ADSS optical cable defect automatic identification method and device

Abstract

The invention relates to the technical field of image processing, in particular to an ADSS optical cable defect automatic identification method and device, and the method comprises the steps: obtaining a to-be-detected optical cable image, carrying out the feature point detection, and according to the difference between each feature point and the gray value of a pixel point in a local range, combining with the discrete features of the gray value of the pixel point in the local range, carrying out the recognition of an ADSS optical cable defect. Obtaining a first defect characteristic value; according to the change characteristics of the gray value between adjacent pixels in each row in the to-be-detected optical cable image, obtaining a gray fluctuation value, and analyzing the difference between the gray fluctuation values between each row and the adjacent row to obtain a defect judgment value; obtaining a second defect characteristic value by combining the gray scale run matrix of the to-be-detected optical cable image; and according to the two defect feature values, combining defect features in the historical optical cable image to perform defect identification on the to-be-detected optical cable image. According to the method, the threshold values suitable for different defect types are determined by analyzing the characteristics of the optical cable defects, so that the accuracy of optical cable defect identification is improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to an automatic identification method and device for ADSS optical cable defects. Background Art

[0002] With the advancement of science and technology, optical fiber communication technology has become one of the main pillars of modern communication and plays a vital role in modern communication networks. However, during the production or use of optical fiber cables, various factors may cause different types of defects on the surface of the optical cable, which may affect the life of the product. In order to ensure the quality of optical fiber communication, it is often necessary to identify defects on the optical cable.

[0003] Defect extraction methods based on threshold segmentation are often widely used in defect identification because of their simple calculation and high computational efficiency. However, there are many types of optical cable defects, such as pinholes, necking and other defects. Fixed thresholds are difficult to adapt to these changes, making the fixed thresholds of conventional threshold segmentation methods inaccurate in distinguishing different types of defects, thus affecting the accuracy of the overall identification results. Summary of the invention

[0004] In view of the above, it is necessary to provide an automatic identification method and device for ADSS optical cable defects to solve the above problems.

[0005] The first aspect of the present application provides an automatic identification method for ADSS optical cable defects, the method comprising:

[0006] Acquire an image of the optical cable to be detected, wherein the optical cable to be detected is placed horizontally, and the axis thereof is parallel to the upper and lower edge lines of the image of the optical cable to be detected;

[0007] Perform feature point detection on the optical cable image to be inspected, and obtain the first defect feature value of each feature point based on the difference between the grayscale value of each feature point and the pixel point in the local range and the discrete features of the grayscale value of the pixel point in the local range;

[0008] According to the variation characteristics of the grayscale values ​​between adjacent pixels in each row of the optical cable image to be detected, the grayscale fluctuation value of each row is obtained; according to the difference between the grayscale fluctuation values ​​of each row and the adjacent rows, the defect judgment value of each row is obtained; according to the grayscale fluctuation value and the defect judgment value of the row where each feature point is located, combined with the grayscale run matrix of the optical cable image to be detected, the second defect feature value of each feature point is obtained;

[0009] According to the first defect feature value and the second defect feature value, combined with the defect features in the historical optical cable image, defect thresholds corresponding to various defect feature values ​​are obtained, and defects are identified on the optical cable image to be inspected.

[0010] The first defect characteristic value of each characteristic point is obtained as follows:

[0011] According to the overall grayscale difference between each feature point and the pixels in its neighborhood, combined with the discrete degree of the grayscale values ​​of the pixels in the row where each feature point is located, the first defect discrimination value of each feature point is obtained;

[0012] Among them, the first defect discrimination value is positively correlated with the difference and the discrete degree; the first defect discrimination value is negatively correlated with the gray value of the corresponding feature point.

[0013] The grayscale fluctuation value of each row is obtained as follows:

[0014] For each row of the optical cable image to be detected, a first-order difference sequence of each row is obtained based on the grayscale value, the difference between the number of positive numbers and the number of negative numbers in the first-order difference sequence is counted, and the grayscale fluctuation value of each row is obtained in combination with the number of non-zero elements in the first-order difference sequence.

[0015] The defect judgment value of each row is obtained as follows:

[0016] According to the fusion result of the difference in grayscale fluctuation value between each row of the optical cable image to be detected and the adjacent rows, the defect judgment value of each row is obtained; wherein the defect judgment value is negatively correlated with the fusion result.

[0017] The process of obtaining the second defect characteristic value of each characteristic point includes:

[0018] Obtain the direction value of each feature point in the feature point detection process; obtain the contrast value of each feature point according to the position characteristics of the direction value of each feature point in the vertical direction;

[0019] Scan the optical cable image to be detected in a vertical direction to obtain a grayscale run-length matrix of the optical cable image to be detected;

[0020] The second defect characteristic value of the i-th characteristic point is recorded as H i , its formula form is: Among them, D i Indicates the defect judgment value of the row where the i-th feature point is located; Indicates the number of non-zero values ​​in the row corresponding to the gray level of the i-th feature point in the gray run matrix; Δ i Represents the contrast value of the i-th feature point.

[0021] The specific process of obtaining the contrast value of each feature point is as follows:

[0022] Two feature points with the same ordinate as each feature point and the largest direction value are obtained, and the absolute value of the difference between the abscissas of the two feature points in the optical cable image to be detected is used as the comparison value of each feature point.

[0023] The step of obtaining the defect thresholds corresponding to various defect characteristic values ​​is as follows:

[0024] Obtain a preset number of historical optical cable defect images;

[0025] According to the distribution of various defect feature values ​​in each historical optical cable defect image, a segmentation threshold value corresponding to the defect feature value of each historical optical cable defect image is obtained;

[0026] According to the average level of the segmentation thresholds of various defect feature values ​​of all optical cable defect images, defect thresholds of various defect feature values ​​of the optical cable defect image to be detected are obtained; wherein the defect thresholds include a first defect threshold and a second defect threshold.

[0027] The segmentation threshold for obtaining the defect feature value corresponding to each historical optical cable defect image is specifically:

[0028] For any defect feature value, threshold segmentation is performed on the any defect feature value of all feature points of each optical cable defect image to obtain a segmentation threshold of the any defect feature value of each optical cable defect image.

[0029] The process of identifying defects on the optical cable image to be inspected includes:

[0030] When the mean of all the first defect feature values ​​in the optical cable image to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is greater than the second defect threshold, the defect recognition result of the optical cable to be detected is that a pinhole defect and a necking defect exist;

[0031] When the mean of all the first defect feature values ​​in the optical cable image to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is less than or equal to the second defect threshold, the defect recognition result of the optical cable to be detected is a pinhole defect;

[0032] When the mean value of all the first defect feature values ​​in the optical cable image to be detected is less than or equal to the first defect threshold, and the mean value of all the second defect feature values ​​is greater than the second defect threshold, the defect identification result of the optical cable to be detected is a necking defect; otherwise, the defect identification result of the optical cable to be detected is that these two defects do not exist.

[0033] In a second aspect, an embodiment of the present application also provides an ADSS optical cable defect automatic identification device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0034] In the above scheme, the image of the optical cable to be detected is first obtained, and feature point detection is performed on it. According to the grayscale difference between the feature point and the nearby pixel points, and the discrete characteristics of the grayscale values ​​of the nearby pixel points, a first feature defect value is constructed to characterize the characteristics of the small hole defect in the optical cable image. Its beneficial effect is that for each feature point in the optical cable image to be detected, the possibility of it falling into the small hole defect area is analyzed, thereby improving the accuracy of corresponding defect recognition; further, the grayscale change analysis is performed on each row in the optical cable image to be detected, and the difference in grayscale changes between adjacent rows is performed to obtain the grayscale fluctuation value and defect judgment value of each row. Its beneficial effect is that through the grayscale fluctuation value, it is judged whether there is a necking defect in the optical cable image to be detected based on the change of texture features, and further through the defect judgment value The texture features of each row are compared. If the texture fluctuation features are obvious and have similar fluctuation features, it can be preliminarily determined that there is a defect at the position corresponding to the feature point. The grayscale fluctuation value and the defect judgment value are combined with the grayscale run matrix to construct the second defect feature value. The beneficial effect is that the texture features and shape features of the necking defect in the optical cable are comprehensively considered to more accurately describe the defect features, so that the identification of defects in the optical cable to be detected is more accurate. The two defect features of the historical optical cable defect image are analyzed. The beneficial effect is that different defect thresholds are obtained for different defects, which solves the problem that the same threshold cannot effectively distinguish these different types of defects, so that the defect identification result of the optical cable image to be detected is more accurate and the efficiency of defect identification is higher. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 A flowchart of the steps of an automatic identification method for ADSS optical cable defects provided by one embodiment of the present application;

[0036] Figure 2 A flowchart for obtaining a defect threshold value is provided for one embodiment of the present application. DETAILED DESCRIPTION

[0037] In the description of the embodiments of the present application, words such as "exemplary", "or", "for example" and the like are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "or", "for example" and the like is intended to present related concepts in a concrete manner.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art in the present application. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0039] It should also be noted that the terms "first" and "second" in this application and the accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The method disclosed in the embodiments of the present application or the method shown in the flow chart includes one or more steps for implementing the method. Without departing from the scope of protection of the present application, the execution order of multiple steps can be interchanged with each other, and some steps can also be deleted.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0041] The specific scheme of the ADSS optical cable defect automatic identification method and device provided by the present application is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of a method for automatically identifying defects in an ADSS optical cable provided by an embodiment of the present application, the method comprising the following steps:

[0043] The first step is to obtain an image of the optical cable to be detected, wherein the optical cable to be detected is placed horizontally, and the axis thereof is parallel to the upper and lower edge lines of the image of the optical cable to be detected.

[0044] A CCD camera is used to capture an ADSS optical cable image, and the acquired ADSS optical cable image is rectangular. During the acquisition process, the optical cable is kept horizontal in the image, and the central axis of the optical cable is parallel to the upper and lower edges of the ADSS optical cable image. In this embodiment, the ADSS optical cable image is converted into grayscale, an adaptive histogram equalization algorithm is used to enhance the image, and a median filtering algorithm is used to denoise the image to obtain an optical cable image to be detected; wherein, the adaptive histogram equalization algorithm and the median filtering algorithm are both existing publicly known technologies, and this application will not elaborate on them; in addition, the implementer may choose other methods for image processing, and this application does not limit this.

[0045] The second step is to perform feature point detection on the optical cable image to be inspected, and obtain the first defect feature value of each feature point based on the difference in grayscale value between each feature point and the pixel point in the local range and the discrete features of the grayscale value of the pixel point in the local range.

[0046] In the process of defect detection of optical cable images, the traditional threshold segmentation method relies on comparing the segmentation threshold with the grayscale value of the pixel to identify the defect area. However, the grayscale value alone is not enough to accurately distinguish the defect area from the non-defect area. This is because the grayscale value in the optical cable image may be affected by many factors, resulting in similar grayscale performance of the defect and normal areas, which makes the defect recognition result of the optical cable image inaccurate. Based on this, the feature point detection of the optical cable image to be detected is first performed, and the small hole defect features of the optical cable image are analyzed to obtain the first defect feature value of each feature point.

[0047] In this embodiment, the Speeded-Up Robust Features (SURF) algorithm is used to detect feature points of the optical cable image to obtain the direction assignment of each feature point; wherein, the SURF algorithm is an existing well-known technology, which is not elaborated in this application; in addition, the implementer can select a suitable feature point detection algorithm to obtain the feature points and their direction assignment, which is not limited in this application.

[0048] Among all the feature points obtained, there are both feature points of non-defective areas and feature points of defective areas, so these feature points need to be further distinguished to obtain feature points of defective areas. For small hole type defects, the color of the feature points at the small hole is darker, and the color around the small hole is relatively lighter, so the grayscale value of the small hole feature point is smaller, and the grayscale value of the small hole feature point will be different from the grayscale value of the pixel points around the small hole feature point; in addition, due to the influence of light, a reflective area will appear on the surface of the optical cable, while the small hole feature point will not reflect light. When the small hole is located in the reflective area, there will be a large grayscale difference in the pixel points in the row where the small hole feature point is located.

[0049] Taking each feature point as the center, the neighborhood of each feature point is obtained; according to the difference between the average level of grayscale values ​​of all pixels in the neighborhood of each feature point and the grayscale value of the corresponding feature point, combined with the discrete degree of grayscale values ​​of pixels in the row where each feature point is located, the first defect discrimination value of each feature point is obtained, wherein the first defect discrimination value is positively correlated with the difference and the discrete degree, and the first defect discrimination value is negatively correlated with the grayscale value of the corresponding feature point.

[0050] It should be noted that the difference between two variables can be obtained by forward mapping of subtraction or division; the discreteness between multiple variables can be obtained by calculating the extreme value, variance, etc. As an embodiment, the first defect discrimination value of the i-th feature point is recorded as K i , its formula form is: in, represents the mean gray value of all pixels except the central pixel in the 11×11 neighborhood centered on the i-th feature point; Qi Represents the gray value of the i-th feature point; d i Represents a sequence of grayscale values ​​of all pixels in the row where the i-th feature point is located; R() represents the extreme value function; γ is a parameter adjustment factor preset to be greater than zero to prevent the denominator from being 0, and its value is 0.01; implementers can adjust it at their own discretion.

[0051] It should be understood that when When the value of is large, it means that the gray value of the i-th feature point is significantly different from the gray mean of the pixels in its surrounding 11×11 neighborhood. i ) is large, indicating that the grayscale values ​​of all pixels in the row where the i-th feature point is located are more discrete. i When the value of is small, it means that the gray value of the i-th feature point is small. In summary, the possibility that the i-th feature point has a small hole defect feature is greater.

[0052] The third step: according to the change characteristics of the grayscale values ​​between adjacent pixels in each row of the optical cable image to be detected, the grayscale fluctuation value of each row is obtained; according to the difference between the grayscale fluctuation values ​​between each row and the adjacent rows, the defect judgment value of each row is obtained; according to the grayscale fluctuation value and the defect judgment value of the row where each feature point is located, combined with the grayscale run matrix of the optical cable image to be detected, the second defect feature value of each feature point is obtained.

[0053] For the cable necking type defect, the cable diameter at the necking location is smaller than that at the non-necking location. In this case, the distance between the two pixels at the junction of the cable and the background in the column where the necking feature point is located is closer. The surface color of the cable at the necking location is in the form of black and white vertical stripes. In this case, there will be more data fluctuations of similar frequencies in the sequence composed of the grayscale values ​​of each row of pixels in the row where the necking feature point is located and in the adjacent rows. Based on this, the grayscale fluctuation values ​​of each row of the image to be detected are obtained:

[0054] For each row of the optical cable image to be detected, a first-order difference sequence of each row is obtained based on the grayscale value, the difference between the number of positive numbers and the number of negative numbers in the first-order difference sequence is counted, and the grayscale fluctuation value of each row is obtained by combining the total number of positive numbers and negative numbers in the first-order difference sequence.

[0055] It should be noted that the statistical difference between the number of positive numbers and the number of negative numbers in the first-order difference sequence refers to the absolute value of the difference between the number of positive numbers and the number of negative numbers in the first-order difference sequence; the grayscale fluctuation value of each row is specifically the ratio of the total number to the absolute value of the difference; wherein, in order to avoid the situation where the denominator is zero, it is necessary to add a preset value greater than zero to the denominator, and the value is 0.01.

[0056] Furthermore, the defect judgment value of each row is obtained:

[0057] According to the fusion result of the difference of the grayscale fluctuation value between each row of the optical cable image to be detected and the adjacent rows, the defect judgment value of each row is obtained.

[0058] As an embodiment, taking the i-th row as an example, the absolute value of the difference between the grayscale fluctuation values ​​of the x-th row and the x-1-th row and the x+1-th row is calculated respectively; the reciprocal of the sum of the two absolute values ​​of the difference is used as the defect judgment value of each row; in order to avoid the situation where the denominator is zero, a preset value greater than zero is added to the denominator, and the value is 0.01. For the first and last rows of the image, only the adjacent rows where they exist are calculated.

[0059] It should be understood that when a necking feature appears in an optical cable, multiple values ​​greater than 0 and less than 0 will appear in the first-order difference sequence of the sequence composed of the grayscale values ​​of each row of pixels in the corresponding row and adjacent rows where the necking feature appears, and the values ​​greater than 0 and less than 0 will appear alternately.

[0060] There may be fine vertical stripes at the necking of the optical cable, and the lengths of these stripes may be different. The optical cable image is scanned in the vertical direction to obtain the grayscale run-length matrix of the optical cable image to be detected. Several runs of different lengths will appear in the grayscale level where the feature points located in the necking defect area are located. Among them, the calculation of the grayscale run-length matrix is ​​a well-known technology, and the specific process will not be repeated.

[0061] As an embodiment, taking the i-th feature point as an example, the second defect feature value of the i-th feature point is recorded as H i , its formula form is: Among them, D i Indicates the defect judgment value of the row where the i-th feature point is located; Indicates the number of non-zero values ​​in the row corresponding to the gray level of the i-th feature point in the gray run matrix; Δ i Represents the contrast value of the i-th feature point.

[0062] Among them, two feature points with the same ordinate as the i-th feature point and the largest direction value are obtained; the absolute value of the difference between the abscissas of the two feature points in the optical cable image to be detected is used as the comparison value of each feature point.

[0063] It should be understood that when D i When the value of is large, it means that the gray value sequence of all pixels in the row where the i-th feature point is located and the adjacent rows has more frequent data fluctuations; When the value of is large, it means that the gray level of the i-th feature point has multiple runs of different lengths; when Δ i When the value of is small, it means that the optical cable diameter at the column where the i-th feature point is located is small, and the possibility that the i-th feature point is in the necking defect area is greater.

[0064] The fourth step: according to the first defect feature value and the second defect feature value, combined with the defect features in the historical optical cable image, the defect thresholds corresponding to various defect feature values ​​are obtained, and the defect recognition is performed on the optical cable image to be detected.

[0065] A preset number of historical optical cable defect images are obtained, wherein the historical optical cable defect images contain at least one of a pinhole defect and a necking defect; for any defect feature value, threshold segmentation is performed on the any defect feature value of all feature points of each optical cable defect image to obtain a segmentation threshold of the any defect feature value of each optical cable defect image; a defect threshold of the any defect feature value of the optical cable defect image to be detected is obtained according to an average level of the segmentation thresholds of all the any defect feature values ​​of all optical cable defect images; the defect threshold includes a first defect threshold and a second defect threshold.

[0066] As an embodiment, the preset number is 600, and the implementer can adjust it according to the actual situation; the method for threshold segmentation is Otsu threshold segmentation; the average level of multiple variables can be measured by the average value of multiple variables, that is, the first defect threshold of the optical cable image to be detected is specifically the average value of the first defect feature values ​​corresponding to the segmentation threshold of all optical cable defect images; the second defect threshold of the optical cable image to be detected is specifically the average value of the second defect feature values ​​corresponding to the segmentation threshold of all optical cable defect images.

[0067] Among them, the defect threshold acquisition flow chart is as follows: Figure 2 shown.

[0068] When the mean of all the first defect feature values ​​in the image of the optical cable to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is greater than the second defect threshold, the defect recognition result of the optical cable to be detected is the presence of a pinhole defect and a necking defect; when the mean of all the first defect feature values ​​in the image of the optical cable to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is less than or equal to the second defect threshold, the defect recognition result of the optical cable to be detected is a pinhole defect; when the mean of all the first defect feature values ​​in the image of the optical cable to be detected is less than or equal to the first defect threshold, and the mean of all the second defect feature values ​​is greater than the second defect threshold, the defect recognition result of the optical cable to be detected is a necking defect; otherwise, the defect recognition result of the optical cable to be detected is the absence of these two defects.

[0069] Based on the same inventive concept as the above method, an embodiment of the present application also provides an ADSS optical cable defect automatic identification device, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned ADSS optical cable defect automatic identification methods are implemented.

[0070] In summary, the embodiment of the present application first obtains the image of the optical cable to be detected, and performs feature point detection on it. According to the grayscale difference between the feature point and the nearby pixel points, and the discrete characteristics of the grayscale values ​​of the nearby pixel points, a first feature defect value is constructed to characterize the characteristics of the small hole defect in the optical cable image. The beneficial effect is that for each feature point in the optical cable image to be detected, the possibility of it falling into the small hole defect area is analyzed, thereby improving the accuracy of corresponding defect recognition; further, the grayscale change analysis is performed on each row in the optical cable image to be detected, and the difference in grayscale changes between adjacent rows is performed to obtain the grayscale fluctuation value and the defect judgment value of each row. The beneficial effect is that through the grayscale fluctuation value, it is judged whether there is a necking defect in the optical cable image to be detected based on the change of texture features, and further through the defect judgment The texture features of each row are compared by the broken value. If the texture fluctuation feature is obvious and has similar fluctuation features, it can be preliminarily determined that there is a defect at the position corresponding to the feature point; the grayscale fluctuation value and the defect judgment value are combined with the grayscale run matrix to construct the second defect feature value. Its beneficial effect is that it comprehensively considers the texture features and shape features of the necking defect in the optical cable, and more accurately describes the defect features, so that the identification of defects in the optical cable to be detected is more accurate; the two defect features of the historical optical cable defect image are analyzed, and its beneficial effect is that different defect thresholds are obtained for different defects, which solves the problem that the same threshold cannot effectively distinguish these different types of defects, so that the defect identification result of the optical cable image to be detected is more accurate and the efficiency of defect identification is higher.

[0071] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two continuous operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

[0072] For those skilled in the art, it is obvious that the present application is not limited to the details of the above exemplary embodiments, and the present application can be implemented in other specific forms without departing from the basic features of the present application. Therefore, no matter from which point of view, the above embodiments of the present application should be regarded as exemplary and non-restrictive; the technical solutions recorded in the above embodiments are modified, or some of the technical features are replaced by equivalents, which does not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for automatic identification of ADSS optical cable defects. It is characterized in that The method comprises the following steps: Acquire an image of the optical cable to be detected, wherein the optical cable to be detected is placed horizontally, and the axis thereof is parallel to the upper and lower edge lines of the image of the optical cable to be detected; Perform feature point detection on the optical cable image to be inspected, and obtain the first defect feature value of each feature point based on the difference between the grayscale value of each feature point and the pixel point in the local range and the discrete features of the grayscale value of the pixel point in the local range; According to the variation characteristics of the grayscale values ​​between adjacent pixels in each row of the optical cable image to be detected, the grayscale fluctuation value of each row is obtained; according to the difference between the grayscale fluctuation values ​​of each row and the adjacent rows, the defect judgment value of each row is obtained; according to the grayscale fluctuation value and the defect judgment value of the row where each feature point is located, combined with the grayscale run matrix of the optical cable image to be detected, the second defect feature value of each feature point is obtained; According to the first defect feature value and the second defect feature value, combined with the defect features in the historical optical cable image, defect thresholds corresponding to various defect feature values ​​are obtained, and defects are identified on the optical cable image to be inspected.

2. The ADSS optical cable defect automatic identification method according to claim 1, It is characterized in that The first defect characteristic value of each characteristic point is obtained as follows: According to the overall grayscale difference between each feature point and the pixels in its neighborhood, combined with the discrete degree of the grayscale values ​​of the pixels in the row where each feature point is located, the first defect discrimination value of each feature point is obtained; Among them, the first defect discrimination value is positively correlated with the difference and the discrete degree; the first defect discrimination value is negatively correlated with the gray value of the corresponding feature point.

3. The ADSS optical cable defect automatic identification method according to claim 1, It is characterized in that The grayscale fluctuation value of each row is obtained as follows: For each row of the optical cable image to be detected, a first-order difference sequence of each row is obtained based on the grayscale value, the difference between the number of positive numbers and the number of negative numbers in the first-order difference sequence is counted, and the grayscale fluctuation value of each row is obtained in combination with the number of non-zero elements in the first-order difference sequence.

4. The ADSS optical cable defect automatic identification method according to claim 1, It is characterized in that The defect judgment value of each row is obtained as follows: According to the fusion result of the difference in grayscale fluctuation value between each row of the optical cable image to be detected and the adjacent rows, the defect judgment value of each row is obtained; wherein the defect judgment value is negatively correlated with the fusion result.

5. The ADSS optical cable defect automatic identification method according to claim 1, It is characterized in that The process of obtaining the second defect characteristic value of each characteristic point includes: Obtain the direction value of each feature point in the feature point detection process; obtain the contrast value of each feature point according to the position characteristics of the direction value of each feature point in the vertical direction; Scan the optical cable image to be detected in a vertical direction to obtain a grayscale run-length matrix of the optical cable image to be detected; The second defect characteristic value of the i-th characteristic point is recorded as H i , its formula form is: Among them, D i Indicates the defect judgment value of the row where the i-th feature point is located; Indicates the number of non-zero values ​​in the row corresponding to the gray level of the i-th feature point in the gray run matrix; Δ i Represents the contrast value of the i-th feature point.

6. The ADSS optical cable defect automatic identification method according to claim 5, It is characterized in that The specific process of obtaining the contrast value of each feature point is as follows: Two feature points with the same ordinate as each feature point and the largest direction value are obtained, and the absolute value of the difference between the abscissas of the two feature points in the optical cable image to be detected is used as the comparison value of each feature point.

7. The ADSS optical cable defect automatic identification method according to claim 1, It is characterized in that The steps of obtaining the defect thresholds corresponding to various defect characteristic values ​​are: Obtain a preset number of historical optical cable defect images; According to the distribution of various defect feature values ​​in each historical optical cable defect image, a segmentation threshold value corresponding to the defect feature value of each historical optical cable defect image is obtained; According to the average level of the segmentation thresholds of various defect feature values ​​of all optical cable defect images, defect thresholds of various defect feature values ​​of the optical cable defect image to be detected are obtained; wherein the defect thresholds include a first defect threshold and a second defect threshold.

8. The ADSS optical cable defect automatic identification method according to claim 7, It is characterized in that The segmentation threshold of the defect feature value corresponding to each historical optical cable defect image is obtained as follows: For any defect feature value, threshold segmentation is performed on the any defect feature value of all feature points of each optical cable defect image to obtain a segmentation threshold of the any defect feature value of each optical cable defect image.

9. The ADSS optical cable defect automatic identification method according to claim 7, It is characterized in that The process of performing defect recognition on the optical cable image to be inspected comprises: When the mean of all the first defect feature values ​​in the optical cable image to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is greater than the second defect threshold, the defect recognition result of the optical cable to be detected is that a pinhole defect and a necking defect exist; When the mean of all the first defect feature values ​​in the optical cable image to be detected is greater than the first defect threshold, and the mean of all the second defect feature values ​​is less than or equal to the second defect threshold, the defect recognition result of the optical cable to be detected is a pinhole defect; When the mean value of all the first defect feature values ​​in the optical cable image to be detected is less than or equal to the first defect threshold, and the mean value of all the second defect feature values ​​is greater than the second defect threshold, the defect identification result of the optical cable to be detected is a necking defect; otherwise, the defect identification result of the optical cable to be detected is that these two defects do not exist.

10. An ADSS optical cable defect automatic identification device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images