Intelligent Detection Method for Broken Textile Fibers Based on Ultrasonic Imaging Analysis

Abstract

The present invention relates to the field of fiber detection, and discloses an intelligent detection method for broken textile fibers based on ultrasonic imaging analysis, which is used to solve the problem that when analyzing broken textile fibers through ultrasonic imaging, the broken fiber signals caused by small or local wear due to interference of ultrasonic signals are ignored. The method includes: using an ultrasonic imaging device to obtain the operation state image of the textile fiber, analyzing the operation state image of the textile fiber to obtain a fiber break index, making a first fiber break judgment according to the fiber break index. If the first fiber break judgment is that the fiber is broken, a break warning reminder is directly issued; if the first fiber break judgment is that the fiber is normal, the current environmental data is collected, the ultrasonic influence index is analyzed, the actual break judgment threshold is obtained according to the ultrasonic influence index, and a second fiber break judgment is made, effectively reducing the situation where the break signal is ignored due to the smallness of fiber breakage.

Description

Technical Field

[0001] The present invention relates to the field of fiber detection, and more particularly to an intelligent detection method for broken textile fibers based on ultrasonic imaging analysis. Background Art

[0002] Ultrasonic imaging analysis is a non-contact imaging technology based on ultrasonic signals, used to generate images by analyzing the propagation, reflection, and attenuation characteristics of ultrasonic waves in a medium. The ultrasonic waves are emitted by a transmitter and will be reflected or scattered when encountering objects with different density or elastic characteristics. The receiver captures the echo signals and reconstructs them into two-dimensional or three-dimensional images through algorithms. Ultrasonic imaging evaluates the structure and state of an object by analyzing the characteristics of the generated images.

[0003] With the rapid development of the textile industry, the detection of broken fibers has become an important link to ensure production efficiency and textile quality. The existing intelligent detection method for broken textile fibers based on ultrasonic imaging analysis is to obtain the echo signals during the operation of the fibers in real time through ultrasonic sensors, and use imaging algorithms to generate images of the broken areas of the fibers. The detection system combines image processing techniques (such as edge detection, morphological analysis, etc.) to identify the break points and discontinuous areas of the fibers, and realizes the automatic identification and alarm of broken textile fibers.

[0004] For example, a multi-region visual detection and monitoring system and method for a spinning frame disclosed in the invention patent announcement with the publication number of CN113174667B includes: judging the position and time of single-spindle yarn breakage through visual recognition, judging the position and time of roller winding through visual recognition, judging the position and time of empty roving bobbins, and breaking the roving. It preferably further includes judging the spindles with relatively large rotational speed deviations through a stroboscope in cooperation with visual recognition. The advantages of the present invention are: reasonable structural design, capable of effectively detecting situations such as broken yarn in the winding area, cotton sliver winding around the front roller in the front roller area, and the roving being used up in the roving area, automatically controlling the stop and notifying the doffer to handle, with high intelligence, and preferably having a spindle speed detection function, capable of timely repairing the faulty single spindles, effectively ensuring the continuity of production and improving production efficiency.

[0005] However, in the process of implementing the technical solutions of the invention in the above application embodiments, it is found that the above technologies have at least the following technical problems:

[0006] In practical applications, due to the complex textile production environment, signal interference problems are inevitable. Signal interference may lead to a decline in the quality of ultrasonic imaging. In this case, the fiber break signals of fine or local wear may be ignored, and the discontinuity of the fiber edges in the image cannot be accurately reflected, resulting in fiber breaks being misdetected as normal states and missed detections. Moreover, the existing break determination criteria of detection methods are usually based on fixed image feature thresholds and cannot dynamically adjust the thresholds according to environmental changes in an environment with large interference, failing to meet the actual needs of high-precision textile detection. Summary of the Invention

[0007] To overcome the above defects of the prior art, the present invention provides an intelligent detection method for broken textile fibers based on ultrasonic imaging analysis to solve the problems in the above background art.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] An intelligent detection method for broken textile fibers based on ultrasonic imaging analysis, comprising the following steps:

[0010] Step 1: Use an ultrasonic imaging device to obtain an operating state image of textile fibers, analyze the operating state image of textile fibers to obtain a fiber break index, set an initial break judgment threshold, and perform a first fiber break judgment based on the fiber break index and the initial break judgment threshold;

[0011] Step 2: If the first fiber break judgment is a fiber break, directly issue a break warning reminder; if the first fiber break judgment is a normal fiber, collect the current environmental data;

[0012] Step 3: Analyze the current environmental data to obtain an ultrasonic influence index;

[0013] Step 4: Dynamically adjust the initial break judgment threshold according to the ultrasonic influence index to obtain an actual break judgment threshold;

[0014] Step 5: Perform a second fiber break judgment through the dynamically adjusted actual break judgment threshold;

[0015] The step of collecting the current environmental data is as follows:

[0016] Set a vibration sensor at the ultrasonic imaging device to obtain vibration data in real time through the vibration sensor. The vibration data includes vibration acceleration, vibration frequency, and vibration displacement amplitude;

[0017] Set a noise measurement device at the ultrasonic imaging device to obtain noise data in real time through the noise measurement device. The noise data includes noise intensity and noise frequency;

[0018] A particulate matter mass sensor and an infrared gas sensor are set at the ultrasonic imaging device, and the particulate matter concentration and impurity volume fraction in the air are collected through the particulate matter mass sensor and the infrared gas sensor;

[0019] The steps for analyzing the current environmental data to obtain the ultrasonic influence index are as follows:

[0020] Obtain the vibration influence coefficient according to the vibration acceleration, vibration frequency, and vibration displacement amplitude;

[0021] Obtain the noise influence coefficient according to the noise intensity and noise frequency;

[0022] Obtain the air impurity influence coefficient according to the particulate matter concentration and impurity volume fraction in the air;

[0023] Evaluate and obtain the ultrasonic influence index according to the vibration influence coefficient, noise influence coefficient, and air impurity influence coefficient. The specific acquisition method is as follows: ; In the formula, represents the ultrasonic influence index, represents the vibration influence coefficient, represents the noise influence coefficient, represents the air impurity influence coefficient, , , represent the weight coefficients of the vibration influence coefficient, noise influence coefficient, and air impurity influence coefficient. Preferably, the steps for obtaining the fiber breakage index are as follows:

[0024] Use the connected region labeling algorithm to label the continuous regions of the fibers, and label all connected regions through a recursive method to obtain the number of connected regions;

[0025] Obtain the coordinates of the edge points of the connected regions and calculate the minimum gap width between the fiber connected regions;

[0026] Extract the morphological features of the fibers through image recognition technology. The morphological features include the fiber area and fiber perimeter, and obtain the shape factor according to the fiber area and fiber perimeter;

[0027] Evaluate and obtain the fiber breakage index according to the number of connected regions, gap width, and shape factor. The specific acquisition method is as follows: ; In the formula, represents the fiber breakage index, represents the number of connected regions, represents the minimum gap width, represents the shape factor. Preferably, the steps for the first fiber breakage judgment according to the fiber breakage index and the initial breakage judgment threshold are as follows:

[0028] Compare the fiber breakage index with the initial breakage judgment threshold. If the fiber breakage index is greater than or equal to the initial breakage judgment threshold, it is determined that the current textile fiber has broken, and the breakage position is located, and a breakage warning is issued; if the fiber breakage index is less than the initial breakage judgment threshold, it is preliminarily determined that the current state of the textile fiber is normal.

[0029] Preferably, the step of obtaining the air impurity influence coefficient according to the particulate matter concentration and impurity volume fraction in the air is as follows:

[0030] Divide the space from the ultrasonic imaging device to the detected textile fiber into n sub-spaces on average;

[0031] Calculate the air impurity degree of each sub-space according to the particulate matter concentration and impurity volume fraction in the air. The specific acquisition method is as follows: ; In the formula, represents the air impurity degree, represents the particulate matter concentration, represents the impurity volume fraction; perform clustering processing on the air impurity degree of each sub-space, and obtain the air impurity influence coefficient according to the clustering result.

[0032] Preferably, the step of performing clustering processing on the air impurity degree of each sub-space is as follows:

[0033] Step 3.1: Use the air impurity degree as the clustering feature, and use the air impurity degrees of all sub-spaces as the data set. Each air impurity degree in the data set is a data point;

[0034] Step 3.2: Use the distance statistics method to determine the optimal number of clusters K of the data set;

[0035] Step 3.3: Randomly select K data points in the data set as the initial cluster centers. For each data point, calculate its Euclidean distance to each initial cluster center. For each data point, traverse the K initial cluster centers and assign it to the cluster corresponding to the initial cluster center with the closest Euclidean distance;

[0036] Step 3.4: After traversing all data points, obtain the initial clusters. For each initial cluster, calculate the mean value of the data points in it to obtain a new cluster center;

[0037] Step 3.5: Repeat Step 3.3 and Step 3.4 until the cluster centers no longer change, and obtain the final clusters and final cluster centers.

[0038] Preferably, the step of obtaining the air impurity influence coefficient according to the clustering result is as follows:

[0039] Calculate the weight of each final clustering cluster by calculating the ratio of the number of data points in each final clustering cluster to the total number in the dataset;

[0040] Obtain it by performing a weighted sum of the weight of each final clustering cluster and the final clustering center.

[0041] Preferably, the step of dynamically adjusting the initial wire break judgment threshold according to the ultrasonic influence index to obtain the actual wire break judgment threshold is as follows:

[0042] Set an influence threshold, compare the ultrasonic influence index with the influence threshold. If the ultrasonic influence index is less than the influence threshold, it is determined that the current environment has little influence on the ultrasonic wave, and it is determined that the current state of the textile fiber is normal, and continue to detect the fiber at the next location;

[0043] If the ultrasonic influence index is greater than or equal to the influence threshold, it is determined that the current environment has a large influence on the ultrasonic wave, then calculate the ratio of the influence threshold to the ultrasonic influence index to obtain an adjustment factor;

[0044] Calculate the product of the adjustment factor and the initial wire break judgment threshold to obtain the actual wire break judgment threshold.

[0045] Preferably, the step of performing the second fiber wire break judgment using the actually adjusted wire break judgment threshold is as follows:

[0046] Compare the fiber wire break index with the actual wire break judgment threshold. If the fiber wire break index is less than the actual wire break judgment threshold, it is determined that the current state of the textile fiber is normal, no warning is issued, and continue to detect the fiber at the next location; if the fiber wire break index is greater than or equal to the actual wire break judgment threshold, it is determined that the current textile fiber has broken, locate the break position, and issue a wire break warning reminder.

[0047] The technical effects and advantages of the present invention:

[0048] Use an ultrasonic imaging device to obtain the running state image of the textile fiber, analyze the running state image of the textile fiber to obtain the fiber wire break index, perform the first fiber wire break judgment according to the fiber wire break index. If the first fiber wire break judgment is that the fiber is broken, directly issue a wire break warning reminder. If the first fiber wire break judgment is that the fiber is normal, collect the current environmental data, analyze to obtain the ultrasonic influence index, obtain the actual wire break judgment threshold according to the ultrasonic influence index, and perform the second fiber wire break judgment, effectively reducing the situation where the fracture signal is ignored due to the smallness of the fiber fracture. Description of the Drawings

[0049] Figure 1 It is a flowchart of the intelligent detection method for textile fiber wire break based on ultrasonic imaging analysis provided by the embodiment of the present application. Specific Embodiments

[0050] Next, in combination with the accompanying drawings in the present invention, the technical solutions in the present invention will be clearly and completely described. In addition, the forms of each structure described in the following embodiments are merely examples, and the intelligent detection method for broken textile fibers based on ultrasonic imaging analysis involved in the present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0051] The present invention provides an intelligent detection method for broken textile fibers based on ultrasonic imaging analysis, including the following steps:

[0052] Step 1: Use an ultrasonic imaging device to obtain an operating state image of the textile fiber, analyze the operating state image of the textile fiber to obtain a fiber breakage index, set an initial breakage judgment threshold, and perform a first fiber breakage judgment based on the fiber breakage index and the initial breakage judgment threshold;

[0053] In this embodiment, it should be specifically noted that the step of using an ultrasonic imaging device to obtain an operating state image of the textile fiber is as follows:

[0054] Obtain an image of the fiber through an ultrasonic imaging device, use the histogram equalization method to enhance the image contrast and improve the distinguishability between the fiber and the background. Histogram equalization is an image processing technique used to enhance image contrast. Its principle is to redistribute the gray values of the image to make the original gray histogram more uniform, thereby expanding the dynamic range of the gray values;

[0055] Use a filter to remove noise and improve the clarity of the fiber edge. A filter is an image processing tool used to reduce noise or interference in an image while retaining or enhancing important image features;

[0056] Use an edge detection algorithm to extract the boundary of the fiber and separate the fiber image. The edge detection algorithm is an image processing method used to identify regions in an image where the brightness changes sharply, thereby extracting the boundary of an object. Through edge detection, the boundary of the fiber can be accurately separated, highlighting its shape features, and providing a basis for further analysis (such as connectivity detection or gap calculation).

[0057] In this embodiment, it should be specifically noted that the step of obtaining the fiber breakage index is as follows:

[0058] Define the connectivity between a pixel and its neighboring pixels, such as 4-connectivity or 8-connectivity. Use the connected-component labeling algorithm to label the continuous regions of the fibers. Label all connected regions through a recursive method to obtain the number of connected components. The number of connected components represents the number of regions into which the fibers are segmented. The connected-component labeling algorithm is an image processing method used to identify and label the connected regions formed by adjacent pixels in an image. It assigns the same label to the pixels belonging to the same connected region by checking the connectivity between a pixel and its neighborhood, such as 4-connectivity or 8-connectivity. This algorithm can be used to extract the continuous regions of the fibers and help determine the integrity and breakage status of the fibers;

[0059] 4-connectivity and 8-connectivity are connectivity concepts in image processing used to define whether pixels are adjacent. 4-connectivity means that a pixel is adjacent to its four neighboring pixels above, below, left, and right, and is applicable to considering connectivity only in the horizontal and vertical directions. 8-connectivity extends this concept and means that a pixel is adjacent to its four neighboring pixels above, below, left, and right as well as the four diagonal pixels, covering connectivity in more directions.

[0060] Obtain the coordinates of the edge points of the connected regions and calculate the minimum gap width between the connected regions of the fibers. The specific obtaining method is as follows: ; where is expressed as the minimum gap width, and are expressed as the coordinates of the edge points of two connected regions;

[0061] Extract the morphological features of the fibers through image recognition technology. The morphological features include the fiber area, that is, the number of pixel points within the region, and the fiber perimeter, that is, the boundary length. Obtain the shape factor based on the fiber area and the fiber perimeter. The specific obtaining method is as follows: ; where is expressed as the shape factor, is expressed as the fiber perimeter, is expressed as the fiber area;

[0062] Evaluate and obtain the fiber breakage index based on the number of connected components, the gap width, and the shape factor. The specific obtaining method is as follows: ; where is expressed as the fiber breakage index, is expressed as the number of connected components. The number of connected components represents how many discontinuous regions the fibers are segmented into and is an important indicator for evaluating fiber breakage. Broken fibers are usually segmented into multiple isolated regions. Therefore, the larger the number of connected components, the higher the likelihood and severity of fiber breakage. This direct correlation can reflect the overall integrity of the fibers, Denoted as the minimum gap width, the minimum gap width is the core parameter for measuring the degree of fiber continuity interruption, representing the shortest distance between fiber-connected regions. It directly reflects the spatial characteristics of fracture: if the fiber breaks, obvious gaps will appear between the connected regions, and the larger the gap width, the more severe the fracture and the less stable the overall structure of the fiber. The larger the gap, the higher the probability and severity of fracture. This relationship ensures that the broken wire index can sensitively capture the degree of structural damage. Denoted as the shape factor, the shape factor is an index for measuring the geometric regularity and integrity of the fiber-connected region, reflecting whether the fiber maintains a coherent and regular morphology. For unbroken fibers, the shape factor is usually high, close to the value of a regular shape. In the case of fracture or wear, the edges of the fiber become irregular and the shape factor will decrease significantly. As an overall feature, the shape factor makes up for the minor anomalies that may be overlooked by relying solely on gap width or connectivity, playing a role in stabilizing the judgment.

[0063] In this embodiment, it should be specifically noted that the first fiber breakage judgment step according to the fiber broken wire index and the initial broken wire judgment threshold is as follows:

[0064] Compare the fiber broken wire index with the initial broken wire judgment threshold. If the fiber broken wire index is greater than or equal to the initial broken wire judgment threshold, it is determined that the current textile fiber has broken, and the break position is located, and a broken wire warning reminder is issued; if the fiber broken wire index is less than the initial broken wire judgment threshold, it is preliminarily determined that the current state of the textile fiber is normal.

[0065] Step 2: If the first fiber breakage judgment is that the fiber has broken, a broken wire warning reminder is directly issued. If the first fiber breakage judgment is that the fiber is normal, the current environmental data is collected;

[0066] In this embodiment, it should be specifically noted that the step of collecting the current environmental data is as follows:

[0067] A vibration sensor is set at the ultrasonic imaging device, and vibration data is obtained in real time through the vibration sensor. The vibration data includes vibration acceleration, vibration frequency, and vibration displacement amplitude;

[0068] A noise measurement device is set at the ultrasonic imaging device, and noise data is obtained in real time through the noise measurement device. The noise data includes noise intensity and noise frequency;

[0069] A particulate matter mass sensor and an infrared gas sensor are set at the ultrasonic imaging device, and the particulate matter concentration and impurity volume fraction in the air are collected through the particulate matter mass sensor and the infrared gas sensor.

[0070] Step 3: Analyze the current environmental data to obtain the ultrasonic influence index;

[0071] In this embodiment, it should be specifically noted that the steps for analyzing the current environmental data to obtain the ultrasonic influence index are as follows:

[0072] Obtain the vibration influence coefficient based on the vibration acceleration, vibration frequency, and vibration displacement amplitude. The specific acquisition method is as follows: ; where represents the vibration influence coefficient, represents the vibration acceleration, represents the vibration frequency, represents the vibration displacement amplitude;

[0073] Obtain the noise influence coefficient based on the noise intensity and noise frequency. The specific acquisition method is as follows: ; where represents the noise influence coefficient, represents the noise intensity, represents the noise frequency;

[0074] Obtain the air impurity influence coefficient based on the particulate matter concentration and impurity volume fraction in the air;

[0075] Evaluate and obtain the ultrasonic influence index based on the vibration influence coefficient, noise influence coefficient, and air impurity influence coefficient. The specific acquisition method is as follows: ; where represents the ultrasonic influence index, represents the vibration influence coefficient. When the vibration intensity increases, it may cause displacement or instability of the ultrasonic transmitter or receiver, resulting in problems such as signal deviation, noise superposition, or imaging blurring, thereby reducing the accuracy of fiber breakage detection. This proportional relationship indicates that the vibration intensity directly affects the quality of ultrasonic signals and the reliability of the detection system. represents the noise influence coefficient. Environmental noise can originate from mechanical operation, electromagnetic devices, or other acoustic interferences. These noises may be superimposed on the ultrasonic echo signal in the receiver, forming indistinguishable pseudo-signals or noise ripples. This interference reduces image contrast, blurs fiber boundaries, and even masks fine fracture features. In addition, noise may also cause misjudgment, such as ignoring the actual fracture area. represents the air impurity influence coefficient. Impurities in the air have various adverse effects on the propagation path of ultrasonic waves, such as scattering, absorption, or reflection, which may lead to attenuation of signal energy, deviation of propagation direction, or loss of echo signals. In addition, the attachment of impurities on the sensor surface may further reduce the sensitivity of the sensor, increase the noise level of the detection system, interfere with the clarity and coherence of the image, and thus affect the accurate identification of fiber fracture characteristics. 、 、 The weighting coefficients represented as the vibration influence coefficient, the noise influence coefficient, and the air impurity influence coefficient, and , , , the specific values are determined by professionals according to the actual situation. For example, , , they can be 0.4, 0.4, 0.2. In this embodiment, it should be specifically noted that the steps to obtain the air impurity influence coefficient according to the particulate matter concentration and the impurity volume fraction in the air are as follows:

[0076] The space from the ultrasonic imaging device to the detected textile fibers is evenly divided into n sub-spaces;

[0077] Calculate the air impurity degree of each sub-space according to the particulate matter concentration and the impurity volume fraction in the air. The specific acquisition method is as follows: ; In the formula, represents the air impurity degree, represents the particulate matter concentration, represents the impurity volume fraction. When the impurity volume fraction is small, the integration result is approximately linear, reflecting the direct contribution of low-concentration impurities to the air impurity degree. When the impurity volume fraction is large, the integration result approaches logarithmic growth, controlling the over-amplification of high-concentration impurities to the result; perform clustering processing on the air impurity degree of each sub-space, and obtain the air impurity influence coefficient according to the clustering result.

[0078] In this embodiment, it should be specifically noted that the steps to perform clustering processing on the air impurity degree of each sub-space are as follows:

[0079] Step 3.1: Use the air impurity degree as the clustering feature, and use the air impurity degrees of all sub-spaces as the data set. Each air impurity degree in the data set is a data point;

[0080] Step 3.2: Use the distance statistics method to determine the optimal number of clusters K of the data set;

[0081] The distance statistics method is a statistical method for evaluating clustering performance. It determines the optimal number of clusters K by comparing the within-cluster dispersion of the actual data with that of the random reference data. Specifically, the distance statistics method calculates the within-cluster dispersion of the actual data (such as the average distance from the samples within the cluster to the cluster center) and the within-cluster dispersion of the randomly generated data for different values of K, and evaluates the significance of the clustering result through the logarithmic difference between the two. The optimal number of clusters K is usually the number of clusters that maximizes the logarithmic difference, or the point where the logarithmic difference increases significantly. This method reduces the randomness of the clustering result affected by the distribution characteristics of the data set by introducing a random data benchmark.

[0082] Step 3.3: Randomly select K data points in the dataset as the initial cluster centers. For each data point, calculate its Euclidean distance to each initial cluster center. The specific acquisition method is as follows: ; In the formula, represents the Euclidean distance from the data point to the cluster center, where represents the data point, represents the initial cluster center. For each data point, traverse the K initial cluster centers and assign it to the cluster corresponding to the initial cluster center with the closest Euclidean distance; The Euclidean distance is a commonly used distance metric method for calculating the straight-line distance between two points in Euclidean space, reflecting the shortest path length between two points, and is an intuitive and physically meaningful distance metric method. In cluster analysis, the Euclidean distance is used to measure the similarity between data points and cluster centers. The smaller the distance, the closer the data point is to the cluster center.

[0083] Step 3.4: After traversing all data points, obtain the initial clusters. For each initial cluster, calculate the mean value of the data points within it to obtain a new cluster center;

[0084] Step 3.5: Repeat Step 3.3 and Step 3.4 until the cluster centers no longer change, and obtain the final clusters and the final cluster centers.

[0085] In this embodiment, it should be specifically noted that the steps to obtain the air impurity influence coefficient according to the clustering results are as follows:

[0086] Calculate the ratio of the number of data points in each final cluster to the total number in the dataset to obtain the weight of each final cluster;

[0087] Perform a weighted sum of the weight of each final cluster and the final cluster center to obtain the air impurity influence coefficient. Its calculation formula is , where represents the air impurity influence coefficient, represents the weight of the jth final cluster, represents the jth final cluster center, and K is the optimal number of clusters in the dataset.

[0088] By clustering the degree of air impurities in each subspace, the local characteristics of the air impurity distribution can be captured. Since air impurities may be unevenly distributed in different subspaces, this method effectively incorporates these differences into the influence coefficient, making the calculation results more comprehensive and accurate. The weight of each clustering cluster is based on the ratio of the number of data points to the total number, which directly reflects the representativeness of the clustering cluster in the overall space. This way of weight assignment can give priority to the influence of areas with denser impurity data distribution, thus avoiding being dominated by a few discrete points or noise points in the results. The clustering center represents the typical characteristics of air impurities within each cluster. By weighted summing the clustering center and the weight, the influence degree of the overall air impurities can be effectively summarized. This method can reduce the influence of local abnormal data points (such as extreme high or low values), making the air impurity influence coefficient more stable and robust.

[0089] Step 4: Dynamically adjust the initial wire break judgment threshold according to the ultrasonic influence index to obtain the actual wire break judgment threshold.

[0090] In this embodiment, it should be specifically noted that the step of dynamically adjusting the initial wire break judgment threshold according to the ultrasonic influence index to obtain the actual wire break judgment threshold is as follows:

[0091] Set the influence threshold, and compare the ultrasonic influence index with the influence threshold. If the ultrasonic influence index is less than the influence threshold, it is determined that the current environment has little influence on the ultrasonic wave, and the current state of the textile fiber is normal, and the fiber detection at the next location continues.

[0092] If the ultrasonic influence index is greater than or equal to the influence threshold, it is determined that the current environment has a greater influence on the ultrasonic wave. Then, calculate the ratio of the influence threshold to the ultrasonic influence index to obtain the adjustment factor. The specific acquisition method is as follows: ; In the formula, represents the adjustment factor, represents the ultrasonic influence index, represents the influence threshold;

[0093] Multiply the adjustment factor by the initial wire break judgment threshold to obtain the actual wire break judgment threshold. The specific acquisition method is as follows: ; In the formula, represents the actual wire break judgment threshold, represents the adjustment factor, represents the initial wire break judgment threshold. Step 5: Conduct the second fiber wire break judgment through the dynamically adjusted actual wire break judgment threshold.

[0094] In this embodiment, it should be specifically noted that the step of conducting the second fiber wire break judgment through the dynamically adjusted actual wire break judgment threshold is as follows:

[0095] Compare the fiber breakage index with the actual breakage judgment threshold. If the fiber breakage index is less than the actual breakage judgment threshold, it is determined that the current textile fiber state is normal, no warning is issued, and the fiber detection at the next location continues; if the fiber breakage index is greater than or equal to the actual breakage judgment threshold, it is determined that the current textile fiber has broken, the breakage position is located, and a breakage warning is issued to remind the relevant staff to repair the broken textile fiber in time.

[0096] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

[0097] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or replacements, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claimed rights.

Claims

1. An intelligent method for detecting textile fiber breakage based on ultrasonic imaging analysis, characterized in that: The following steps are involved: Step 1: Use ultrasonic imaging equipment to obtain a running state image of the textile fiber, analyze the running state image of the textile fiber, obtain a fiber breakage index, set an initial breakage judgment threshold, and perform the first fiber breakage judgment based on the fiber breakage index and the initial breakage judgment threshold; Step 2: If the first fiber break is judged as a fiber break, a break warning reminder is directly issued; if the first fiber break is judged as a normal fiber, current environmental data is collected; Step 3: Analyze the current environmental data to obtain the ultrasonic impact index; Step 4: Dynamically adjust the initial disconnection judgment threshold according to the ultrasonic impact index to obtain the actual disconnection judgment threshold; Step 5: Perform a second fiber breakage judgment using the dynamically adjusted actual breakage judgment threshold; The steps of collecting current environment data are: A vibration sensor is provided at the ultrasonic imaging device to obtain vibration data in real time through the vibration sensor, the vibration data including vibration acceleration, vibration frequency and vibration displacement amplitude; A noise measuring device is arranged at the ultrasonic imaging device, and noise data is obtained in real time through the noise measuring device, wherein the noise data includes noise intensity and noise frequency; A particle mass sensor and an infrared gas sensor are arranged at the ultrasonic imaging device to collect the particle concentration and impurity volume fraction in the air through the particle mass sensor and the infrared gas sensor; The steps of analyzing the current environmental data to obtain the ultrasonic impact index are: The vibration influence coefficient is obtained according to the vibration acceleration, vibration frequency and vibration displacement amplitude; The noise influence coefficient is obtained according to the noise intensity and the noise frequency; The air impurity influence coefficient is obtained according to the particle concentration and impurity volume fraction in the air; The ultrasonic impact index is obtained by evaluating the vibration impact coefficient, noise impact coefficient and air impurity impact coefficient. The specific method of obtaining it is as follows: ; In the formula, Expressed as the ultrasound impact index, Expressed as the vibration influence coefficient, Expressed as the noise influence coefficient, Expressed as the air impurity influence coefficient, , , It is expressed as the weight coefficient of vibration influence coefficient, noise influence coefficient and air impurity influence coefficient.

2. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 1 is characterized in that: The fiber breakage index acquisition steps are: Use the connected region marking algorithm to mark the continuous region of the fiber, mark all connected regions through a recursive method, and obtain the number of connected regions; Obtain the edge point coordinates of the connected area and calculate the minimum gap width between fiber connected areas; The morphological characteristics of the fibers are extracted by image recognition technology. The morphological characteristics include the fiber area and the fiber perimeter. The shape factor is obtained according to the fiber area and the fiber perimeter. The fiber breakage index is obtained by evaluating the number of connected areas, gap width and shape factor. The specific method is as follows: ; In the formula, Expressed as the fiber breakage index, It is expressed as the number of connected regions, Expressed as the minimum gap width, Expressed as a shape factor.

3. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 1 is characterized in that: The first fiber breakage judgment step according to the fiber breakage index and the initial breakage judgment threshold is as follows: The fiber breakage index is compared with the initial breakage judgment threshold. If the fiber breakage index is greater than or equal to the initial breakage judgment threshold, it is determined that the current textile fiber is broken, the breakage position is located, and a breakage warning reminder is issued; If the fiber breakage index is less than the initial breakage judgment threshold, it is preliminarily determined that the current textile fiber state is normal.

4. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 1 is characterized in that: The step of obtaining the air impurity influence coefficient according to the particle concentration and impurity volume fraction in the air is: Divide the space from the ultrasonic imaging device to the detection textile fiber into n subspaces on average; The air impurity level of each subspace is calculated based on the particle concentration and impurity volume fraction in the air. The specific acquisition method is as follows: ; In the formula, Expressed as the degree of air impurities, Expressed as the particle concentration, It is expressed as the impurity volume fraction; the air impurity degree of each subspace is clustered, and the air impurity influence coefficient is obtained according to the clustering result.

5. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 4 is characterized in that: The steps of clustering the air impurity level of each subspace are as follows: Step 3.1: Take the air impurity level as the clustering feature, take the air impurity level of all subspaces as the data set, and each air impurity level in the data set as a data point; Step 3.2: Use distance statistics to determine the optimal number of clusters K for the data set; Step 3.3: Randomly select K data points in the data set as initial cluster centers. For each data point, calculate its Euclidean distance to each initial cluster center. For each data point, traverse the K initial cluster centers and assign it to the cluster corresponding to the initial cluster center with the closest Euclidean distance. Step 3.4: After traversing all data points, the initial clusters are obtained. For each initial cluster, the mean of the data points in it is calculated to obtain a new cluster center. Step 3.5: Repeat steps 3.3 and 3.4 until the cluster center no longer changes, and obtain the final cluster and the final cluster center.

6. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 4 is characterized in that: The step of obtaining the air impurity influence coefficient according to the clustering result is: The weight of each final cluster is calculated by calculating the ratio of the number of data points in each final cluster to the total number of data points in the data set; The weight of each final cluster is obtained by weighted summing the weight of the final cluster center.

7. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 1 is characterized in that: The steps of dynamically adjusting the initial disconnection judgment threshold according to the ultrasonic wave influence index to obtain the actual disconnection judgment threshold are as follows: Set the affected threshold, compare the ultrasonic affected index with the affected threshold, if the ultrasonic affected index is less than the affected threshold, it is determined that the current environment has little impact on the ultrasonic wave, the current textile fiber state is determined to be normal, and continue to the next fiber detection; If the ultrasonic wave influence index is greater than or equal to the influence threshold, it is determined that the current environment has a great influence on the ultrasonic wave, and the adjustment factor is obtained by calculating the ratio of the influence threshold to the ultrasonic wave influence index; The actual disconnection judgment threshold is obtained by multiplying the adjustment factor by the initial disconnection judgment threshold.

8. The intelligent detection method for textile fiber breakage based on ultrasonic imaging analysis according to claim 1 is characterized in that: The steps of performing the second fiber breakage judgment using the dynamically adjusted actual breakage judgment threshold are as follows: The fiber breakage index is compared with the actual breakage judgment threshold. If the fiber breakage index is less than the actual breakage judgment threshold, the current textile fiber status is judged to be normal, no warning is issued, and the next fiber detection is continued; if the fiber breakage index is greater than or equal to the actual breakage judgment threshold, the current textile fiber is judged to be broken, the breakage position is located, and a breakage warning reminder is issued.

Images