Textile water absorption detection system and detection method

Abstract

This application relates to a textile water absorption testing system and method. The method involves using a water droplet applicator to drip a predetermined amount of water droplets onto the center of a textile sample. A camera monitors the water droplet diffusion process, and deep learning-based computer vision technology is used to analyze dynamic video of the water droplet diffusion. The system then extracts the water droplet diffusion boundary features within each video frame. By measuring the difference between the water droplet diffusion states at each adjacent time point, the system determines the time it takes for the water droplet to be completely absorbed. After water absorption is complete, the system measures the weight change of the textile sample before and after water absorption to determine the water absorption rate of the textile sample. This method enables rapid testing of textile water absorption and provides a more accurate and comprehensive assessment of water absorption performance.

Description

Technical Field

[0001] The present application relates to the technical field of water absorption performance of textiles, and in particular, to a system and method for detecting water absorption performance of textiles. Background Art

[0002] In the textile industry, a textile's water absorption capacity is a key indicator of its quality and application. This performance not only directly impacts a textile's comfort, breathability, and durability, but also profoundly impacts its application in a variety of fields, including medical, sportswear, outdoor apparel, and household goods. Therefore, accurate and efficient testing of textile water absorption is crucial for product development, quality control, and market positioning.

[0003] Traditionally, the water absorption properties of textiles have been tested using the static immersion method, which involves completely immersing a textile sample in water for a period of time, then removing it and measuring the weight change to assess its water absorption. However, this method is difficult to accurately determine when the textile has reached water saturation, is time-consuming, and is not suitable for rapid testing or in-line quality control. Furthermore, after removing the textile sample, the removal of surface moisture is difficult to control, which can easily lead to inaccurate test results.

[0004] Therefore, an optimized method for testing the water absorption of textiles is desired. Summary of the Invention

[0005] This summary is provided to briefly introduce concepts that will be described in detail in the detailed description below. This summary is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to limit the scope of the claimed technical solution.

[0006] In a first aspect, the present application provides a method for detecting water absorbency of textiles, the method comprising:

[0007] Provide textile samples to be tested;

[0008] Using a water drop application device to drop a predetermined amount of water droplets on the center of the textile sample to be tested;

[0009] Use the camera to collect dynamic video of water droplet diffusion;

[0010] Determining the time it takes for the water drop to be completely absorbed based on the water droplet diffusion dynamic video;

[0011] After the water absorption is completed, the weight of the textile sample to be tested after water absorption is measured by a balance;

[0012] The water absorption rate of the textile sample to be tested is determined based on the weight after water absorption and the weight of the textile sample to be tested before water absorption.

[0013] Optionally, based on the water droplet diffusion dynamic video, the time when the water droplet is completely absorbed is determined, including: extracting the water droplet boundary features of each image frame in the water droplet diffusion dynamic video to obtain a time series of water droplet diffusion boundary feature maps; performing foreground saliency processing on the time series of water droplet diffusion boundary feature maps to obtain a time series of water droplet diffusion boundary foreground saliency feature maps; performing diffusion state change measurement on each water droplet diffusion boundary foreground saliency feature map in the time series of water droplet diffusion boundary foreground saliency feature maps to obtain a time series of diffusion state change measurement coefficients; and determining the time when the water droplet is completely absorbed based on a comparison between the time series of the diffusion state change measurement coefficients and a predetermined threshold.

[0014] Optionally, the water droplet boundary features of each image frame in the water droplet diffusion dynamic video are extracted to obtain a time series of water droplet diffusion boundary feature maps, including: performing grayscale processing on each image frame in the water droplet diffusion dynamic video to obtain a time series of water droplet diffusion grayscale maps; and performing boundary feature extraction on each water droplet diffusion grayscale map in the time series of water droplet diffusion grayscale maps to obtain a time series of water droplet diffusion boundary feature maps.

[0015] Optionally, boundary features are extracted for each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image to obtain the time series of the water droplet diffusion boundary feature images, including: each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image is passed through a water droplet boundary feature extractor based on a void convolutional neural network model to obtain the time series of the water droplet diffusion boundary feature images.

[0016] Optionally, the time series of the water droplet spread boundary feature map is subjected to foreground saliency processing to obtain a time series of water droplet spread boundary foreground salient feature maps, including: inputting the time series of the water droplet spread boundary feature map into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain a time series of the water droplet spread boundary foreground salient feature maps.

[0017] Optionally, the time series of the water droplet spread boundary feature map is input into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain a time series of the water droplet spread boundary foreground salient feature map, including: performing layer normalization processing on the water droplet spread boundary feature map to obtain a normalized water droplet spread boundary feature map; performing channel expansion based on point convolution and deep convolution encoding based on a hole convolution layer on the normalized water droplet spread boundary feature map to obtain a water droplet spread boundary deep convolution backup semantic feature map and a water droplet spread boundary deep convolution original semantic feature map. ; Input the water droplet spread boundary depth convolution original semantic feature map into the foreground gated mask module based on the Gelu function to obtain the water droplet spread boundary depth convolution gated mask weight feature map; calculate the position point multiplication between the water droplet spread boundary depth convolution gated mask weight feature map and the water droplet spread boundary depth convolution backup semantic feature map to obtain the gated mask foreground highlighted water droplet spread boundary feature map; perform point convolution-based channel shrinkage on the gated mask foreground highlighted water droplet spread boundary feature map to obtain the water droplet spread boundary foreground salient feature map.

[0018] Optionally, the diffusion state change measurement is performed on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients, including: performing global mean pooling processing on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of water droplet diffusion boundary foreground significant feature vectors; calculating the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vectors to obtain a time series of the diffusion state change measurement coefficients.

[0019] Optionally, the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector is calculated to obtain the time series of the diffusion state change measurement coefficient, including: calculating the positional difference between the next water droplet diffusion boundary foreground significant feature vector of the i-th water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector and the i-th water droplet diffusion boundary foreground significant feature vector to obtain the diffusion state change feature vector; calculating the positional division between the diffusion state change feature vector and the i-th water droplet diffusion boundary foreground significant feature vector to obtain a standardized diffusion state change feature vector; dividing the eigenvalue of the standardized diffusion state change feature vector by the maximum eigenvalue of the diffusion state change representation vector to obtain the diffusion state change measurement coefficient of the i-th water droplet diffusion boundary foreground significant feature vector.

[0020] Optionally, the water droplet complete absorption time is determined based on a comparison between the time series of the diffusion state change measurement coefficient and a predetermined threshold, including: when the diffusion state change measurement coefficient at the last time point in the time series of the diffusion state change measurement coefficient is less than or equal to a predetermined threshold, determining the current time point as the water droplet complete absorption time.

[0021] In a second aspect, the present application provides a textile water absorbency detection system, the system comprising:

[0022] A textile sample providing module to be tested, used for providing textile samples to be tested;

[0023] a water drop application module, configured to use a water drop application device to drop a predetermined amount of water droplets on the center of the textile sample to be tested;

[0024] The water droplet diffusion dynamic video acquisition module is used to capture the water droplet diffusion dynamic video through the camera;

[0025] a water droplet complete absorption time determination module, configured to determine the water droplet complete absorption time based on the water droplet diffusion dynamic video;

[0026] A weight measurement module, configured to measure the weight of the textile sample to be tested after water absorption by using a balance after water absorption is completed;

[0027] The water absorption rate determination module of the textile sample to be tested is used to determine the water absorption rate of the textile sample to be tested based on the weight after water absorption and the weight of the textile sample to be tested before water absorption.

[0028] The above technical solution uses a water droplet application device to drip a predetermined amount of water droplets onto the center of a textile sample. A camera monitors the water droplet diffusion process, and deep learning-based computer vision technology is used to analyze dynamic video of the water droplet diffusion. The feature of the water droplet diffusion boundary in each video image frame is mined. By measuring the difference in the water droplet diffusion state between two adjacent time points, the time required for complete water droplet absorption is determined. After water absorption is complete, the weight change of the textile sample before and after water absorption is measured to determine the water absorption rate of the textile sample. This method enables rapid testing of the water absorption performance of textiles and provides a more accurate and comprehensive water absorption performance assessment.

[0029] Other features and advantages of the present application will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The above and other features, advantages and aspects of the embodiments of the present application will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale. In the drawings:

[0031] Figure 1 The figure is a flow chart of a method for detecting water absorbency of textiles according to an exemplary embodiment.

[0032] Figure 2 is based on Figure 1 The illustrated embodiment shows a flow chart of step S104 of a method for detecting water absorbency of textiles.

[0033] Figure 3 The figure is a block diagram of a textile water absorbency detection system according to an exemplary embodiment.

[0034] Figure 4 It is a block diagram of an electronic device according to an exemplary embodiment.

[0035] Figure 5 This is a diagram showing an application scenario of a method for detecting water absorbency of textiles according to an exemplary embodiment. DETAILED DESCRIPTION

[0036] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0037] It should be understood that the various steps described in the method embodiments of the present application can be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present application is not limited in this respect.

[0038] As used herein, the term "including" and its variations are open-ended, i.e., "including but not limited to." The term "based on" means "based, at least in part, on." The term "one embodiment" means "at least one embodiment," the term "another embodiment" means "at least one additional embodiment," and the term "some embodiments" means "at least some embodiments." Other terms are defined in the following description.

[0039] It should be noted that the concepts of "first" and "second" mentioned in this application are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0040] It should be noted that the modifications of "one" and "multiple" mentioned in this application are illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0041] The names of the messages or information exchanged between multiple devices in the embodiments of the present application are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0042] The specific implementation of this application is described in detail below with reference to the accompanying drawings.

[0043] To address the above technical issues, the technical concept of this application is to use a water droplet application device to drip a predetermined amount of water droplets at the center of a textile sample, while using a camera to monitor the water droplet diffusion process. Deep learning-based computer vision technology is then used to analyze the dynamic video of the water droplet diffusion, mining the water droplet diffusion boundary features in each video image frame. By measuring the difference in the water droplet diffusion state between each two adjacent time points, the time required for the water droplet to be completely absorbed is determined. After the water absorption is completed, the weight change of the textile sample before and after the water absorption is completed is measured to determine the water absorption rate of the textile sample. This allows for rapid testing of the water absorption performance of textiles and provides a more accurate and comprehensive water absorption performance evaluation.

[0044] Figure 1 FIG. 1 is a flow chart of a method for detecting water absorbency of textiles according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0045] Step S101: providing a textile sample to be tested;

[0046] Step S102: using a water droplet applying device to drop a predetermined amount of water droplets on the center of the textile sample to be tested;

[0047] Step S103: Capture a dynamic video of water droplet diffusion through a camera;

[0048] Step S104: determining the time required for the water droplets to be completely absorbed based on the water droplet diffusion dynamic video;

[0049] Step S105: After the water absorption is completed, the weight of the textile sample to be tested is measured by a balance;

[0050] Step S106: Determine the water absorption rate of the textile sample to be tested based on the weight after water absorption and the weight of the textile sample to be tested before water absorption.

[0051] Figure 2 is based on Figure 1 The embodiment shown is a flowchart of step S104 of a method for detecting water absorption of textiles, as shown in FIG. Figure 2 As shown, step S104, determining the time for complete absorption of water droplets based on the water droplet diffusion dynamic video, includes:

[0052] Step S1041: extracting water droplet boundary features of each image frame in the water droplet diffusion dynamic video to obtain a time series of water droplet diffusion boundary feature maps;

[0053] Step S1042: performing foreground saliency processing on the time series of the water droplet diffusion boundary feature map to obtain a time series of water droplet diffusion boundary foreground saliency feature maps;

[0054] Step S1043, performing diffusion state change measurement on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients;

[0055] Step S1044: Determine the complete absorption time of the water drop based on the comparison between the time series of the diffusion state change measurement coefficient and a predetermined threshold.

[0056] Specifically, first, a predetermined amount of water droplets is dripped into the center of the textile sample through a water droplet application device to ensure that the initial conditions of the water droplets are consistent during each test, thereby improving the repeatability and accuracy of the test results. Then, a high-resolution camera is used to capture and record the entire process of water droplet diffusion on the textile surface in real time, generating a dynamic video of water droplet diffusion, which intuitively displays the diffusion behavior of water droplets on the textile surface, so as to facilitate the use of artificial intelligence technology to analyze the time it takes for the water droplets to be completely absorbed in real time, thereby preventing detection errors caused by water evaporation over a long period of time and improving detection efficiency. After the water droplets are completely absorbed, a balance is used to measure the weight of the textile sample after water absorption. At the same time, the weight of the sample before water absorption is used as a comparison benchmark, and the water absorption rate of the textile sample can be obtained by calculating the difference between the weight after water absorption and the weight before water absorption as a percentage of the weight before water absorption.

[0057] In one embodiment of the present application, water droplet boundary features of each image frame in the water droplet diffusion dynamic video are extracted to obtain a time series of water droplet diffusion boundary feature maps, including: grayscale processing of each image frame in the water droplet diffusion dynamic video to obtain a time series of water droplet diffusion grayscale maps; boundary feature extraction of each water droplet diffusion grayscale map in the time series of water droplet diffusion grayscale maps to obtain a time series of water droplet diffusion boundary feature maps.

[0058] More specifically, in the process of determining the time it takes for a water droplet to be completely absorbed based on the water droplet diffusion dynamic video, it is considered that the water droplet diffusion dynamic video may be affected by various factors such as lighting conditions and camera white balance, resulting in deviations in the color of the image frames in the video, thereby affecting the accuracy of the subsequent water droplet boundary feature extraction. Therefore, in the technical solution of the present application, each image frame in the water droplet diffusion dynamic video is first grayscale processed to obtain a time series of water droplet diffusion grayscale images. Through image grayscale processing, the image color information can be effectively removed, and only the brightness information is retained, thereby highlighting the contrast between the water droplet diffusion area and the textile background, making the subsequent water droplet boundary feature extraction more accurate and efficient.

[0059] Next, in order to achieve accurate detection of water droplet boundary features, this application uses a dilated convolutional neural network model as a water droplet boundary feature extractor to process each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image separately. Those skilled in the art should know that the dilated convolutional neural network can effectively capture a wider range of spatial information in the image by introducing a dilated structure in the convolution kernel to increase the feature receptive field, thereby fully capturing the boundary contour information of the water droplet diffusion in the water droplet diffusion grayscale image and generating a time series of water droplet diffusion boundary feature maps.

[0060] In one embodiment of the present application, boundary features are extracted for each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image to obtain the time series of the water droplet diffusion boundary feature image, including: each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image is passed through a water droplet boundary feature extractor based on a void convolutional neural network model to obtain the time series of the water droplet diffusion boundary feature image.

[0061] Then, in order to further highlight the boundary contour of the water droplet diffusion and remove the interference of background information on the water droplet boundary detection, this application introduces a feature foreground mask salient device based on a convolutional gated feedforward mechanism to perform mask processing on the time series of the water droplet diffusion boundary feature map to generate a time series of water droplet diffusion boundary foreground salient feature maps. Specifically, the feature foreground mask salient device first performs layer normalization processing on the water droplet diffusion boundary feature map to eliminate the scale difference between layers. Subsequently, deep convolution coding is performed on it to fully extract the spatial and depth information of the water droplet diffusion boundary feature map as the basis for feature foreground salient processing. Then, based on the gating mechanism, adaptive feature learning is performed on the water droplet diffusion boundary features after deep convolution coding to generate a feature mask, weightedly emphasize the boundary features, and suppress background noise. Finally, a point convolution operation is performed on the masked weighted feature map to achieve channel shrinkage, so that its feature dimension is consistent with the original feature map, thereby restoring a high-resolution water droplet diffusion boundary foreground salient feature map.

[0062] In one embodiment of the present application, the time series of the water droplet diffusion boundary feature map is subjected to foreground saliency processing to obtain a time series of the water droplet diffusion boundary foreground salient feature map, including: inputting the time series of the water droplet diffusion boundary feature map into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain a time series of the water droplet diffusion boundary foreground salient feature map.

[0063] Furthermore, in one embodiment of the present application, the time series of the water droplet spread boundary feature map is input into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain a time series of the water droplet spread boundary foreground salient feature map, including: performing layer normalization processing on the water droplet spread boundary feature map to obtain a normalized water droplet spread boundary feature map; performing channel expansion based on point convolution and depth convolution encoding based on a hole convolution layer on the normalized water droplet spread boundary feature map to obtain a water droplet spread boundary depth convolution backup semantic feature map and a water droplet spread boundary depth convolution. Original semantic feature map; inputting the water droplet spread boundary depth convolution original semantic feature map into the foreground gated mask module based on the Gelu function to obtain the water droplet spread boundary depth convolution gated mask weight feature map; calculating the position point multiplication between the water droplet spread boundary depth convolution gated mask weight feature map and the water droplet spread boundary depth convolution backup semantic feature map to obtain the gated mask foreground highlighting water droplet spread boundary feature map; performing channel shrinkage based on point convolution on the gated mask foreground highlighting water droplet spread boundary feature map to obtain the water droplet spread boundary foreground salient feature map.

[0064] Specifically, the water droplet diffusion boundary feature map is processed using the following foreground mask highlighting formula to obtain the water droplet diffusion boundary foreground salient feature map, wherein the foreground mask highlighting formula is:

[0065] F ln =LayerNormalization(F i )

[0066] F m =Conv 3×3,2 (Conv 1×1 (F ln ))

[0067] F r =Conv 3×3,2 (Conv 1×1 (F ln ))

[0068] F a =mask[Gelu(F m )]

[0069]

[0070] F t =F a ⊙F r

[0071] F′ i =Conv 1×1 (F t )

[0072] Among them, F i represents the water droplet diffusion boundary feature map, Layer Normalization (·) represents the layer normalization operation, F ln Represents the normalized water droplet diffusion boundary feature map, Conv 1×1 Represents point convolution, Conv 3×3,2 represents a 3×3 dilated convolution with a dilation number of 2, F m Represents the original semantic feature map of the water droplet diffusion boundary depth convolution, F r represents the depth convolution backup semantic feature map of the water droplet diffusion boundary, Gelu represents the GELU activation function, mask(·) represents the gated mask processing, θ is the preset gate threshold, F a represents the water droplet diffusion boundary depth convolution gated mask weight feature map, ⊙ represents the position point multiplication operation, F t Denotes the gated mask foreground highlighting the water droplet diffusion boundary feature map, F′ i Represents the foreground salient feature map of the water droplet diffusion boundary.

[0073] Furthermore, in order to improve the efficiency of feature representation, global mean pooling processing is further performed on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to integrate the spatial information of the water droplet diffusion boundary in each image frame and ignore local irregular changes, so as to obtain a more stable time series of water droplet diffusion boundary foreground significant feature vectors and improve the generalization ability of the model.

[0074] Next, to further analyze the diffusion dynamics of water droplets from initial application to complete absorption, the technical solution of this application calculates the diffusion state change metric coefficient of each water droplet diffusion boundary foreground salient feature vector in the time series of the water droplet diffusion boundary foreground salient feature vector, thereby quantifying the speed and extent of water droplet diffusion on the textile. Specifically, in the process of calculating the real-time diffusion state change metric coefficient, the characteristic difference between the water droplet diffusion boundary foreground salient feature vectors at two adjacent time points is first calculated to quantify the change amplitude of the water droplet diffusion state. The obtained difference vector is then divided by the water droplet diffusion boundary foreground salient feature vector at the previous time point by position to reveal the change rate of the water droplet diffusion state and achieve data normalization. Furthermore, the ratio between the characteristic mean and maximum value of the standardized difference vector is used as the diffusion state change metric coefficient, thereby reflecting the diffusion dynamics of water droplets at different time points.

[0075] In one embodiment of the present application, a diffusion state change measurement is performed on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients, including: performing global mean pooling processing on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of water droplet diffusion boundary foreground significant feature vectors; calculating the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vectors to obtain a time series of the diffusion state change measurement coefficients.

[0076] Furthermore, in one embodiment of the present application, the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector is calculated to obtain the time series of the diffusion state change measurement coefficient, including: calculating the positional difference between the next water droplet diffusion boundary foreground significant feature vector of the i-th water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector and the i-th water droplet diffusion boundary foreground significant feature vector to obtain the diffusion state change feature vector; calculating the positional division between the diffusion state change feature vector and the i-th water droplet diffusion boundary foreground significant feature vector to obtain a standardized diffusion state change feature vector; dividing the eigenvalue of the standardized diffusion state change feature vector by the maximum eigenvalue of the diffusion state change representation vector to obtain the diffusion state change measurement coefficient of the i-th water droplet diffusion boundary foreground significant feature vector.

[0077] Specifically, the time series of the water droplet diffusion boundary foreground significant feature vectors is processed using the following diffusion state change measurement formula to obtain the time series of the diffusion state change measurement coefficients, wherein the diffusion state change measurement formula is:

[0078]

[0079] Among them, V i and V i+1 denote the i-th and i+1-th water droplet diffusion boundary foreground significant feature vectors in the time series of the water droplet diffusion boundary foreground significant feature vectors, respectively. average(·) denotes the mean calculation, max(·) denotes the maximum value calculation, and D i is the diffusion state change measurement coefficient of the foreground significant feature vector of the i-th water droplet diffusion boundary.

[0080] Subsequently, the complete absorption time of the water droplet is determined based on a comparison between the time series of the diffusion state change metric coefficient and a predetermined threshold. It should be understood that in the initial diffusion phase, due to the interaction between the water droplet and the textile surface, the water droplet diffuses rapidly, and the value of the diffusion state change metric coefficient is generally large. As the water droplet is gradually absorbed by the textile, the diffusion rate gradually slows, and the value of the diffusion state change metric coefficient also gradually decreases. Therefore, when the diffusion state change metric coefficient falls below the predetermined threshold, the water droplet can be considered to have been completely absorbed by the textile. The moment when the diffusion state change metric coefficient first falls below the predetermined threshold is determined as the complete absorption time of the water droplet. After determining the complete absorption time of the water droplet, the weight of the textile sample after water absorption can be measured using a balance and compared with the weight of the sample before water absorption to calculate the water absorption rate, which reflects the textile's ability to absorb and retain water.

[0081] In one embodiment of the present application, the water droplet complete absorption time is determined based on a comparison between the time series of the diffusion state change measurement coefficient and a predetermined threshold, including: when the diffusion state change measurement coefficient of the last time point in the time series of the diffusion state change measurement coefficient is less than or equal to the predetermined threshold, determining the current time point as the water droplet complete absorption time.

[0082] In the embodiment of the present application, it is considered that each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map respectively represents the image semantic foreground significant features of each water droplet diffusion grayscale map in the time series of the water droplet diffusion grayscale map. When the global mean pooling processing is performed on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map, although the amount of calculation is reduced, the time series of the water droplet diffusion boundary foreground significant feature vector will also have information structure collapse caused by feature dimensionality reduction relative to the original feature distribution, which will reduce the calculation accuracy of the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector.

[0083] Therefore, in the technical solution of the present application, before performing global mean pooling processing on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map, it also includes: performing feature optimization on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain an optimized time series of water droplet diffusion boundary foreground significant feature map.

[0084] The feature optimization process includes the following steps:

[0085] Calculating a feature mean of the water droplet diffusion boundary foreground salient feature map, and dividing the feature mean by a difference between a maximum eigenvalue and a minimum eigenvalue of the water droplet diffusion boundary foreground salient feature map to obtain a water droplet diffusion boundary foreground salient distribution representation value;

[0086] The water droplet diffusion boundary foreground saliency distribution characterization value is subtracted from one and divided by the water droplet diffusion boundary foreground saliency distribution characterization value to obtain a water droplet diffusion boundary foreground saliency distribution modulation value;

[0087] activating the water droplet diffusion boundary foreground salient feature map through a probabilistic function to obtain a probabilistic water droplet diffusion boundary foreground salient feature map;

[0088] performing pointwise subtraction of the probabilistic water droplet spread boundary foreground salient feature map and the water droplet spread boundary foreground salient distribution modulation value, taking the absolute value and calculating the negative of the logarithmic value with base 2 to obtain a probabilistic water droplet spread boundary foreground salient distribution modulation information feature map;

[0089] After dividing the water droplet spread boundary foreground saliency distribution characterization value by one minus the difference between each eigenvalue of the probabilistic water droplet spread boundary foreground saliency feature map, summing all eigenvalues ​​of the probabilistic water droplet spread boundary foreground saliency feature map, and dividing the sum by the scale of the water droplet spread boundary foreground saliency feature map to obtain a probabilistic water droplet spread boundary foreground saliency distribution modulation bias value;

[0090] The optimized water droplet spread boundary foreground salient feature map is obtained by multiplying the probabilistic water droplet spread boundary foreground salient distribution modulation information feature map and the probabilistic water droplet spread boundary foreground salient distribution modulation bias value with the weight as a hyperparameter.

[0091] The optimized water droplet diffusion boundary foreground salient feature map is expressed as:

[0092]

[0093] And among them:

[0094]

[0095] in, represents the feature mean of the foreground salient feature map of the water droplet diffusion boundary, f max and f min represents the maximum eigenvalue and minimum eigenvalue of the water droplet diffusion boundary foreground salient feature map, respectively; p represents the water droplet diffusion boundary foreground salient distribution representation value; F represents the probabilistic water droplet diffusion boundary foreground salient feature map obtained by activating the water droplet diffusion boundary foreground salient feature map through the probabilistic function; f represents the probabilistic water droplet diffusion boundary foreground salient feature map obtained by activating the water droplet diffusion boundary foreground salient feature map through the probabilistic function; i represents the i-th eigenvalue of the probabilistic water droplet spread boundary foreground salient feature map, log represents the logarithmic function with base 2, ε is the weight as a hyperparameter, S is the scale of the water droplet spread boundary foreground salient feature map, that is, the width of the feature matrix of the water droplet spread boundary foreground salient feature map multiplied by the height and then multiplied by the number of channels of the water droplet spread boundary foreground salient feature map, F' represents the optimized water droplet spread boundary foreground salient feature map, represents positional subtraction, and ⊕ represents positional addition.

[0096] That is, in the preferred example described above, the eigenvalue-based probability information distribution planning of the water droplet diffusion boundary foreground salient feature map is performed by using the Bernoulli probability modulation distribution of the water droplet diffusion boundary foreground salient feature map relative to the eigenvalue distribution, and the probability reverse mapping of the overall probability characteristics of the water droplet diffusion boundary foreground salient feature map is used as an expanded coverage of the set mapping space of the water droplet diffusion boundary foreground salient feature map to independently understand the interaction path between the intuitive probability information distribution and the abstract probability space mapping of the water droplet diffusion boundary foreground salient feature map, so as to avoid the adverse effects on the calculation of the subsequent diffusion state change measurement coefficient by avoiding the information collapse caused by the global mean pooling process of the water droplet diffusion boundary foreground salient feature map.

[0097] In summary, the above scheme uses a water droplet applicator to drip a predetermined amount of water droplets onto the center of a textile sample. A camera monitors the water droplet diffusion process, and deep learning-based computer vision technology is used to analyze the dynamic video of the water droplet diffusion. This data mines the water droplet diffusion boundary features in each video image frame. By measuring the difference in the water droplet diffusion state between two adjacent time points, the time it takes for the water droplet to be completely absorbed is determined. After water absorption is complete, the weight change of the textile sample before and after water absorption is measured to determine the water absorption rate of the textile sample. This method enables rapid testing of the water absorption properties of textiles and provides a more accurate and comprehensive water absorption performance assessment.

[0098] Figure 3 FIG. 1 is a block diagram of a textile water absorption detection system according to an exemplary embodiment. Figure 3 As shown, the system 200 includes:

[0099] The textile sample providing module 201 is used to provide the textile sample to be tested;

[0100] A water drop application module 202 is configured to use a water drop application device to drop a predetermined amount of water droplets on the center of the textile sample to be tested;

[0101] The water droplet diffusion dynamic video acquisition module 203 is used to acquire the water droplet diffusion dynamic video through a camera;

[0102] A water droplet complete absorption time determination module 204 is configured to determine the water droplet complete absorption time based on the water droplet diffusion dynamic video;

[0103] A weight measurement module 205 is configured to measure the weight of the textile sample after water absorption by using a balance after water absorption is completed;

[0104] The water absorption rate determination module 206 of the textile sample to be tested is configured to determine the water absorption rate of the textile sample to be tested based on the weight after water absorption and the weight before water absorption of the textile sample to be tested.

[0105] Reference below Figure 4 , which shows a schematic structural diagram of an electronic device 600 suitable for implementing an embodiment of the present application. The terminal device in the embodiment of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0106] like Figure 4 As shown, the electronic device 600 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage device 608 into a random access memory (RAM) 603. Various programs and data required for the operation of the electronic device 600 are also stored in the RAM 603. The processing device 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0107] Typically, the following devices may be connected to the I / O interface 605: an input device 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 608 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 609. The communication device 609 may allow the electronic device 600 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 600 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead.

[0108] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present application are performed.

[0109] It should be noted that the computer-readable medium mentioned above in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or device. In this application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. This propagated data signal can take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0110] In some embodiments, the client and server can communicate using any currently known or future developed network protocol, such as HTTP (HyperText Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0111] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0112] Computer program code for performing the operations of the present application can be written in one or more programming languages, or a combination thereof, including, but not limited to, object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0113] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0114] The modules described in the embodiments of this application may be implemented in software or hardware. In some cases, the name of a module does not necessarily limit the module itself. For example, a test parameter acquisition module may also be described as a "module for acquiring device test parameters corresponding to a target device."

[0115] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0116] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] Figure 5 FIG. 1 is an application scenario diagram of a method for detecting water absorption of textiles according to an exemplary embodiment. Figure 5 As shown, in this application scenario, first, a textile sample to be tested is provided (for example, Figure 5 C1 shown in the figure); capture the dynamic video of water droplet diffusion through the camera (for example, Figure 5 Then, the obtained textile sample to be tested and the water droplet diffusion dynamic video are input to a server deployed with a textile water absorption detection algorithm (for example, Figure 5 In S) shown in , the server is capable of processing the textile sample to be detected and the water droplet diffusion dynamic video based on a textile water absorption detection algorithm to determine the water absorption rate of the textile sample to be detected.

[0118] The above description is merely a preferred embodiment of the present application and an illustration of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this application is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but also encompasses other technical solutions formed by any combination of the above-mentioned technical features or their equivalents without departing from the above-mentioned disclosed concepts. For example, a technical solution formed by replacing the above-mentioned features with (but not limited to) technical features with similar functions disclosed in this application.

[0119] In addition, although adopting specific order to describe each operation, this should not be interpreted as requiring these operations to be executed in the specific order shown or in sequential order.Under certain environment, multitasking and parallel processing may be advantageous.Similarly, although comprising some specific implementation details in the above discussion, these should not be interpreted as limiting the scope of the application.Some features described in the context of separate embodiment can also be implemented in a single embodiment in combination.On the contrary, the various features described in the context of a single embodiment also can be implemented in multiple embodiments individually or in the mode of any suitable subcombination.

[0120] Although the subject matter has been described using language specific to structural features and / or method logic, it should be understood that the subject matter as defined is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely exemplary implementations. Regarding the apparatus in the above-described embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method and will not be further elaborated here.

Claims

1. A method for detecting water absorption of textiles, characterized in that: include: Provide textile samples to be tested; Using a water drop application device to drop a predetermined amount of water droplets on the center of the textile sample to be tested; Use the camera to collect dynamic video of water droplet diffusion; Determining the time it takes for the water drop to be completely absorbed based on the water droplet diffusion dynamic video; After the water absorption is completed, the weight of the textile sample to be tested after water absorption is measured by a balance; Determining the water absorption rate of the textile sample to be tested based on the weight after water absorption and the weight of the textile sample to be tested before water absorption; The method of determining the time required for complete absorption of the water droplets based on the water droplet diffusion dynamic video includes: Extracting water drop boundary features of each image frame in the water drop diffusion dynamic video to obtain a time series of water drop diffusion boundary feature maps; performing foreground saliency processing on the time series of the water droplet diffusion boundary feature map to obtain a time series of water droplet diffusion boundary foreground saliency feature maps; Performing diffusion state change measurement on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients; The water droplet complete absorption time is determined based on a comparison between the time series of the diffusion state change metric coefficient and a predetermined threshold.

2. The method for detecting water absorption of textiles according to claim 1, wherein: Extracting water drop boundary features of each image frame in the water drop diffusion dynamic video to obtain a time series of water drop diffusion boundary feature maps, including: Performing grayscale processing on each image frame in the water droplet diffusion dynamic video to obtain a time series of water droplet diffusion grayscale images; Boundary feature extraction is performed on each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image to obtain the time series of the water droplet diffusion boundary feature image.

3. The method for detecting water absorption of textiles according to claim 2, wherein: Extracting boundary features of each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image to obtain the time series of the water droplet diffusion boundary feature image includes: Each water droplet diffusion grayscale image in the time series of the water droplet diffusion grayscale image is respectively passed through a water droplet boundary feature extractor based on a hole convolutional neural network model to obtain the time series of the water droplet diffusion boundary feature image.

4. The method for detecting water absorption of textiles according to claim 3, wherein: Performing foreground saliency processing on the time series of the water droplet diffusion boundary feature map to obtain a time series of water droplet diffusion boundary foreground saliency feature maps, including: The time series of the water droplet diffusion boundary feature map is input into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain the time series of the water droplet diffusion boundary foreground salient feature map.

5. The method for detecting water absorption of textiles according to claim 4, wherein: Inputting the time series of the water droplet spread boundary feature map into a feature foreground mask salient device based on a convolutional gated feedforward mechanism to obtain the time series of the water droplet spread boundary foreground salient feature map, comprising: performing layer normalization processing on the water droplet spread boundary feature map to obtain a normalized water droplet spread boundary feature map; Performing point convolution-based channel expansion and dilated convolutional layer-based depth convolution coding on the normalized water droplet spread boundary feature map to obtain a water droplet spread boundary depth convolution backup semantic feature map and a water droplet spread boundary depth convolution original semantic feature map; The water droplet diffusion boundary depth convolution original semantic feature map is input based on The foreground gated mask module of the function is used to obtain the water droplet diffusion boundary depth convolution gated mask weight feature map; Calculating the position-wise multiplication between the water droplet spread boundary depth convolution gated mask weight feature map and the water droplet spread boundary depth convolution backup semantic feature map to obtain a gated mask foreground highlighting water droplet spread boundary feature map; The gated mask foreground salient water droplet diffusion boundary feature map is subjected to channel shrinkage based on point convolution to obtain the water droplet diffusion boundary foreground salient feature map.

6. The method for detecting water absorption of textiles according to claim 5, characterized in that: Performing diffusion state change measurement on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients includes: performing global mean pooling processing on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of water droplet diffusion boundary foreground significant feature vectors; The diffusion state change metric coefficient of each water droplet diffusion boundary foreground salient feature vector in the time series of the water droplet diffusion boundary foreground salient feature vector is calculated to obtain the time series of the diffusion state change metric coefficient.

7. The method for detecting water absorption of textiles according to claim 6, wherein: Calculating the diffusion state change measurement coefficient of each water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector to obtain the time series of the diffusion state change measurement coefficient includes: Calculating a positional difference between a water droplet diffusion boundary foreground significant feature vector subsequent to an i-th water droplet diffusion boundary foreground significant feature vector in the time series of the water droplet diffusion boundary foreground significant feature vector and the i-th water droplet diffusion boundary foreground significant feature vector to obtain a diffusion state change feature vector; Calculating the position-based division between the diffusion state change feature vector and the i-th water droplet diffusion boundary foreground salient feature vector to obtain a normalized diffusion state change feature vector; The characteristic mean of the standardized diffusion state change characteristic vector is divided by the maximum eigenvalue of the diffusion state change representation vector to obtain the diffusion state change measurement coefficient of the i-th water droplet diffusion boundary foreground significant characteristic vector.

8. The method for detecting water absorption of textiles according to claim 7, wherein: Determining the complete absorption time of the water droplet based on a comparison between the time series of the diffusion state change measurement coefficient and a predetermined threshold value includes: When the diffusion state change metric coefficient at the last time point in the time series of the diffusion state change metric coefficient is less than or equal to a predetermined threshold, the current time point is determined to be the complete absorption time of the water droplet.

9. A textile water absorption detection system, characterized in that: include: A textile sample providing module to be tested, used for providing textile samples to be tested; a water drop application module, configured to use a water drop application device to drop a predetermined amount of water droplets on the center of the textile sample to be tested; The water droplet diffusion dynamic video acquisition module is used to capture the water droplet diffusion dynamic video through the camera; a water droplet complete absorption time determination module, configured to determine the water droplet complete absorption time based on the water droplet diffusion dynamic video; A weight measurement module, configured to measure the weight of the textile sample to be tested after water absorption by using a balance after water absorption is completed; a water absorption rate determination module for the textile sample to be tested, configured to determine the water absorption rate of the textile sample to be tested based on the weight after water absorption and the weight before water absorption of the textile sample to be tested; The water droplet complete absorption time determination module includes: Extracting water drop boundary features of each image frame in the water drop diffusion dynamic video to obtain a time series of water drop diffusion boundary feature maps; performing foreground saliency processing on the time series of the water droplet diffusion boundary feature map to obtain a time series of water droplet diffusion boundary foreground saliency feature maps; Performing diffusion state change measurement on each water droplet diffusion boundary foreground significant feature map in the time series of the water droplet diffusion boundary foreground significant feature map to obtain a time series of diffusion state change measurement coefficients; The water droplet complete absorption time is determined based on a comparison between the time series of the diffusion state change metric coefficient and a predetermined threshold.

Images