​​of all the pixels in the neighborhood is calculated as the neighborhood grayscale fluctuation value of each pixel point; Perform threshold segmentation on the neighborhood grayscale fluctuation values ​​of all pixels in each grayscale surface image, output the segmentation threshold, and mark all pixels greater than the segmentation threshold as mildew pixels; The method for obtaining the mildew spot distribution coefficient is: For each mildew spot pixel point, the distance between the mildew spot pixel point and each crack straight line is calculated, and the minimum value of all distances to the crack straight line is taken as the crack distance of each mildew spot pixel point; Clustering algorithms are used to cluster all the mildew spots pixels to obtain clusters; For the grayscale surface image at each acquisition moment, the distance between the cluster centers of any two clusters in the grayscale surface image is calculated as the first distance, and the sum of all the first distances is used as the mildew spot distribution coefficient; The method for obtaining the texture distribution coefficient is: For the grayscale surface image at each acquisition moment, the mean of the crack distances of all the mildew spot pixel points in the grayscale surface image is calculated as the first mean; the result of an exponential function with a natural constant as the base and the sum of the number of the crack straight lines and the first mean as the exponent is calculated as the first exponent, and the product of the first exponent, the mildew spot distribution coefficient and the color change coefficient is taken as the texture distribution coefficient at each acquisition moment; The method of obtaining a prediction coefficient based on the texture distribution coefficient and testing the performance of the home furnishing material includes: Obtain the grain distribution coefficient at the last acquisition moment in the standard weathering test for qualified waterproof wood; The maximum value of the texture distribution coefficients of the qualified waterproof wood at the last collection moment for a preset number of times is recorded as the standard texture distribution coefficient; For the weather resistance test of each batch of waterproof wood to be tested, the sequence composed of the texture distribution coefficients at each acquisition moment and all previous acquisition moments is used as the texture distribution sequence at each acquisition moment; For each acquisition moment, the texture distribution sequence and the preset smoothing coefficient are used as input, and the exponential smoothing method is used to predict the texture distribution coefficient at the last acquisition moment as the prediction coefficient at each acquisition moment; When the prediction coefficient is greater than the standard texture distribution coefficient, the performance of the current batch of waterproof wood is unqualified; otherwise, the performance of the current batch of waterproof wood is qualified.

2. A performance testing method for moisture-proof and waterproof furniture materials as claimed in claim 1, characterized in that: The method for obtaining the material color change rate is: Calculate the first-order difference sequence of the grayscale difference sequence at each acquisition moment, perform curve fitting on the first-order difference sequence to obtain a fitting curve function, and perform straight line fitting to obtain a fitting straight line function; For the fitted straight line function and curve function at each acquisition moment, the absolute value of the difference between the fitted straight line function value and the curve function value at the same acquisition moment is calculated as the first absolute value of the difference. For each acquisition moment, the variance of all the first absolute values ​​of the difference is calculated as the material color change rate at each acquisition moment.

3. A performance detection system for moisture-proof and waterproof furniture materials, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of a performance testing method for moisture-proof and waterproof furniture materials as described in any one of claims 1-2 are implemented.

Images