WO2023126030A1

WO2023126030A1 - Switch interface integrity inspection method

Info

Publication number: WO2023126030A1
Application number: PCT/CN2023/086364
Authority: WO
Inventors: 韦振; 魏荣生
Original assignee: 探长信息技术(苏州)有限公司
Priority date: 2022-09-28
Filing date: 2023-04-05
Publication date: 2023-07-06
Also published as: LU504273B1; CN115471486A

Abstract

Disclosed in the present invention is a switch interface integrity inspection method, relating to the field of machine vision. The method mainly comprises: obtaining a grayscale image of a switch image containing a network interface, and performing bilinear interpolation on the grayscale image to obtain an up-sampling image; performing histogram equalization on the up-sampling image to obtain an enhanced image, and in the histogram equalization process, according to a frequency ratio of the enhanced image to the up-sampling image and a frequency ratio of the enhanced image to all new grayscale levels, respectively determining a grayscale level of each new grayscale level after mapping, the new grayscale levels being the grayscale levels existing in the up-sampling image but does not exist in the grayscale image; and matching by means of a standard network interface template, so as to determine whether an abnormal network interface exists in the enhanced image, and locate the abnormal network interface. By means of embodiments of the present invention, all network interfaces contained in the target image can be inspected at the same time without sequential visual inspection, such that the duration required for inspection is shortened.

Description

A method for detecting integrity of switch interface

technical field

The present application relates to the field of machine vision, in particular to a method for detecting the integrity of a switch interface.

Background technique

The network interface of the switch may be damaged to a certain extent due to bumps and other reasons in the transportation process, resulting in poor contact at the network interface and affecting subsequent use.

Therefore, after batches of switches are shipped to the point of sale or inventory point, relevant personnel need to select a certain proportion of the switches for visual inspection, and check the integrity of each network interface one by one during the visual inspection process to determine the integrity of all switches. There are no abnormal network interfaces in the inspected batch.

However, the size of the network interface is small. In order to ensure the accuracy of the test results, the inspectors tend to spend a lot of time on the inspection. Even so, it may be difficult to ensure the accuracy of the inspection. Inspection efficiency and accuracy.

Contents of the invention

In view of the above technical problems, the present invention provides a switch interface integrity detection method, which obtains an upsampled image by grayscale and bilinear interpolation on the front image of the switch containing the network interface, and detects new The gray level and the gray level other than the new gray level respectively determine the gray level after mapping, obtain the target image with more details, and finally realize the integrity detection of the network interface in the switch by means of template matching. It is possible to test all the network interfaces contained in the target image at the same time without visual inspection in sequence, thereby shortening the time required for inspection, and effectively avoiding the visual fatigue of manual visual inspection, thereby improving the network interface of the switch. Accuracy and efficiency of integrity checks.

The embodiment of the present invention proposes a switch interface integrity detection method, including:

The front image of the switch including the network interface is collected and grayscaled to obtain a grayscale image, and the grayscale image is upsampled by bilinear interpolation to obtain an upsampled image.

Perform histogram equalization on the upsampled image to obtain an enhanced image, and in the process of histogram equalization, determine each new gray level in the upsampled image and the gray level after gray level mapping other than the new gray level, Wherein, the gray level after mapping of each new gray level is determined according to its frequency ratio in the upsampled image and its frequency ratio to all new gray levels in the upsampled image. A grayscale is a grayscale that exists in the upsampled image but not in the grayscale image.

Template matching is performed on the enhanced image by using the template corresponding to the standard network interface, judging whether there is an abnormal network interface in the enhanced image according to the matching result, and locating the abnormal network interface if there is an abnormal network interface.

Further, in the switch interface integrity detection method, the gray level after each new gray level is mapped according to its frequency ratio in the upsampled image and its relationship with all new gray levels in the upsampled image. Frequency ratios are determined, including:

According to the frequency ratio of each new gray level in the upsampled image and the ratio of each new gray level to the frequency of all new gray levels in the upsampled image, the cumulative distribution of each new gray level is determined respectively The value after function mapping, wherein the larger the ratio of each new gray level to the frequency of all new gray levels in the upsampled image, the smaller the value after mapping the cumulative distribution function corresponding to each new gray level.

According to the mapped value of the cumulative distribution function of each newborn gray level, the mapped gray level of each newborn gray level is respectively determined.

Further, in the switch interface integrity detection method, the enhanced image is template-matched using the template corresponding to the standard network interface, and whether there is an abnormal network interface in the enhanced image is judged according to the matching result, including:

The similarity between the template corresponding to the standard network interface and the area to be tested in the enhanced image is obtained, and the area to be tested is an area in the enhanced image that is as large as the template.

In a case where the similarity is greater than the preset first threshold, it is determined that the area to be tested is an area where a normal network interface is located.

In the case that the similarity is not greater than the preset first threshold, it is judged whether the similarity is greater than the preset second threshold, and if the judgment result is yes, the area to be tested is determined to be the area where the abnormal interface is located, wherein the preset The second threshold is smaller than the preset first threshold.

Further, in the method for detecting the integrity of the switch interface, the similarity is the structural similarity SSIM.

Further, in the method for detecting the integrity of the switch interface, the similarity is a Pearson correlation coefficient.

Further, in the method for detecting the integrity of the switch interface, before collecting the front image of the switch including the network interface, the method further includes: setting a parallel light to illuminate the switch including the network interface.

Further, in the switch interface integrity detection method, after performing histogram equalization on the upsampled image to obtain the enhanced image, the method further includes: performing edge detection on the enhanced image, and using the edge detection result as a new enhanced image.

Further, in the switch interface integrity detection method, the operator used for edge detection is any one of Roberts operator, Prewitt operator, Sobel operator and Canny operator.

Further, the method for detecting the integrity of the switch interface also includes: locating the abnormal network interface if there is an abnormal network interface.

The embodiment of the present invention provides a method for detecting the integrity of a switch interface. Compared with the prior art, the beneficial effect of the embodiment of the present invention is that all network interfaces contained in the target image can be tested simultaneously without visual inspection in sequence Inspection, thereby shortening the time required for the inspection, and effectively avoiding visual fatigue caused by manual visual inspection, thereby improving the accuracy and efficiency of the integrity inspection of the network interface of the switch.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is a schematic flowchart of a method for detecting the integrity of a switch interface provided by an embodiment of the present invention.

Detailed ways

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.

The terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of these features; in the description of this embodiment, unless otherwise specified, the meaning of "plurality" is two or more.

The embodiment of the present invention provides a switch interface integrity detection method, as shown in Figure 1, including:

Step S101 , collecting a front image of a switch including a network interface and performing grayscale conversion to obtain a grayscale image, and performing upsampling on the grayscale image through bilinear interpolation to obtain an upsampled image.

Step S102: Perform histogram equalization on the upsampled image to obtain an enhanced image, and in the process of histogram equalization, respectively determine each new gray level in the upsampled image and gray levels after gray level mapping other than the new gray level. The new gray level is a gray level that exists in the upsampled image but not in the grayscale image.

Wherein, the gray level after mapping of each new gray level is determined according to its frequency ratio in the upsampled image and its frequency ratio to all new gray levels in the upsampled image.

Step S103 , use the template corresponding to the standard network interface to perform template matching on the enhanced image, judge whether there is an abnormal network interface in the enhanced image according to the matching result, and locate the abnormal network interface if there is an abnormal network interface.

Through the above steps, all network interfaces included in the enhanced image can be tested simultaneously without visual inspection in sequence, thereby shortening the time required for inspection, and effectively avoiding visual fatigue caused by manual visual inspection, thereby improving the reliability of the switch. The accuracy and efficiency of the integrity verification of the network interface.

The purpose of the embodiment of the present invention is to realize the integrity detection of the rated network interface existing in the switch through technical means such as machine vision and image processing, shorten the time required for detection and improve the detection accuracy.

Further, step S101 , collecting a front image of a switch including a network interface and performing grayscale conversion to obtain a grayscale image, and performing upsampling on the grayscale image through bilinear interpolation to obtain an upsampled image.

Multi-layer shelves can be set up to store switches, and at the same time, the switches are oriented uniformly so as to collect images on the side of the switch including the network interface. The resolution performance of the camera needs to be high enough, and the indoor ambient light can be set in advance to ensure that the ambient light Bright and uniform enough to make the details of the captured image appear more clearly.

Since the positions of different switches on the shelves may be different, there may be blind spots in the image acquisition device, so that a complete frontal image of the switch containing the network interface may not be captured, so multiple cameras can be set in the embodiment of the present invention , and fix the shooting angle of each camera separately, and each camera only collects the switch image within the horizontal viewing angle range of the camera lens.

In the embodiment of the present invention, it is also possible to arrange the image acquisition device in parallel on the front of the switch containing the network interface, move the image acquisition device and take images, and finally splicing all the captured images to obtain the front image of the switch. In this way, it is possible Effectively avoid the different lens distances between different positions in the switchboard and the image acquisition device, which may have adverse effects on the collected images.

Optionally, before collecting the front image of the switch including the network interface, parallel light perpendicular to the switch including the network interface can also be set to irradiate the switch, which helps to better present the switch in the collected front image Details of the interface, so as to improve the subsequent integrity detection accuracy of the network interface.

Perform conventional grayscale processing on the collected images to remove the influence of redundant color factors, which can effectively reduce the amount of calculation in the detection process. The grayscale process includes: taking the maximum value of pixel values in the three channels of RGB in the front image of the switch including the network interface as the grayscale value of the pixel in the grayscale image.

For the appearance inspection of batch switch interfaces, it is mainly aimed at the deformation and defect of metal copper contacts. However, it is difficult to effectively find possible deformation or damage in a short period of time through visual observation, making it difficult for the integrity of the network interface of the switch. Detection efficiency is limited.

In the same image, there are multiple network interfaces on the side of the network interface of the switch. Therefore, if it is directly processed, the area corresponding to each network interface is relatively small, and it is impossible to obtain more accurate detection results. Therefore, the present invention The embodiment expects to retain the information in the original image while obtaining a larger resolution image and processing it. At the same time, compared with the nearest neighbor interpolation method, bilinear interpolation has less aliasing in the obtained up-sampled image, which can The information in the original image is preserved more completely. Therefore, in the embodiment of the present invention, the bilinear interpolation method is used to realize the upsampling processing of the grayscale image, and the upsampled The upsampled image is larger in size after sampling, and the bilinear interpolation process in the embodiment of the present invention adopts the same upsampling ratio for rows and columns.

It should be noted that image scaling can be achieved by using bilinear interpolation. The steps of image scaling using bilinear interpolation are as follows: calculate the scaling factor between the target picture and the original picture, where the scaling factors are respectively included in the row The ratio of the direction and the ratio of the column direction; using the scaling factor, the pixel position of the target image is pushed back to the virtual pixel position in the original image; the four pixel points adjacent to the width and height directions are found from the virtual pixel position, and four pixels The point is calculated by bilinear interpolation to obtain the pixel value in the target image. Specifically, each pixel of the image output by the bilinear interpolation algorithm is the result of an operation of four pixels (2×2) in the original image.

Further, in step S102, perform histogram equalization on the upsampled image to obtain an enhanced image, and in the process of histogram equalization, determine each new gray level in the upsampled image and the gray level mapping other than the new gray level grayscale after.

Wherein, the gray level after mapping of each new gray level is determined according to its frequency ratio in the upsampled image and its frequency ratio to all new gray levels in the upsampled image. A grayscale is a grayscale that exists in the upsampled image but not in the grayscale image.

In the embodiment of the present invention, histogram equalization is used to enhance the contrast of the image to improve the definition of blurred edge contours. However, when the histogram is equalized, some gray levels with a small number of pixels may be lost indiscriminately, or These small number of gray levels are not obvious, which may lead to inconspicuous gray levels of the copper contacts themselves and edges with less pixel distribution, resulting in inability to effectively perform subsequent integrity detection of the network interface.

After bilinear interpolation, the generated pixels may be in the edge high-frequency area or uniform low-frequency area of the upsampled image, while the edge high-frequency area in the grayscale image is more likely to be born in the upsampled image. Gray level, that is, the gray level that does not exist in the original grayscale image, because these pixels are located at the high-frequency edge with a gray gradient, and its adjacent pixels are more likely to appear when performing bilinear interpolation calculations The new gray level, on the contrary, in the low-frequency area where the gray value is relatively uniform, the gray value of adjacent pixels is relatively close, and the gray level generated after bilinear interpolation is relatively close to these adjacent pixels. grayscale.

Therefore, in the embodiment of the present invention, the first grayscale histogram corresponding to the upsampled image and the second grayscale histogram corresponding to the grayscale image can be obtained respectively, so as to filter out the grayscale histogram that exists in the first grayscale histogram but does not exist in the grayscale histogram. Each nascent gray level in the second gray level histogram.

In the embodiment of the present invention, the gray levels after mapping of each new gray level account for all new gray levels in the first gray histogram according to the frequency of each new gray level in the first gray histogram. The scale of the frequencies in the graph is determined.

Histogram equalization is a simple and effective image enhancement technology, which changes the gray level of each pixel in the image by changing the histogram of the image, and is mainly used to enhance the contrast of images with a small dynamic range. Because the gray distribution in the original image may be concentrated in a narrow range, the image is not clear enough. For example, the gray level of an overexposed image is concentrated in the high brightness range, while underexposure will make the gray level of the image concentrated in the low brightness range.

Using histogram equalization, the histogram of the original image can be transformed into a form of uniform distribution, which increases the dynamic range of the gray value difference between pixels, thereby achieving the effect of enhancing the overall contrast of the image. In other words, the basic principle of histogram equalization is to widen the gray value with a large number of pixels in the image (that is, the gray value that plays a major role in the picture), and to widen the gray value with a small number of pixels (that is, the gray value that plays a major role in the picture). Merge the gray values that do not play a major role in the picture), thereby increasing the contrast, making the image clear, and achieving the purpose of enhancement.

However, for the upsampled image obtained in the embodiment of the present invention, there are new gray levels that do not exist in the original gray scale image. The existence of these new gray levels can make the network in the embodiment of the present invention The details of the interface can be completely preserved, but if the gray level histogram equalization is directly performed on the upsampled image, these new gray levels in the enhanced image after histogram equalization are retained in the enhanced image due to their small frequency. It is not obvious, so that it is still impossible to perform integrity detection of the network interface on the basis of the enhanced image.

In the embodiment of the present invention, each new gray level and the gray level after gray level mapping other than the new gray level are respectively determined, including:

According to the frequency ratio of each new gray level in the upsampled image and the ratio of each new gray level to the frequency of all new gray levels in the upsampled image, the cumulative distribution of each new gray level is determined respectively The value after function mapping, wherein the larger the ratio of the frequency of each new gray level to all new gray levels in the upsampled image, the smaller the value after mapping the cumulative distribution function corresponding to each new gray level; And according to the mapped value of the cumulative distribution function of each new gray level, the gray level after mapping of each new gray level is respectively determined.

At the same time, according to the frequency ratio of the gray levels other than each new gray level in the up-sampled image, the values after the cumulative distribution function mapping of each gray level other than the new gray level are respectively determined.

Specifically, the process of obtaining the value after the cumulative distribution function mapping of each gray level in the upsampled image includes:

Among them, s _k is the value mapped by the cumulative distribution function corresponding to the kth gray level, k is an integer in the range of [0, L-1], j is an integer in the range of [0, k], and L is the above The total number of gray levels in the sampled image, G _j is the frequency of the jth gray level, and M and N are the length and width of the image, respectively.

At the same time, when the kth gray level is a new-born gray level, the value of g(k) is the reciprocal of the frequency ratio of the new-born gray level in all new-born gray levels. It should be noted that the new-born gray level The frequency of the degree level refers to the number of pixels of the new gray level in the upsampled image; in this way, the values after the cumulative distribution function mapping of each gray level in the upsampled image are respectively obtained.

Finally, after the mapped values of the cumulative distribution functions of all gray levels are determined, the mapped values of the cumulative distribution functions of all gray levels are normalized to the range [0, 255] to obtain the above The histogram-equalized gray level corresponding to each gray level in the sampled image.

In this way, the image resolution is improved, and the image is enhanced, and the detail information that is less distributed in the image is retained as much as possible. The detail information refers to the tiny damage in the image or the edge information of the copper contact.

Optionally, after the histogram-equalized enhanced image is obtained, edge detection may be performed on the enhanced image, and the edge detection result may be used as a new enhanced image. In this way, the amount of calculation in the subsequent template matching process can be reduced.

The image edge is the most basic feature of the image, and the so-called edge (Edge) refers to the discontinuity of the local characteristics of the image. The sudden change of information such as grayscale or structure is called edge. For example, the sudden change of gray level, the sudden change of color, the sudden change of texture structure, etc. An edge is the end of one region and the beginning of another, and this feature can be used to segment an image. The edge of an image has two attributes, direction and magnitude. Edges can usually be detected by first or second derivatives. The first derivative takes the maximum value as the corresponding edge position, while the second derivative takes the zero-crossing point as the corresponding edge position.

The process of edge detection by the edge operator of the first derivative includes: using the template as a kernel to perform convolution and operation with each pixel of the image, and then selecting an appropriate threshold to extract the edge of the image. Common marginal operators of the first derivative include: Roberts operator, Prewitt operator, Sobel operator, Canny operator, etc. The process of edge detection by the edge operator of the second derivative is based on the characteristic of the zero crossing point of the second derivative. The common The edge operator with second order derivative has Laplacian operator.

The operator used in the edge detection process in the embodiment of the present invention is any one of Roberts operator, Prewitt operator, Sobel operator and Canny operator.

Optionally, for the enhanced image, there are both rich image details and clear contrast. At the same time, the Canny operator has the advantage of dual threshold detection, so the Canny operator can be preferentially used for edge detection.

Further, in step S103 , template matching is performed on the enhanced image using the template corresponding to the standard network interface, judging whether there is an abnormal network interface in the enhanced image according to the matching result, and locating the abnormal network interface if there is an abnormal network interface.

Template matching is a process of traversing the enhanced image using a template. At the same time, during the traversal process, if the area to be tested is the area where the network interface is normal, that is, there is no defect, the similarity between the area to be tested and the template will be expressed as greater than the preset first threshold; otherwise, if the similarity between the area to be tested and the template is not greater than the preset threshold, the area to be tested may be the area where the defective network interface is located, or the area to be tested may not be the network interface However, since there is a certain degree of similarity between the region where the defective network interface is located and the template, therefore, when the similarity between the template and the region to be tested is greater than the preset second threshold, the The area to be tested is determined to be the area where the abnormal network interface is located. In this way, the location of the possible abnormal network interface can be realized while implementing the integrity test on the network interface.

Specifically, the similarity in the embodiment of the present invention may be the structural similarity SSIM, the Pearson correlation coefficient, or other feature values used to calculate the similarity between images.

Structural similarity SSIM (structural similarity index) is an index to measure the similarity of two images. From the perspective of image composition, structural information is defined as independent of brightness and contrast, reflecting the attributes of the object structure in the scene, and Distortion is modeled as a combination of three different factors: brightness, contrast, and structure, where the mean is an estimate of brightness, the standard deviation is an estimate of contrast, and the covariance is a measure of structural similarity.

In statistics, Pearson correlation coefficient (Pearson correlation coefficient), also known as Pearson product-moment correlation coefficient (PPMCC or PCCs for short), can be applied to calculate the similarity between two images sex.

To sum up, the embodiment of the present invention provides a method for detecting the integrity of a switch interface. The upsampling image is obtained by grayscale and bilinear interpolation on the front image of the switch including the network interface, and the upsampling image The new gray level and the gray level other than the new gray level respectively determine the gray level after mapping, obtain the target image with more details, and finally realize the integrity detection of the network interface in the switch by means of template matching .

The embodiment of the present invention can test all the network interfaces contained in the target image at the same time without visual inspection sequentially, thereby shortening the time required for inspection, and effectively avoiding visual fatigue caused by manual visual inspection, thereby improving the reliability of the switch. The accuracy and efficiency of the integrity verification of the network interface.

Words such as "including", "comprising", "having" and so on in the present invention are open words, mean "including but not limited to", and can be used interchangeably. As used herein, the words "or" and "and" refer to the word "and/or" and are used interchangeably therewith, unless the context clearly dictates otherwise. As used herein, the word "such as" refers to the phrase "such as but not limited to" and can be used interchangeably therewith.

It should also be pointed out that in the method and system of the present invention, each component or each step can be decomposed and/or reassembled. These decompositions and/or recombinations should be considered equivalents of the present disclosure.

The above-mentioned embodiments are only examples for clear description, and do not constitute limitations on the protection scope of the present invention. For those of ordinary skill in the art, on the basis of the above description, other changes or changes in different forms can also be made, and it is not necessary and impossible to exhaustively enumerate all the implementation manners here. All designs identical or similar to the present invention fall within the protection scope of the present invention.

Claims

A switch interface integrity detection method, characterized in that, comprising:

Collect the front image of the switch including the network interface and grayscale to obtain a grayscale image, and upsample the grayscale image through bilinear interpolation to obtain an upsampled image;

Perform histogram equalization on the upsampled image to obtain an enhanced image, and in the process of histogram equalization, determine each new gray level in the upsampled image and the gray level after gray level mapping other than the new gray level, Wherein, the gray level after mapping of each new gray level is determined according to its frequency ratio in the upsampled image and its frequency ratio to all new gray levels in the upsampled image. Grayscale is the grayscale present in the upsampled image but not present in the grayscale image;

Template matching is performed on the enhanced image by using the template corresponding to the standard network interface, and whether there is an abnormal network interface in the enhanced image is judged according to the matching result.
A method for detecting the integrity of a switch interface according to claim 1, wherein the gray level after each new gray level is mapped according to its frequency ratio in the upsampled image, and its relationship with the above The frequency ratio of all newborn gray levels in the sampled image is determined, including:

According to the frequency ratio of each new gray level in the upsampled image and the ratio of each new gray level to the frequency of all new gray levels in the upsampled image, the cumulative distribution of each new gray level is determined respectively The value after function mapping, wherein the larger the ratio of the frequency of each new gray level to all new gray levels in the upsampled image, the smaller the value after mapping the cumulative distribution function corresponding to each new gray level;

According to the mapped value of the cumulative distribution function of each newborn gray level, the mapped gray level of each newborn gray level is respectively determined.
A method for detecting the integrity of a switch interface according to claim 1, wherein the enhanced image is template-matched using a template corresponding to a standard network interface, and judging whether there is an abnormal network interface in the enhanced image according to the matching result, including:

Obtaining the similarity between the template corresponding to the standard network interface and the region to be tested in the enhanced image, where the region to be tested is a region of the same size as the template in the enhanced image;

When the similarity is greater than the preset first threshold, it is determined that the area to be tested is the area where the normal network interface is located;

In the case that the similarity is not greater than the preset first threshold, it is judged whether the similarity is greater than the preset second threshold, and if the judgment result is yes, the area to be tested is determined to be the area where the abnormal interface is located, wherein the preset The second threshold is smaller than the preset first threshold.
A method for detecting integrity of a switch interface according to claim 3, wherein the similarity is structural similarity SSIM.
A method for detecting integrity of a switch interface according to claim 3, wherein the similarity is a Pearson correlation coefficient.
The method for detecting the integrity of a switch interface according to claim 1, wherein before collecting the front image of the switch including the network interface, further comprising: setting parallel light to illuminate the switch including the network interface.
A method for detecting the integrity of a switch interface according to claim 1, wherein after performing histogram equalization on the upsampled image to obtain the enhanced image, further comprising: performing edge detection on the enhanced image, and using the edge detection result as a new enhanced image.
The method for detecting integrity of a switch interface according to claim 7, wherein the operator used for edge detection is any one of Roberts operator, Prewitt operator, Sobel operator and Canny operator.
The method for detecting the integrity of a switch interface according to claim 1, further comprising: locating the abnormal network interface if there is an abnormal network interface.