Image resolution monofilament detection system and method

Abstract

The invention discloses an image resolution monofilament detection system and method, and relates to the technical field of monofilament detection. According to the image resolution monofilament detection system and method, a high-resolution camera is used for collecting a marked area image caused by an identification accessory on a chemical fiber godet wheel in real time, an image processing module is used for carrying out noise removal and color space conversion on the image, and a Canny edge detection algorithm is used for extracting the edge contour of the image. And the color identification module analyzes the color change of the marked area through a contour tracking algorithm, calculates to obtain a broken wire index, and judges whether the marked area is a broken wire area according to a preset threshold value. The broken wire judgment module judges whether the wire guide wheel rotates or not according to the color recognition result so as to judge whether the wire is broken or not, and early warning information is sent out when the wire is broken. The system can accurately monitor the yarn breakage problem in the chemical fiber production process, and has efficient and real-time yarn breakage detection capability.

Description

Technical Field

[0001] The present invention relates to the technical field of monofilament detection, and particularly to an image resolution monofilament detection system and method. Background Art

[0002] With the rapid development of industrial automation and intelligent manufacturing, the textile industry has increasingly higher requirements for production quality. Especially in the process of chemical fiber production, the quality and stability of monofilaments directly affect the performance and appearance of the final products. Therefore, the monitoring of chemical fiber godet wheels and broken filament detection have become important links in the quality control of production lines. Traditional manual detection methods are not only time-consuming and laborious, but also easily affected by human factors, and the accuracy and real-time nature of the detection results cannot meet the requirements of modern production lines for efficient and high-precision monitoring. Therefore, an automated and intelligent image resolution monofilament detection system has emerged, which realizes the real-time monitoring of the operating state of chemical fiber godet wheels through efficient and accurate image processing technology, so as to timely detect broken filament problems, prevent production accidents, and ensure product quality.

[0003] At present, the broken filament detection of chemical fiber godet wheels mainly relies on traditional sensor technologies, mechanical monitoring devices or manual inspections. Traditional sensor technologies and mechanical devices may not be able to accurately capture the minute changes on the godet wheels, resulting in low sensitivity of broken filament detection and unable to detect all broken filament situations in a timely manner. Mechanical devices and manual detection methods usually cannot provide real-time feedback, making it difficult to detect broken filament problems in a timely manner during the production process, thus affecting production efficiency and product quality. Sensors and mechanical monitoring devices usually require regular maintenance and replacement, increasing the operating costs of the production line. False detections or missed detections are likely to occur in complex production environments, affecting the accuracy of the detection results.

[0004] Therefore, there is an urgent need for a more accurate, real-time and automated detection means for the existing broken filament detection technologies. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides an image resolution monofilament detection system and method. By using a high-resolution camera to collect images in real time and performing precise image processing and color change analysis, it can automatically detect and judge the breakage of filaments, reducing the risk of manual intervention and missed detections, and ensuring the efficiency, precision and stability of the production process. Such a system and method can timely identify the broken filament area and issue a warning, improving the automation level of the production line and enhancing the overall production efficiency and product quality. The problems in the above background art are solved.

[0006] To achieve the above objectives, the present invention is realized through the following technical solutions: An image resolution single-filament detection system, including the following modules: an image acquisition module, an image processing module, a color recognition module, and a broken-filament determination module; the image acquisition module is used to collect, in real time through a high-resolution camera, the image of the marked area caused by the recognition accessory on the chemical fiber guide wheel; the image processing module is used to remove noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, extract the edge of the marked area in the image through the Canny edge detection algorithm, and obtain the edge contour image of the marked area; the color recognition module is used to analyze the change in the marked color in the edge contour image of the marked area through the contour tracking algorithm, identify the change amount of the marked area image color, obtain the broken-filament index, and identify whether the marked area is a broken-filament area according to the preset broken-filament index threshold; the broken-filament determination module judges whether the guide wheel rotates according to the recognition result of the marked area by the color recognition module, and judges whether the filament is broken according to the rotation state, and issues a warning message when the filament is broken; the broken-filament index is used to quantify the degree of change in the marked color in the marked area image.

[0007] Further, the specific process of collecting the image of the marked area caused by the recognition accessory on the chemical fiber guide wheel is as follows: The chemical fiber includes filament and spandex; the camera transmits the collected marked area image to the image processing module, and according to the setting of the optical sensor of the camera, ensures that the image can be accurately captured under different working conditions, and converts the real-time collected marked area image into a digital signal.

[0008] Further, the specific process of removing noise from the marked area image is as follows: Receive the marked area image transmitted by the image acquisition module; Smooth the marked area image through the Gaussian blur algorithm to remove the random noise in the image; Remove the salt-and-pepper noise in the image through the median filter algorithm and enhance the contrast, and retain the details of the marked area image.

[0009] Further, the specific process of extracting the edge of the marked area in the image through the Canny edge detection algorithm and obtaining the edge contour image of the marked area is as follows: Convert the marked area image after noise removal into a grayscale image, perform Gaussian filtering on the grayscale image to remove noise and smooth the image; Calculate the gradient value of the grayscale image, calculate the gradients in the horizontal and vertical directions respectively through the Sobel operator, and compare each pixel in the grayscale image with its adjacent pixels, and apply the non-maximum suppression algorithm to retain the edge pixels with the set gradient amplitude; Through the double-threshold determination method, set the low threshold and the high threshold, screen out the strong edges and the weak edges, and connect the adjacent weak edges to output the edge contour image of the marked area.

[0010] Further, the specific process of analyzing the marker color change in the edge contour image of the marked area through the contour tracing algorithm and identifying the color change amount of the marked area image is as follows: Extract the edge contour image of the marked area and convert it into a format for processing by the contour tracing algorithm; Analyze the edge features of the marked area through the contour tracing algorithm, identify the color change at the edge, and calculate the amplitude of the color change within the marked area.

[0011] Further, the specific process of obtaining the broken wire index is as follows: Extract the edge features of the marked area through the contour tracing algorithm, analyze according to the color change amount, calculate the amplitude of the color change within the marked area by weighted average, normalize the obtained amplitude of the color change, map it to a set range, and obtain the broken wire index.

[0012] Further, the specific process of identifying whether the marked area is a broken wire area according to the preset broken wire index threshold is as follows: When the broken wire index is greater than the preset broken wire index threshold, the marked area is identified as a broken wire area; when it is less than the preset broken wire index threshold, the marked area is identified as an unbroken wire area.

[0013] Further, the specific process of judging whether the wire guiding wheel rotates and judging whether the wire is broken according to the rotation state is as follows: Identify the marked area according to the broken wire index. When the marked area is identified as a broken wire area, it means that the color of the marked area image is a mixture of red and blue, the wire guiding wheel does not rotate, and the wire is broken; when the marked area is identified as an unbroken wire area, it means that the color of the marked area image is purple, the wire guiding wheel rotates, and the wire is not broken.

[0014] An image resolution single wire detection method includes the following steps: S1. Real-time collect the marked area image caused by the identification attachment on the chemical fiber wire guiding wheel through a high-resolution camera; S2. Remove the noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, and extract the edge of the marked area in the image by applying the Canny edge detection algorithm to obtain the edge contour image of the marked area; S3. Analyze the marker color change in the edge contour image of the marked area through the contour tracing algorithm, identify the color change amount of the marked area image, obtain the broken wire index, and identify whether the marked area is a broken wire area according to the preset broken wire index threshold; S4. According to the recognition result of the marked area, judge whether the wire guiding wheel rotates and judge whether the wire is broken according to the rotation state, and send a warning message when the wire is broken.

[0015] The present invention has the following beneficial effects: (1) The image resolution single - filament detection system collects, in real - time, the image of the marked area on the chemical fiber guide wheel caused by the identification accessory through a high - resolution camera, and converts it into a format suitable for further processing. Through image pre - processing steps such as noise removal, color space conversion, and edge detection, the system can effectively improve the image quality, extract the precise edge contour of the marked area, and provide high - quality data support for subsequent color recognition and broken - filament judgment.

[0016] (2) The image resolution single - filament detection method analyzes the color change amount in the marked area image through a contour tracking algorithm, quantifies it as a broken - filament index, and combines a preset threshold to determine whether the marked area is a broken - filament area. On this basis, the system further judges whether the filament is broken according to the rotation state of the guide wheel, and issues a warning message in a timely manner when a filament break is detected, ensuring that abnormal situations in the production process are quickly responded to and processed, improving the intelligence and automation level of the production line, and effectively reducing the cost and error rate of manual monitoring.

[0017] Of course, it is not necessary for any product implementing the present invention to achieve all of the above - mentioned advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of an image resolution single - filament detection system of the present invention.

[0019] Figure 2 It is a flowchart of an image resolution single - filament detection method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] The embodiments of the present application solve the problems of reduced production efficiency and unstable quality caused by filament breakage in the production process of chemical fiber guide wheels through an image resolution single - filament detection system and method.

[0021] The general idea for the problems in the embodiments of the present application is as follows: Collect, in real - time, the image of the marked area on the chemical fiber guide wheel caused by the identification accessory through a high - resolution camera.

[0022] Perform noise removal on the marked area image, convert the marked area image from the RGB color space to the HSV color space, and extract the edge of the marked area in the image by applying the Canny edge detection algorithm to obtain the edge contour image of the marked area.

[0023] Analyze the marked color change in the edge contour image of the marked area through a contour tracking algorithm, identify the color change amount of the marked area image, obtain the broken - filament index, and identify whether the marked area is a broken - filament area according to the preset broken - filament index threshold.

[0024] According to the recognition result of the marked area, it is judged whether the guide wire wheel rotates, and whether the wire is broken is judged according to the rotation state. When the wire is broken, a warning message is sent.

[0025] Please refer to Figure 1 , an embodiment of the present invention provides a technical solution: an image resolution single wire detection system, including the following modules: an image acquisition module, an image processing module, a color recognition module, and a broken wire determination module; the image acquisition module is used to collect the marked area image caused by the recognition attachment on the chemical fiber guide wire wheel in real time through a high-resolution camera; the image processing module is used to remove noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, extract the marked area edge in the image through the Canny edge detection algorithm, and obtain the marked area edge contour image; the color recognition module is used to analyze the marked color change in the marked area edge contour image through the contour tracking algorithm, identify the marked area image color change amount, obtain the broken wire index, and identify whether the marked area is a broken wire area according to the preset broken wire index threshold; the broken wire determination module judges whether the guide wire wheel rotates according to the recognition result of the marked area by the color recognition module, and judges whether the wire is broken according to the rotation state. When the wire is broken, a warning message is sent; the broken wire index is used to quantify the degree of marked color change in the marked area image.

[0026] In this implementation scheme, the image acquisition module uses a high-resolution camera to collect images of the marked area on the chemical fiber guide wheel in real time. The camera needs to be able to accurately capture the subtle changes on the guide wheel to ensure that the image quality is high enough to support subsequent processing. High-resolution camera: It refers to a camera device that can collect images with high details and low noise, usually having a high number of pixels to accurately capture the details in the image. Marked area image: It refers to the image area generated on the guide wheel by identifying accessories (reflective sheets, markers). These marked areas are used to identify the running state of the wire. Image processing module: The task of this module is to preprocess the collected marked area images, remove noise, and convert the color space of the images to facilitate subsequent edge detection and analysis. Noise removal: This step removes random noise in the image through an algorithm to ensure image quality and improve the accuracy of subsequent algorithms. RGB to HSV color space conversion: RGB is a common color representation method, while the HSV color space can process color information more effectively, especially when dealing with color changes in images. HSV (hue H, saturation S, value V) can separate color from brightness, making the detection of color changes more stable and accurate. Canny edge detection algorithm: The Canny algorithm is a common image edge detection technique. It finds the edge positions by calculating the gradients of the image and accurately extracts the edges of the image through multiple filtrations and threshold judgments. RGB color space: It is a standard method of representing the colors of an image as a combination of red (R), green (G), and blue (B). HSV color space: It is a space that represents colors through hue (H), saturation (S), and value (V). Compared with the RGB space, HSV is more suitable for color segmentation and color analysis. Canny edge detection: By calculating the gradient values of the pixels in the image, the edges of the image are found. This method is widely used in edge recognition to detect the details of the image contour. Color recognition module: This module analyzes the color changes in the marked area, quantifies them into a broken wire index, and is used to determine whether the marked area is a broken wire area. The module uses a contour tracking algorithm to extract edge features and analyze the amplitude of color changes. Contour tracking algorithm: This algorithm tracks and records the positions and features of the edge contours according to the edge information in the image. It is commonly used to detect line or area changes in images. Broken wire index: It is a quantified index used to describe the amplitude of color changes in the marked area, usually represented by the ratio of color changes. The larger the value, the more significant the color change and the greater the possibility of a broken wire. Broken wire determination module: This module further determines whether the guide wheel is rotating based on the output result of the color recognition module, and judges whether the wire is broken according to the rotation state. If the wire is broken, the system will trigger a warning message. Guide wheel rotation state: It refers to whether the guide wheel is in a rotating state during operation. Whether the wire is broken is judged based on the rotation state.Warning information: It refers to the message that the system notifies the operator by means of alarm, signal lamp or other ways when an abnormal state (such as wire breakage) is detected.

[0027] Specifically, the specific process of collecting the image of the marked area caused by the identification attachment on the chemical fiber godet wheel is as follows: Chemical fibers include filament yarns and spandex filaments; The camera transmits the collected image of the marked area to the image processing module. According to the optical sensor settings of the camera, it is ensured that the image can be accurately captured under different working conditions, and the real-time collected image of the marked area is converted into a digital signal.

[0028] In this implementation, chemical fiber: Chemical fiber is a fiber made from natural or synthetic polymers through chemical methods and is widely used in the manufacture of various textiles. Filament: It refers to a type of continuous fiber in the textile industry, usually having a relatively long length, capable of being stretched within a certain range, and used in the production of fabrics and other textiles. Spandex filament: Spandex is an elastic fiber with strong elasticity and extensibility, usually used in making highly elastic fabrics such as tight-fitting clothes and sportswear. During the processing of these chemical fiber products, different production abnormalities may occur due to fiber breakage, material wear, or other factors. Therefore, it is necessary to monitor and detect these chemical fiber materials in real time to ensure production quality. Role of the camera: The camera, as the core device for image acquisition, captures the image of the marked area on the chemical fiber godet wheel through its optical sensor. In this system, the marked area is the image area generated on the godet wheel by accessories (such as reflectors, identifiers, etc.), and these accessories are used to help identify the operating state of the godet wheel, detect the quality of the filament, etc. Optical sensor settings: The optical sensor is a key component in the camera that captures light and converts it into digital signals. The accuracy and sensitivity of the sensor directly affect the image quality. According to the settings and configurations of the camera (such as focal length, shutter speed, exposure time, etc.), the camera can adapt to different working conditions and ensure accurate capture of the marked area image under various lighting conditions. For example, in strong light or low light environments, the sensor of the camera will automatically adjust its exposure amount to ensure that the image is not overexposed or too dim, thus ensuring the clarity and accuracy of the image. Transmission of the marked area image to the image processing module: The camera transmits the captured marked area image to the image processing module through an interface (such as USB, Ethernet, etc.). This transmission process is usually real-time, thus ensuring the timeliness of the detection process. During the transmission process, the image is usually encoded into digital signals for subsequent processing and analysis. Conversion of real-time image into digital signals: The optical sensor of the camera converts the captured image into digital signals by sensing light. The process of converting an image into digital signals involves sampling the intensity and color of light and representing each pixel in the image with discrete digital values. These digital signals can contain rich image information, such as features like color, brightness, contrast, edges, etc., for subsequent image processing algorithms to use.

[0029] Specifically, the specific process of removing noise from the marked area image is as follows: Receive the marked area image transmitted by the image acquisition module; Smooth the marked area image through the Gaussian blur algorithm to remove random noise in the image; Remove salt-and-pepper noise in the image through the median filter algorithm and enhance the contrast, retaining the details of the marked area image.

[0030] In this implementation, noise: Noise generally refers to random interference points or unnecessary elements in an image. These interferences may be caused by various reasons, such as environmental light changes, sensor errors, or other external factors. Noise needs to be effectively removed during the image processing process to avoid affecting subsequent steps such as edge detection and color change analysis. Gaussian blur: Gaussian blur is a commonly used image smoothing algorithm that blurs image details through a convolution operation on the image, thereby removing some small random noises. This algorithm is calculated based on the Gaussian function (normal distribution), and its smoothing effect has a good mathematical basis. The Gaussian filter weights and averages the pixel values around each pixel. The pixels closer to the center have higher weights, and the pixels farther from the center have lower weights. Applying Gaussian blur: When applying Gaussian blur to the marked area image, each pixel value in the image is weighted and averaged with the pixels in its neighborhood, resulting in a smoothing effect. In this way, random noises in the image (such as salt-and-pepper noise and noise caused by camera sensor errors) will be smoothed out, thus removing most of the random noises. This process helps to make the image smoother and provides more stable input data for subsequent processing algorithms. Median filtering algorithm (removing salt-and-pepper noise and enhancing contrast): Salt-and-pepper noise: Salt-and-pepper noise is a form of image noise, manifested as extremely bright or extremely dark pixel points in the image, similar to the dispersion of "salt grains" and "pepper grains" in the image. This kind of noise is usually caused by factors such as sensor errors and transmission errors, and has a greater impact on the image quality. Median filtering: Median filtering is a non-linear filtering technique mainly used to remove salt-and-pepper noise in images. Its principle is to select a fixed-size window around each pixel in the image, arrange all the pixel values in the window in ascending or descending order, and select the middle value as the new value of the pixel at the center of the window. This method is particularly effective for salt-and-pepper noise because the noise usually causes extreme pixel values, and median filtering can effectively eliminate these extreme values by selecting the middle value. Enhancing contrast: While removing salt-and-pepper noise, median filtering can also enhance the contrast of the image to a certain extent because it can smooth out the extreme values in the image while retaining the edge information in the image. This is very helpful for retaining image details and the accuracy of subsequent processing. Retaining details: During the noise removal process, especially when applying median filtering, although the noise is removed, important details in the image, such as the contours and textures of the marked area, are retained. This is because median filtering is different from Gaussian blur and can effectively avoid blurring the important structural features in the image. For a single-filament detection system, the details of the marked area usually include key information such as the contours and color changes of the markers on the wire guiding wheel, and these information are crucial for broken wire detection. Therefore, during the noise removal process, the detail information in the image needs to be retained so that subsequent algorithms such as color recognition and edge detection can accurately identify the changes in the marked area.

[0031] Specifically, the edge of the marked area in the image is extracted by the Canny edge detection algorithm, and the specific process of obtaining the edge contour image of the marked area is as follows: the marked area image after noise removal is converted into a grayscale image, and the grayscale image is Gaussian filtered to remove noise and smooth the image; the gradient value of the grayscale image is calculated, and the gradients in the horizontal and vertical directions are calculated respectively by the Sobel operator, and each pixel in the grayscale image is compared with its adjacent pixels, and the non-maximum suppression algorithm is applied to retain the edge pixels with the set gradient amplitude; through the double threshold judgment method, the low threshold and the high threshold are set to screen out strong edges and weak edges, and the adjacent weak edges are connected to output the edge contour image of the marked area.

[0032] In this embodiment, the marked area image after noise removal is converted into a grayscale image: Grayscale image: A grayscale image is an image after color information is removed, and each pixel contains only brightness information. The color of an image is usually represented by the numerical values of the three color channels RGB (red, green, and blue), while a grayscale image is represented by the brightness value of a single channel. Conversion to a grayscale image is a common step in image preprocessing to reduce computational complexity and focus on the detection of brightness changes (i.e., edges). Gaussian filtering is performed on the grayscale image to remove noise and smooth the image: Gaussian filtering: Gaussian filtering is a low-pass filter used to smooth images and remove high-frequency noise. It smoothes the image and eliminates noise by taking a weighted average of the neighborhood pixels around each pixel. Gaussian kernel function: The weights of the Gaussian filter are calculated based on the Gaussian function, and the formula is as follows: ;in: is the value of the Gaussian kernel function; is the standard deviation, which controls the degree of filtering, The larger the value, the smoother the image. are the coordinates of the pixel. The purpose of Gaussian filtering is to remove small-scale noise in the image, smooth the details in the image, and prepare a clearer image for subsequent edge detection. Gradient calculation: The edge is the area in the image where the brightness changes most dramatically, and the gradient describes the degree and direction of the image brightness change. In order to calculate the gradient, the Sobel operator is usually used, which is a common discrete operator used to calculate the gradient of the image in the horizontal and vertical directions. Sobel operator: The Sobel operator calculates the local gradient of the image through convolution operations. The commonly used Sobel operators are as follows: Sobel operator in the horizontal direction: ; Sobel operator in the vertical direction: ; By applying these two operators respectively, the gradient values of each pixel in the image in the horizontal and vertical directions are calculated: the horizontal gradient Calculation formula: Vertical gradient Calculation formula ;in: is the brightness value of a certain pixel in the image. and are the weights of the Sobel operator. Apply the non-maximum suppression algorithm to retain the edge pixels with the largest gradient magnitude: Non-maximum suppression: After calculating the gradient magnitude, the algorithm generates a large number of possible edge points. The purpose of the non-maximum suppression algorithm is to retain the points with the largest gradient magnitude in the image and suppress other insignificant edge points. The specific steps are: Calculate the gradient magnitude (Magnitudc) of each pixel point: ; Calculate the gradient direction of each pixel point ( ): , and check whether the pixel value is a local maximum along the gradient direction. If it is not a local maximum, suppress the point and set it to 0. Non-maximum suppression ensures the refinement of the edge contour, making the detected edges clearer and more accurate. Through the double-threshold decision method, strong edges and weak edges are screened out, and adjacent weak edges are connected: Double-threshold decision method: The double-threshold decision method is used to determine which edges are strong edges and which are weak edges and classify them: Strong edges: The gradient magnitude is greater than the high threshold, and these are considered reliable edges. Weak edges: The gradient magnitude is between the low threshold and the high threshold, and these are considered potential edges. Non-edges: The gradient magnitude is less than the low threshold, and these are considered not to be edges. Connecting weak edges: For weak edges, if it is adjacent to a strong edge, it is considered a real edge and retained; if it is not adjacent to a strong edge, it is removed. After screening out strong edges and weak edges through the double-threshold decision method, a more accurate edge image is obtained.

[0033] Specifically, by analyzing the color change of the marker in the edge contour image of the marker area through the contour tracing algorithm, the specific process of identifying the color change amount of the marker area image is as follows: Extract the edge contour image of the marker area and convert it into a format processed by the contour tracing algorithm; Analyze the edge features of the marker area through the contour tracing algorithm, identify the color change at the edge, and calculate the amplitude of the color change within the marker area.

[0034] In this implementation scheme, the contour tracking algorithm is a method commonly used in computer vision to extract and analyze the edge features of images. This algorithm helps detect and analyze image changes by tracking the edge contours in the image and identifying and analyzing the color changes in the edge regions. In the single filament detection of image resolution, the contour tracking algorithm is used to identify the amount of color change in the marked area image and further obtain the broken filament index. The specific steps are as follows: Edge extraction: First, obtain the denoised marked area image from the image processing module and use an edge detection algorithm (such as the Canny edge detection) to extract the edge information of the marked area. Edge detection helps distinguish the contours between different regions in the image. Format conversion: Convert the extracted edge image into a format that the contour tracking algorithm can process, usually a binary image or an image processed with an appropriate threshold. This image only contains the edge information of the marked area, facilitating subsequent algorithms to track and analyze the edges. Contour tracking: Use the contour tracking algorithm (the findContours algorithm in OpenCV) to continuously track the edges of the marked area. The purpose of this step is to detect features such as the shape, position, and size of the edges and obtain the contour information of the edges. Edge feature recognition: The algorithm determines the continuity, change rules, etc. of the edges based on the edge features extracted from the contours. On this basis, the algorithm can analyze the color change trend within the marked area, especially the transition process of the marked color from red-blue mixture to purple.

[0035] Color change detection: The contour tracking algorithm identifies the color changes at the edges based on the direction and position of the edges. Within the marked area, due to the rotation or non-rotation of the wire guiding wheel, the color may change significantly (such as the red-blue mixture changing to purple). By analyzing the amount of color change, the amplitude of the color change can be obtained. Analyzing the marked color change: The algorithm quantifies the amplitude of the color change by comparing the colors in the edge regions at different time points. Generally, the larger the amplitude of the color change, the more significant the change within the marked area, which is related to whether the wire is broken.

[0036] Specifically, the specific process of obtaining the broken filament index is as follows: Extract the edge features of the marked area through the contour tracking algorithm, analyze according to the amount of color change, calculate the amplitude of the color change within the marked area through weighted average, normalize the obtained amplitude of the color change, map it to the set range, and obtain the broken filament index.

[0037] In this implementation scheme, when extracting the edge features of the marked area through the contour tracking algorithm and analyzing according to the amount of color change, the weighted average method is used to calculate the amplitude of the color change, considering the contribution of different edge regions to the overall color change. The amount of color change at each edge point will be assigned different weights according to its importance (such as the area, position, or depth of the edge point). Through the weighted average method, the interference of noise in the edge regions can be effectively avoided, and the overall color change situation can be accurately reflected. Weighted average formula: ; wherein: is the weighted average of color changes (i.e., the amplitude of color change), is the weight of the th edge point, usually defined according to the importance of this point in the marked area. is the th color change amount of the edge point, indicating the amplitude of color change of this point. is the total number of edge points. After obtaining the weighted average amplitude of color change, normalization is performed. The purpose of normalization is to adjust the amplitude of color change to a standard range ([0,1]) so that the amplitudes of color change under different conditions can be compared. Normalization formula: ; wherein: is the normalized amplitude of color change, with a value in the range of [0,1] is the amplitude of color change obtained by weighted average. and are respectively the minimum and maximum values of the amplitude of color change, used to determine the normalization range. Finally, the normalized amplitude of color change is mapped to a set range, as the broken wire index. This is for facilitating broken wire judgment and the generation of warning information. Mapping formula: ; wherein: is the mapped broken wire index, indicating the quantization value of the amplitude of color change. is the normalized amplitude of color change. and are the minimum and maximum values of the preset broken wire index, used to map to the final broken wire index range. Specifically, according to the preset broken wire index threshold, the specific process of identifying whether the marked area is a broken wire area is as follows: when the broken wire index is greater than the preset broken wire index threshold, the marked area is identified as a broken wire area; when it is less than the preset broken wire index threshold, the marked area is identified as an unbroken wire area.

[0038] In this implementation scheme, the process of identifying whether the marked area is a wire break area is achieved by judging the wire break index according to a preset wire break index threshold. The wire break index is the amplitude of the color change in the marked area calculated in the previous steps, indicating the degree of color change. When the wire break index of the marked area is greater than the preset wire break index threshold, it indicates that the amplitude of the color change in the marked area is large, usually meaning that the breakage of the wire causes an obvious color change, so this area is identified as the "wire break area". On the contrary, when the wire break index is less than the preset threshold, it means that the color change is small and the amplitude of the color change does not reach the standard of breakage, so the marked area is identified as the "non-wire break area". The core of this process is to accurately judge whether the wire is broken by comparing the wire break index with the preset threshold, and then decide whether the system needs to issue an alarm.

[0039] Specifically, the specific process of judging whether the guide wire wheel rotates and judging whether the wire is broken according to the rotation state is as follows: Identify the marked area according to the wire break index. When the marked area is identified as the wire break area, it means that the color of the marked area image is a mixture of red and blue, the guide wire wheel does not rotate, and the wire is broken; when the marked area is identified as the non-wire break area, it means that the color of the marked area image is purple, the guide wire wheel rotates, and the wire is not broken.

[0040] In this implementation scheme, when judging whether the guide wire wheel rotates and whether the wire is broken, it depends on the recognition result of the wire break area in the previous step. If the marked area is identified as the "wire break area", it means that the color of the image in this area presents a mixture of red and blue, indicating that the guide wire wheel does not rotate. At this time, due to the breakage of the wire, the system will judge the breakage of the wire and issue a warning signal. On the contrary, if the marked area is identified as the "non-wire break area", the color of the image is purple, which means that the guide wire wheel is rotating and the wire is not broken. At this time, the system recognizes that the wire is intact and no warning is required. This process determines the rotation state of the guide wire wheel and the breakage state of the wire through the color change pattern, providing a basis for the intelligent monitoring of the entire system. An image resolution single wire detection method includes the following steps: S1. Real-time collect the image of the marked area caused by the identification attachment on the chemical fiber guide wire wheel through a high-resolution camera; S2. Remove the noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, and extract the edge of the marked area in the image by applying the Canny edge detection algorithm to obtain the edge contour image of the marked area; S3. Analyze the marked color change in the edge contour image of the marked area through the contour tracking algorithm, identify the color change amount of the marked area image, obtain the wire break index, and identify whether the marked area is a wire break area according to the preset wire break index threshold; S4. According to the recognition result of the marked area, judge whether the guide wire wheel rotates and judge whether the wire is broken according to the rotation state, and issue a warning message when the wire is broken.

[0041] In this implementation scheme, S1. The image of the marked area caused by the identification accessory on the wire guide wheel is collected in real time through a high-resolution camera to ensure that the changes on the wire guide wheel can be accurately captured. S2. Noise removal and edge extraction: The collected image is subjected to noise removal, the color space is converted to enhance the image quality, and the edge contour of the marked area is extracted through the Canny edge detection algorithm to provide clear boundary data for subsequent analysis. S3. Color change analysis and broken wire index calculation: The color changes in the edge image are analyzed through the contour tracking algorithm, the amplitude of the color change is calculated to generate a broken wire index, and a preset threshold is used to determine whether the marked area is a broken wire area. S4. Judgment of the rotation state of the wire guide wheel and wire breakage: According to the recognition result of the marked area, it is judged whether the wire guide wheel rotates and whether the wire breaks. If the wire breaks, the system issues a warning message.

[0042] In summary, the present application has at least the following effects: An image resolution single wire detection system and method, through a high-resolution camera and advanced image processing algorithms, realizes the accurate acquisition and analysis of the marked area on the chemical fiber wire guide wheel, ensuring that the breakage state of the wire can be accurately identified. It can collect and process images in real time, quickly analyze the color changes in the marked area, timely judge the rotation state of the wire guide wheel and whether the wire breaks, and has high-efficiency real-time response capabilities. When a wire break is detected, the system can automatically issue a warning message to avoid delays in manual monitoring and improve the safety and stability in the production process. Through noise removal algorithms such as Gaussian blur and median filtering, the image quality is effectively improved, and misjudgments caused by environmental interference or equipment problems are reduced. By converting the color space and combining edge detection algorithms, the system can operate stably under different working conditions, accurately identify the color changes in the marked area, has a wide adaptability, and is suitable for a variety of different chemical fiber wire production environments.

[0043] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0044] The present invention is described with reference to the flowcharts and / or block diagrams of systems, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general - purpose computer, a special - purpose computer, an embedded processor, or other programmable data - processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data - processing devices produce a means for realizing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0045] These computer program instructions can also be stored in a computer - readable memory that can direct a computer or other programmable data - processing device to work in a specific manner, such that the instructions stored in the computer - readable memory produce a manufactured article including an instruction means that realizes the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0046] These computer program instructions can also be loaded onto a computer or other programmable data - processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer - implemented process, so that the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0047] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concepts. Therefore, the appended claims are intended to be interpreted to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0048] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. An image resolution single filament detection system, characterized in that, It includes the following modules: an image acquisition module, an image processing module, a color recognition module, and a broken wire determination module; The image acquisition module is used to collect, in real time via a high-resolution camera, an image of a marked area caused by an identification attachment on a chemical fiber guide wheel; The image processing module is used to remove noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, extract the edge of the marked area in the image through the Canny edge detection algorithm, and obtain the edge contour image of the marked area; The color recognition module is used to analyze the color change of the mark in the edge contour image of the marked area through the contour tracking algorithm, identify the color change amount of the marked area image, obtain a broken wire index, and identify whether the marked area is a broken wire area according to a preset broken wire index threshold; The broken wire determination module determines whether the guide wheel rotates according to the recognition result of the color recognition module on the marked area, and determines whether the wire is broken according to the rotation state. When the wire is broken, a warning message is sent; The broken wire index is used to quantify the degree of color change of the mark in the marked area image.

2. The image resolution monofilament detection system according to claim 1, characterized in that: The specific process of collecting an image of a marked area caused by an identification attachment on a chemical fiber guide wheel is as follows: The chemical fiber includes filament and spandex; The camera transmits the collected marked area image to the image processing module. According to the settings of the camera's optical sensor, it is ensured that the image can be accurately captured in different working states, and the marked area image collected in real time is converted into a digital signal.

3. An image resolution monofilament detection system according to claim 2, characterized in that: The specific process of removing noise from the marked area image is as follows: Receive the marked area image transmitted by the image acquisition module; Perform smoothing processing on the marked area image through the Gaussian blur algorithm to remove random noise in the image; Remove salt-and-pepper noise in the image through the median filtering algorithm and enhance the contrast, and retain the details of the marked area image.

4. An image resolution monofilament detection system according to claim 3, characterized in that: The specific process of extracting the edge of the marked area in the image through the Canny edge detection algorithm and obtaining the edge contour image of the marked area is as follows: Convert the marked area image after noise removal into a grayscale image, perform Gaussian filtering on the grayscale image to remove noise and smooth the image; Calculate the gradient value of the grayscale image, calculate the gradients in the horizontal and vertical directions respectively through the Sobel operator, compare each pixel in the grayscale image with its adjacent pixels, and apply the non-maximum suppression algorithm to retain the edge pixels with a set amplitude of the gradient; Through the double-threshold determination method, set a low threshold and a high threshold, screen out strong edges and weak edges, and connect adjacent weak edges to output the edge contour image of the marked area.

5. An image resolution monofilament detection system according to claim 4, characterized in that: The specific process of analyzing the color change of the mark in the edge contour image of the marked area through the contour tracking algorithm and identifying the color change amount of the marked area image is as follows: Extract the edge contour image of the marked area and convert it into a format processed by the contour tracking algorithm; Analyze the edge features of the marked area through the contour tracking algorithm, identify the color change at the edge, and calculate the amplitude of the color change within the marked area.

6. The image resolution monofilament detection system according to claim 5, characterized in that: The specific process of obtaining the broken wire index is as follows: Extract the edge features of the marked area through the contour tracking algorithm, analyze according to the color change amount, calculate the color change amplitude within the marked area through weighted average, normalize the obtained color change amplitude, map it to the set range, and obtain the broken wire index.

7. An image resolution monofilament detection system according to claim 6, characterized in that: The specific process of identifying whether the marked area is a broken wire area according to the preset broken wire index threshold is as follows: When the broken wire index is greater than the preset broken wire index threshold, the marked area is identified as a broken wire area; when it is less than the preset broken wire index threshold, the marked area is identified as an unbroken wire area.

8. An image resolution monofilament detection system according to claim 7, characterized in that: The specific process of judging whether the wire guiding wheel rotates and judging whether the wire is broken according to the rotation state is as follows: Identify the marked area according to the broken wire index. When the marked area is identified as a broken wire area, it means that the color of the marked area image is a mixture of red and blue, the wire guiding wheel does not rotate, and the wire is broken; When the marked area is identified as an unbroken wire area, it means that the color of the marked area image is purple, the wire guiding wheel rotates, and the wire is not broken.

9. An image resolution single filament detection method, applying an image resolution single filament detection system according to any one of claims 1-8, characterized in that, It includes the following steps: S1. Real-time collect the image of the marked area caused by the identification attachment on the chemical fiber wire guiding wheel through a high-resolution camera; S2. Remove the noise from the marked area image, convert the marked area image from the RGB color space to the HSV color space, and extract the edge of the marked area in the image by applying the Canny edge detection algorithm to obtain the edge contour image of the marked area; S3. Analyze the marked color change in the edge contour image of the marked area through the contour tracking algorithm, identify the color change amount of the marked area image, obtain the broken wire index, and identify whether the marked area is a broken wire area according to the preset broken wire index threshold; S4. According to the identification result of the marked area, judge whether the wire guiding wheel rotates, and judge whether the wire is broken according to the rotation state. When the wire is broken, a warning message is sent.

Images