Carbon fiber defect detection method, device, equipment and medium

Abstract

The invention relates to the technical field of carbon fiber production, in particular to a carbon fiber defect detection method, device and equipment and a medium, and the method comprises the following steps: acquiring an image of carbon fibers under detection light, and converting the image into a grayscale image; calculating a threshold value of each pixel point in the gray level image through a local moving average threshold value, respectively judging whether the pixel point is an image background or a carbon fiber, and extracting pixel points in an area of the carbon fiber; carrying out contour extraction on the extracted carbon fiber region, and calculating a minimum perimeter polygon and a minimum enclosing rectangle of the extracted contour; and calculating the rectangle similarity of the minimum perimeter polygon and the minimum enclosing rectangle, judging whether the rectangle similarity accords with a set threshold value, and if so, judging that defects exist. According to the invention, the method can adapt to illumination changes, avoids false detection caused by environmental factors, judges defects through regional form analysis, does not depend on a single defect shape, and improves the accuracy and stability of carbon fiber defect detection.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon fiber production, and in particular to a carbon fiber defect detection method, device, equipment and medium. Background Art

[0002] Carbon fiber has a series of excellent properties such as high specific strength, high specific modulus, fatigue resistance, creep resistance, corrosion resistance and small thermal expansion coefficient. Therefore, it is widely used as the reinforcement of carbon fiber reinforced resin-based composites (CFRP) in cutting-edge defense technologies such as aerospace and aviation. It is also a new material for the upgrading of civilian industries.

[0003] In the process of carbon fiber production, defects such as hair balls, hairy fibers, broken fibers, and joints will appear. Relying solely on manual visual methods to detect defects in the carbon fiber production process will inevitably result in missed detections, so it is necessary to rely on a visual system to conduct real-time online monitoring of carbon fiber defects in the carbon fiber production process.

[0004] In the online monitoring of the visual system, a fixed threshold is often set to segment defects and identify them. However, as time goes by, the attenuation of the visual light source and the change of the ambient light source will greatly affect the visual imaging system. If the defects are segmented by a fixed threshold alone, it is easy to cause certain false alarms.

[0005] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides a carbon fiber defect detection method, device, equipment and medium, thereby effectively solving the problems in the background technology.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is: a carbon fiber defect detection method, comprising the following steps: Obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; The threshold of each pixel in the grayscale image is calculated by a local moving average threshold, and it is judged to be the image background or the carbon fiber, and the pixel in the carbon fiber area is extracted; Performing contour extraction on the extracted carbon fiber area, and calculating the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; The rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle is calculated, and it is determined whether the rectangle similarity meets a set threshold, and if so, it is determined that there is a defect.

[0008] Furthermore, the calculating of the threshold of each pixel in the grayscale image by using the local moving average threshold includes: Performing a Z-scan on the pixel points in the grayscale image to convert the two-dimensional image into one-dimensional data; Calculate the local moving average threshold of the corresponding pixel points of the one-dimensional data through a window of a set size; The magnitude of the grayscale value of the pixel point and the local moving average threshold is determined, and if the grayscale value is less than the local moving average threshold, it is determined to be the image background.

[0009] Furthermore, the calculating the local moving average threshold of the corresponding pixel points of the one-dimensional data through a window of a set size includes: ; In the formula, k represents the kth point in the one-dimensional data, n is the window size, and Z i is the gray value of the pixel corresponding to the i-th point, and m(k) represents the calculated local moving average threshold of the pixel corresponding to the k-th point.

[0010] Furthermore, the window size n corresponds to the width of the carbon fiber in the grayscale image.

[0011] Furthermore, the contour extraction of the extracted carbon fiber region comprises the following steps: Let the extracted carbon fiber area be g0, and find the pixel point p at the upper left corner of the area; Find the pixel q outside the region g0 in the eight-domain of the pixel p, and start from the pixel q and search clockwise for the first pixel in the eight-domain of the pixel p that is inside the region g0, which is recorded as p1; Repeat the search for the pixel q1 outside the region g0 in the eight-domain of the pixel point p1 and start from the pixel q1, and search clockwise for the first pixel in the eight-domain of the pixel point p1 that is inside the region g0, which is recorded as p2, and repeat the search for p3, p4, ..., pn; The point set [p0, p1, ···, pn] of p is the contour of region g0.

[0012] Furthermore, the calculating of the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle includes: D = S(A) / S(B); Wherein, D is the rectangle similarity, S(A) is the area of ​​the minimum perimeter polygon, and S(B) is the area of ​​the minimum circumscribed rectangle.

[0013] The present invention also includes a carbon fiber defect detection device, using the above method, the device includes: An acquisition unit, used to obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; A dynamic threshold unit, used to calculate the threshold of each pixel in the grayscale image by a local moving average threshold, and to determine whether it is an image background or a carbon fiber, and to extract the pixel points in the carbon fiber area; A contour extraction unit, used to extract the contour of the extracted carbon fiber area, and calculate the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; The defect judgment unit is used to calculate the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle, and judge whether the rectangle similarity meets the set threshold, and if so, judge that there is a defect.

[0014] The present invention also includes a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above method is implemented.

[0015] The present invention also includes a storage medium on which a computer program is stored. When the computer program is executed by a processor, the method described above is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention can adapt to lighting changes through local moving average threshold segmentation to avoid false detection caused by environmental factors, adopts rectangular similarity calculation to judge defects through regional morphological analysis, does not rely on a single defect shape, reduces the risk of missed detection, and can quickly process images to meet the real-time detection needs of carbon fiber production lines. It not only improves the accuracy and stability of carbon fiber defect detection, but also reduces the occurrence of false alarms and missed alarms, so that the system can be more reliably applied in industrial production environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 is a flow chart of the method of the present invention; Figure 2 It is a structural schematic diagram of the device of the present invention; Figure 3 It is a schematic diagram of the structure of the computer device of the present invention. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0020] like Figure 1 As shown: A carbon fiber defect detection method comprises the following steps: Obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; The threshold of each pixel in the grayscale image is calculated by the local moving average threshold, and it is judged as the image background or carbon fiber, and the pixel points in the carbon fiber area are extracted; Performing contour extraction on the extracted carbon fiber area, and calculating the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; Calculate the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle, and determine whether the rectangle similarity meets the set threshold. If so, it is determined that there is a defect.

[0021] The local moving average threshold method is applied to each pixel point. The segmentation threshold of each pixel point is determined by dynamically calculating the grayscale values ​​of adjacent pixel areas. Each pixel point is compared with its local threshold. Pixels greater than the threshold are judged as carbon fiber areas, and pixels less than the threshold are judged as background areas. This method can adapt to changes in light sources and avoid false detections caused by uneven lighting or light source attenuation.

[0022] The extracted contour is further analyzed to calculate its minimum perimeter polygon (MPP) and minimum circumscribed rectangle. When the shape of the carbon fiber area is close to a regular rectangle, the rectangle similarity value is close to 1; if the shape of the area is irregular, such as defects, the similarity value deviates significantly from 1. According to the set rectangle similarity threshold D0, each extracted carbon fiber area is judged based on the overall morphology of the carbon fiber area, and the shape invariance and relative relationship are used to avoid false positives or negatives that may be caused by relying solely on appearance features.

[0023] Through local moving average threshold segmentation, it can adapt to lighting changes and avoid false detection caused by environmental factors. It uses rectangular similarity calculation and regional morphology analysis to judge defects. It does not rely on a single defect shape and reduces the risk of missed detection. It can quickly process images to meet the real-time detection needs of carbon fiber production lines. It not only improves the accuracy and stability of carbon fiber defect detection, but also reduces the occurrence of false alarms and missed alarms, making the system more reliably applied in industrial production environments.

[0024] In this embodiment, the threshold of each pixel in the grayscale image is calculated by using the local moving average threshold, including: Perform a Z-scan on the pixels in the grayscale image to convert the two-dimensional image into one-dimensional data; Calculate the local moving average threshold of the corresponding pixel points of the one-dimensional data through a window of set size; Determine the gray value of the pixel and the size of the local moving average threshold. If the gray value is less than the local moving average threshold, it is judged as the image background.

[0025] The pixels in the grayscale image are scanned along a zigzag path. This scanning method starts from the upper left corner of the image, first scans horizontally line by line to the right side of the image, then scans the next line, and so on, until the entire image is scanned. In this way, the original two-dimensional grayscale image is converted into a one-dimensional pixel data sequence. The purpose of Z-scanning is to simplify image processing so that complex two-dimensional data can be quickly calculated in a linear manner while maintaining the local correlation of the image structure.

[0026] For the one-dimensional data obtained by Z-scanning, a sliding window is used to average the local grayscale value of each pixel. The size of the window can be set according to the actual situation, and is usually set to a size close to the width of the carbon fiber yarn to ensure that the calculated threshold can accurately reflect the grayscale distribution characteristics of the local area. Within the sliding window, the grayscale average of all pixels in the area is calculated and used as the local moving average threshold of the current pixel. Every time the window moves one pixel, the average grayscale value of the point is recalculated.

[0027] By dynamically calculating the threshold for each pixel through a sliding window, it can adapt to changes in illumination in different areas and avoid misjudgment caused by fixed thresholds in complex environments. Z-scan simplifies a two-dimensional image into one-dimensional data, and combined with sliding window technology, it not only ensures the accuracy of the calculation, but also improves the efficiency of image processing. It is suitable for real-time online detection systems. For pixels at the edge of the image, the original value is retained, that is, m1=z1, because the pixels at the edge of the image do not have a great impact on the overall target extraction and can be ignored.

[0028] The local moving average threshold of the corresponding pixel points of the one-dimensional data is calculated by using a window of a set size, including: ; In the formula, k represents the kth point in the one-dimensional data, n is the window size, and Z i is the gray value of the pixel corresponding to the i-th point, and m(k) represents the calculated local moving average threshold of the pixel corresponding to the k-th point.

[0029] As a preference of the above embodiment, the window size n corresponds to the width of the carbon fiber in the grayscale image.

[0030] During the production process, the width of the carbon fiber is usually fixed and continuous, corresponding to a specific grayscale change area in the image. Therefore, the width of the carbon fiber in the image can be obtained by measurement and used as the benchmark for the sliding window. The window size n determines the pixel range referenced when calculating the local moving average threshold. If the window is too large, local details may be smoothed out, resulting in unclear segmentation of the carbon fiber and the background; if the window is too small, the change in the threshold will be too sensitive and easily affected by noise or local illumination changes, resulting in false detection or missed detection. By setting the window size to correspond to the width of the carbon fiber, it can be ensured that the calculated local average value can accurately reflect the brightness characteristics of the area where the carbon fiber is located, effectively distinguish the carbon fiber from the background, and reduce false positives and missed detections.

[0031] In this embodiment, contour extraction is performed on the extracted carbon fiber region, including the following steps: Let the extracted carbon fiber area be g0, and find the pixel point p at the upper left corner of the area; Find the pixel q outside the region g0 in the eight-domain of the pixel p, and start from the pixel q and search clockwise for the first pixel in the eight-domain of the pixel p that is inside the region g0, which is recorded as p1; Repeat the search for the pixel q1 outside the region g0 in the eight-domain of the pixel point p1 and start from the pixel q1, and search clockwise for the first pixel in the eight-domain of the pixel point p1 that is inside the region g0, which is recorded as p2, and repeat the search for p3, p4, ..., pn; The point set [p0, p1, ···, pn] of p is the contour of region g0.

[0032] This method can gradually and accurately extract the complete contour from the boundary of the carbon fiber area, effectively avoiding missing or incorrectly extracting boundary points. The eight-neighborhood search can capture changes in all directions of the edge of the area, ensuring the coherence and integrity of the contour. It is suitable for contour extraction of carbon fiber areas with complex shapes and can handle irregular edges or noise. Through this contour extraction method, the boundary of the carbon fiber area can be accurately identified, providing a reliable basis for subsequent polygon fitting, rectangular bounding box calculation and defect detection. This fine contour extraction method can also improve the detection accuracy of the system, especially when dealing with complex backgrounds or lighting changes, showing higher robustness.

[0033] As a preferred embodiment of the above, calculating the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle includes: D = S(A) / S(B); Where D is the rectangle similarity, S(A) is the area of ​​the minimum perimeter polygon, and S(B) is the area of ​​the minimum circumscribed rectangle.

[0034] like Figure 2 As shown, this embodiment also includes a carbon fiber defect detection device, using the above method, the device includes: An acquisition unit, used to obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; A dynamic threshold unit is used to calculate the threshold of each pixel in the grayscale image through a local moving average threshold, and to determine whether it is the image background or the carbon fiber, and to extract the pixel points in the carbon fiber area; A contour extraction unit, used for performing contour extraction on the extracted carbon fiber region and calculating the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; The defect judgment unit is used to calculate the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle, and judge whether the rectangle similarity meets the set threshold. If it meets the threshold, it is judged that there is a defect.

[0035] See also Figure 3 A computer device 400 provided in an embodiment of the present application includes: a processor 410 and a memory 420, wherein the memory 420 stores a computer program executable by the processor 410, and when the computer program is executed by the processor 410, the above method is executed.

[0036] The embodiment of the present application further provides a storage medium 430 on which a computer program is stored. When the computer program is run by the processor 410, the above method is executed.

[0037] Among them, the storage medium 430 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, disk or optical disk.

[0038] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. "Multiple" means two or more, unless otherwise clearly and specifically defined.

[0039] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0040] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0041] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code that includes one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present invention belong.

[0042] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.

[0043] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0044] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0045] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A carbon fiber defect detection method, characterized in that: The steps include: Obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; The threshold of each pixel in the grayscale image is calculated by a local moving average threshold, and it is judged to be the image background or the carbon fiber, and the pixel in the carbon fiber area is extracted; Performing contour extraction on the extracted carbon fiber area, and calculating the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; The rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle is calculated, and it is determined whether the rectangle similarity meets a set threshold, and if so, it is determined that there is a defect.

2. The carbon fiber defect detection method according to claim 1, characterized in that: The calculating the threshold of each pixel in the grayscale image by using the local moving average threshold comprises: Performing a Z-scan on the pixel points in the grayscale image to convert the two-dimensional image into one-dimensional data; Calculate the local moving average threshold of the corresponding pixel points of the one-dimensional data through a window of a set size; The magnitude of the grayscale value of the pixel point and the local moving average threshold is determined, and if the grayscale value is less than the local moving average threshold, it is determined to be the image background.

3. The carbon fiber defect detection method according to claim 2, characterized in that: The calculating the local moving average threshold of the corresponding pixel points of the one-dimensional data through a window of a set size includes: ； In the formula, k represents the kth point in the one-dimensional data, n is the window size, and Z i is the gray value of the pixel corresponding to the i-th point, and m(k) represents the calculated local moving average threshold of the pixel corresponding to the k-th point.

4. The carbon fiber defect detection method according to claim 3, characterized in that: The window size n corresponds to the width of the carbon fiber in the grayscale image.

5. The carbon fiber defect detection method according to claim 1, characterized in that: The step of extracting the contour of the extracted carbon fiber region comprises the following steps: Let the extracted carbon fiber area be g0, and find the pixel point p at the upper left corner of the area; Find the pixel q outside the region g0 in the eight-domain of the pixel p, and start from the pixel q and search clockwise for the first pixel in the eight-domain of the pixel p that is inside the region g0, which is recorded as p1; Repeat the search for the pixel q1 outside the region g0 in the eight-domain of the pixel point p1 and start from the pixel q1, and search clockwise for the first pixel in the eight-domain of the pixel point p1 that is inside the region g0, which is recorded as p2, and repeat the search for p3, p4, ..., pn; The point set [p0, p1, ···, pn] of p is the contour of region g0.

6. The carbon fiber defect detection method according to claim 1, characterized in that: The calculating of the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle comprises: D = S(A) / S(B); Wherein, D is the rectangle similarity, S(A) is the area of ​​the minimum perimeter polygon, and S(B) is the area of ​​the minimum circumscribed rectangle.

7. A carbon fiber defect detection device, characterized in that: Using the method according to any one of claims 1 to 6, the device comprises: An acquisition unit, used to obtain an image of the carbon fiber under the detection light and convert it into a grayscale image; A dynamic threshold unit, used to calculate the threshold of each pixel in the grayscale image by a local moving average threshold, and to determine whether it is an image background or a carbon fiber, and to extract the pixel points in the carbon fiber area; A contour extraction unit, used to extract the contour of the extracted carbon fiber area, and calculate the minimum perimeter polygon and the minimum circumscribed rectangle of the extracted contour; The defect judgment unit is used to calculate the rectangle similarity between the minimum perimeter polygon and the minimum circumscribed rectangle, and judge whether the rectangle similarity meets the set threshold, and if so, judge that there is a defect.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images