An optical cable fault positioning method and system based on image accurate identification

Abstract

The application discloses a kind of cable fault location method and system based on image accurate identification, it is related to image processing technical field, comprising: establishing fault database, including several fault feature classes in fault database, each fault feature class includes several same type fault features;Obtain optical cable image;According to image splitting algorithm, optical cable image is carried out image splitting fusion, obtains several fusion feature sets;According to matching degree algorithm, the fault matching degree of each fusion feature set is calculated respectively;The fusion feature set corresponding to the maximum value of fault matching degree is screened out, and the fusion feature set is regarded as image region division feature group;Based on the optical cable image region of each fusion feature corresponding in image region division feature group, fault region division is carried out in optical cable image.Guarantee the high precision matching between fault identification and feature of optical cable image, improve the precision of optical cable image fault identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, in particular to a cable fault positioning method and system based on image accurate identification. BACKGROUND

[0002] Optical cable lines play a vital role in modern communication networks, but due to their wide distribution, complex environment, and susceptibility to natural disasters, human damage and other factors, faults occur frequently. Traditional optical cable fault positioning methods rely on manual inspection and optical cable testing equipment, and have problems such as inaccurate positioning and low efficiency. With the rapid development of artificial intelligence and image recognition technology, its application in optical cable fault positioning has become a new trend.

[0003] In the existing optical cable fault positioning and identification system, the depth of image feature analysis is insufficient, and it is difficult to mine the internal correlation degree of multiple fault features, which is prone to error feature recognition matching when positioning optical cable fault features. SUMMARY

[0004] To solve the above technical problems, a cable fault positioning method and system based on image accurate identification are provided, which solves the problem of insufficient depth of image feature analysis in the existing optical cable fault positioning and identification system, difficulty in mining the internal correlation degree of multiple fault features, and error feature recognition matching when generating optical cable fault feature positioning.

[0005] To achieve the above purpose, the technical scheme adopted by the present application is:

[0006] A cable fault positioning method based on image accurate identification, comprising:

[0007] Establishing a fault database;

[0008] Obtaining an optical cable image;

[0009] According to an image splitting algorithm, the optical cable image is split and fused to obtain a plurality of fused feature sets;

[0010] According to a matching degree algorithm, the fault matching degree of each fused feature set is calculated respectively;

[0011] The fused feature set corresponding to the maximum fault matching degree is selected, and the fused feature set is taken as an image region division feature group;

[0012] Based on the optical cable image region corresponding to each fused feature in the image region division feature group, the fault region in the optical cable image is divided.

[0013] Preferably, the image splitting algorithm specifically includes:

[0014] acquiring a total number M of same fault features in the optical cable image;

[0015] sequentially traversing positive integers in a 0-M interval as a feature fusion number n;

[0016] generating a feature fusion restriction condition based on the total number M of same fault features in the optical cable image and the feature fusion number n;

[0017] generating a plurality of feature fusion sets under the restriction of the feature fusion restriction condition;

[0018] extracting a plurality of same fault features with the same boundary from the optical cable image according to numbers in the feature fusion set, and collecting the same fault features into a fusion feature set;

[0019] wherein an expression of the feature fusion restriction condition is:

[0020]

[0021] wherein z i is a number of same fault features included in the i-th fusion feature, and N+ is a set of positive integers.

[0022] Preferably, the acquiring of the total number M of same fault features in the optical cable image specifically comprises:

[0023] performing gray scale processing on the original optical cable image by using a weighted average method to obtain an optical cable gray scale image;

[0024] setting a gray scale similarity deviation threshold, and dividing a continuous interval with a gray scale deviation less than the gray scale similarity deviation threshold in the optical cable gray scale image into same fault features;

[0025] counting a total number of same fault features in the optical cable gray scale image, and denoted as M;

[0026] the weighted average formula is

[0027] Gray(i,j) = α1R(i,j) + α2G(i,j) + α3B(i,j),

[0028] wherein Gray(i,j) is a gray scale of an i-th row and j-th column pixel point in the original optical cable image, α1, α2, and α3 are red, green, and blue three-color sensitivity weighting coefficients respectively, and R(i,j), G(i,j), and B(i,j) are red, green, and blue three-color components of the i-th row and j-th column pixel point respectively.

[0029] Preferably, the extracting of the plurality of same fault features with the same boundary from the optical cable image according to the numbers in the feature fusion set, and the collecting of the same fault features into the fusion feature set specifically comprises:

[0030] S1, selecting an element from the feature fusion set, and denoted as Z;

[0031] S2, selecting one same fault feature from the optical cable image as a reference fault feature;

[0032] S3, setting a fusion index j, and adding an initial value of 1 to j;

[0033] S4, judging whether j is equal to Z, if yes, taking the reference fault feature as a fusion feature, and performing step S6, if no, increasing the value of j by 1, and performing step S5;

[0034] S5, selecting a same fault feature having the same boundary as the reference fault feature, fusing the reference fault feature and the same fault feature to form a new reference fault feature, and returning to step S4;

[0035] S6, judging whether all elements in the feature fusion set are traversed, if yes, performing S7, if no, removing Z from the feature fusion set, and returning to step S1;

[0036] S7, combining all fusion features into a fusion feature set;

[0037] S8, repeating steps S1-S7 until a new fusion feature set cannot be generated under the feature fusion set.

[0038] Preferably, the matching degree algorithm is specifically:

[0039] calculating the similarity of each fusion feature in the fusion feature set and each fault feature in the fault database, and screening out a fault feature with the largest similarity as a target feature of the fusion feature;

[0040] performing feature correlation degree calculation based on the target feature of each fusion feature in the fusion feature set, and obtaining the feature correlation degree of the fusion feature set;

[0041] calculating the fault matching degree of the fusion feature set based on the similarity of each fusion feature in the fusion feature set and the target feature thereof and the feature correlation degree of the fusion feature set through a matching degree calculation formula;

[0042] the matching degree calculation formula is specifically:

[0043]

[0044] In the formula, P is the fault matching degree of the fusion feature set, C hi is the similarity of the i-th fusion feature in the fusion feature set and the target feature thereof, and G l is the feature correlation degree of the fusion feature set.

[0045] Preferably, the feature correlation degree calculation is performed based on the target feature of each fusion feature in the fusion feature set, and the feature correlation degree of the fusion feature set is obtained, specifically including:

[0046] determining the fault feature class corresponding to the target feature of each fusion feature;

[0047] combining the fusion features in the fusion feature set two by two into several fusion feature groups, combining the fusion features with the same boundary into adjacent fusion feature groups, and combining the fusion features without the same boundary into interval fusion feature groups;

[0048] based on the set correlation logic, determining the correlation degree between the fault feature classes corresponding to the target features of the two fusion features of each fusion feature group, denoted as the correlation degree of the fusion feature group;

[0049] based on the correlation degrees of all fusion feature groups, calculating the feature correlation degree of the fusion feature set through a feature correlation degree synthesis formula;

[0050] wherein, the feature correlation degree synthesis formula is specifically:

[0051]

[0052] wherein, g k is the correlation degree of the kth adjacent fusion feature group, g u is the correlation degree of the uth interval fusion feature group, K is the total number of adjacent fusion feature groups, and U is the total number of interval fusion feature groups.

[0053] Further, an optical cable fault positioning system based on image accurate identification is proposed, which is used to realize the optical cable fault positioning method based on image accurate identification as described above, comprising:

[0054] a storage module, the storage module is used for storing a fault database;

[0055] an optical cable image acquisition module, the optical cable image acquisition module is used for acquiring optical cable images;

[0056] a feature splitting module, the feature splitting module is electrically connected with the optical cable image acquisition module, and the feature splitting module is used for image splitting fusion of the optical cable images according to an image splitting algorithm, to obtain several fusion feature sets;

[0057] a matching calculation module, the matching calculation module is electrically connected with the storage module and the feature splitting module, and the matching calculation module is used for calculating the fault matching degree of each fusion feature set according to a matching degree algorithm;

[0058] The screening module is electrically connected with the matching calculation module, and is used for screening a fusion feature set corresponding to the maximum fault matching degree, and taking the fusion feature set as an image region division feature group.

[0059] The region division module is electrically connected with the screening module, and is used for performing fault region division on the optical cable image based on the optical cable image region corresponding to each fusion feature in the image region division feature group.

[0060] Optionally, the feature splitting module comprises:

[0061] The feature extraction unit is used for acquiring the total number M of the same fault features in the optical cable image.

[0062] The fusion condition unit is used for generating a feature fusion restriction condition.

[0063] The feature fusion unit is used for generating a plurality of feature fusion sets under the restriction of the feature fusion restriction condition, extracting a plurality of same fault features with the same boundary from the optical cable image according to the numbers in the feature fusion sets, and collecting the same fault features into a fusion feature set.

[0064] The matching calculation module comprises:

[0065] The similarity fitting unit is used for calculating the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, screening a fault feature with the maximum similarity as a target feature of the fusion feature.

[0066] The association calculation unit is used for performing feature association degree calculation based on the target feature of each fusion feature in the fusion feature set, and acquiring the feature association degree of the fusion feature set.

[0067] The matching calculation unit is used for calculating the fault matching degree of the fusion feature set through a matching degree calculation formula based on the similarity between each fusion feature in the fusion feature set and the target feature thereof and the feature association degree of the fusion feature set.

[0068] Compared with the prior art, the present application has the following beneficial effects:

[0069] The application provides an optical cable fault positioning scheme based on image accurate identification, which splits the optical cable image into a plurality of fusion feature sets in detail, and combines the fusion feature set with the fault feature fitting of the fault database and the correlation degree between the fusion features in the fusion feature set for double identification and judgment, greatly improves the feature analysis depth of the optical cable image, can effectively guarantee the overall matching of the optical cable image when performing fault domain division and positioning, and further guarantees the high-precision matching between the optical cable image and the features in the fault identification and feature analysis, and improves the precision of optical cable image fault identification. BRIEF DESCRIPTION OF DRAWINGS

[0070] Figure 1 A flow chart of the optical cable fault positioning method based on image accurate identification provided by the application is shown.

[0071] Figure 2 A flow chart of the image splitting algorithm method in the application is shown.

[0072] Figure 3 A flow chart of the method for obtaining the same fault feature total number M in the optical cable image in the application is shown.

[0073] Figure 4 A flow chart of the method for obtaining the fusion feature set in the application is shown.

[0074] Figure 5 A flow chart of the matching degree algorithm method in the application is shown.

[0075] Figure 6 A flow chart of the method for obtaining the feature correlation degree of the fusion feature set in the application is shown.

[0076] Figure 7 An architecture schematic diagram of the electronic device provided by the application is shown.

[0077] Figure 8 A computer readable storage medium structure schematic diagram provided by the application is shown. DETAILED DESCRIPTION

[0078] The following description is used to disclose the application so that those skilled in the art can implement the application. The preferred embodiments in the following description are only as examples, and other obvious modifications can be thought of by those skilled in the art.

[0079] Referring to Figure 1 The optical cable fault positioning method based on image accurate identification includes:

[0080] Establishing a fault database;

[0081] Obtaining an optical cable image;

[0082] According to the image splitting algorithm, the optical cable image is image splitting fused to obtain a plurality of fusion feature sets;

[0083] According to the matching degree algorithm, the fault matching degree of each fusion feature set is calculated respectively;

[0084] The fusion feature set corresponding to the maximum fault matching degree is screened out, and the fusion feature set is taken as the image region division feature group;

[0085] Based on the optical cable image region corresponding to each fusion feature in the image region division feature group, the fault region division in the optical cable image is carried out.

[0086] By splitting the optical cable image into a plurality of fusion feature sets in detail, and combining the fusion feature set with the fault feature fitting of the fault database and the correlation degree between the fusion features in the fusion feature set, the depth of feature analysis of the optical cable image is greatly improved, and the overall matching of the optical cable image during fault recognition and positioning can be effectively guaranteed.

[0087] Referring to Figure 2 , the image splitting algorithm specifically includes:

[0088] Obtain the total number M of the same fault features in the optical cable image;

[0089] Iterate the positive integers in the 0-M interval as the feature fusion number n;

[0090] Based on the total number M of the same fault features in the optical cable image and the feature fusion number n, a feature fusion restriction condition is generated;

[0091] Under the restriction of the feature fusion restriction condition, a plurality of feature fusion sets are generated;

[0092] According to the numbers in the feature fusion array, a plurality of same fault features with the same boundary are extracted from the optical cable image and are collected as a fusion feature set;

[0093] Wherein, the expression of the feature fusion restriction condition is:

[0094]

[0095] In the formula, z i is the number of the same fault features included in the i-th fusion feature, and N+ is a set of positive integers.

[0096] Referring to Figure 3 , obtaining the total number M of the same fault features in the optical cable image specifically includes:

[0097] The original optical cable image is processed by using the weighted average method to obtain the optical cable gray image;

[0098] A gray scale similarity deviation threshold is set, and the continuous intervals with a gray scale deviation less than the gray scale similarity deviation threshold in the gray scale image of the optical cable are divided into the same fault features;

[0099] The total number of the same fault features in the gray scale image of the optical cable is counted and recorded as M;

[0100] The weighted average formula is

[0101] Gray(i,j)=α1R(i,j)+α2G(i,j)+α3B(i,j),

[0102] In the formula, Gray(i,j) is the gray scale of the pixel point in the i-th row and the j-th column in the original optical cable image, α1, α2, and α3 are the red, green, and blue three-color sensitivity weighting coefficients, respectively, and R(i,j), G(i,j), and B(i,j) are the red, green, and blue three-color components of the pixel point in the i-th row and the j-th column.

[0103] Referring to FIG. 1, Figure 4 As shown in the figure, a plurality of same fault features with the same boundary are extracted from the optical cable image according to the numbers in the feature fusion array, and are collected into a fusion feature set, which specifically includes:

[0104] S1, an element is selected from the feature fusion set and recorded as Z;

[0105] S2, a same fault feature is randomly selected from the optical cable image as a reference fault feature;

[0106] S3, a fusion index j is set, and an initial value of 1 is added to j;

[0107] S4, it is judged whether j is equal to Z, if yes, the reference fault feature is taken as a fusion feature, and step S6 is performed, if not, the value of j is increased by one, and step S5 is performed;

[0108] S5, a same fault feature with the same boundary as the reference fault feature is selected, and the reference fault feature is fused to form a new reference fault feature, and step S4 is returned;

[0109] S6, it is judged whether all elements in the feature fusion set are traversed, if yes, S7 is performed, if not, Z is removed from the feature fusion set, and step S1 is returned;

[0110] S7, all fusion features are combined into a fusion feature set;

[0111] S8, steps S1-S7 are repeated until no new fusion feature set can be generated under the feature fusion set.

[0112] The implementation process of the above image splitting algorithm is described in specific embodiments;

[0113] In some instances, by fusing the same fault features in the optical cable image, the number of identical features in the resulting optical cable image is 3, and the three identical features have the same boundary with each other. The three identical features are denoted as A, B, and C respectively.

[0114] When determining the number of feature fusions n, the values ​​of n are 1, 2, and 3, which represent extracting 1 fusion feature, 2 fusion features, and 3 fusion features from the optical cable image, respectively.

[0115] Taking the extraction of two fusion features as an example, based on the feature fusion constraints mentioned above, the generated feature fusion set is {1,2}, which means that in the generated fusion feature set, one fusion feature is composed of one identical feature, and one fusion feature is composed of two identical features. Based on the fact that the three identical features in the example have the same boundary with each other, the fusion feature sets corresponding to the feature fusion set {1,2} are {A,BC}, {B,AC}, and {C,AB}.

[0116] Reference Figure 5 As shown, the matching degree algorithm is as follows:

[0117] Calculate the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, and select the fault feature with the highest similarity as the target feature of the fusion feature.

[0118] Based on the target features of each fusion feature in the fusion feature set, the feature correlation degree is calculated to obtain the feature correlation degree of the fusion feature set.

[0119] Based on the similarity between each fused feature in the fused feature set and its target feature, and the feature correlation degree of the fused feature set, the fault matching degree of the fused feature set is calculated using the matching degree calculation formula.

[0120] The specific formula for calculating the matching degree is as follows:

[0121]

[0122] In the formula, P represents the fault matching degree of the fused feature set, and C... hi Let G be the similarity between the i-th fused feature in the fused feature set and its target feature. l To integrate the feature correlation of the feature set.

[0123] Reference Figure 6 As shown, based on the target features of each fused feature in the fused feature set, the feature correlation degree is calculated to obtain the feature correlation degree of the fused feature set. Specifically, this includes:

[0124] Determine the fault feature class corresponding to the target feature of each fused feature;

[0125] The fusion features in the fusion feature set are combined two by two to form a plurality of fusion feature groups, a fusion feature group formed by fusion features with the same boundary is recorded as an adjacent fusion feature group, and a fusion feature group formed by fusion features without the same boundary is recorded as a spaced fusion feature group;

[0126] Based on the set association logic, the association degree between the fault feature classes corresponding to the target feature pairs of the two fusion features of each fusion feature group is determined, which is recorded as the association degree of the fusion feature group;

[0127] Based on the association degrees of all fusion feature groups, the feature association degree of the fusion feature set is calculated through a feature association degree synthesis formula;

[0128] The feature association degree synthesis formula is specifically:

[0129]

[0130] In the formula, g k is the association degree of the kth adjacent fusion feature group, g u is the association degree of the uth spaced fusion feature group, K is the total number of adjacent fusion feature groups, and U is the total number of spaced fusion feature groups.

[0131] The fusion features of the adjacent fusion feature groups are close in distance, and the fusion features of the spaced fusion feature groups are far apart, so in the calculation of the feature association degree, the adjacent fusion feature groups are assigned a weight of 2, and the spaced fusion feature groups are assigned a weight of 1.

[0132] The association logic can be set in advance by the staff based on the mutual association between each fault feature class, for example, the bending angle is too large and the internal structure of the optical cable is damaged under pressure, which may cause the optical cable to break at the same time, so there is a greater association between the feature fault bending angle and the feature fault optical cable breakage. In turn, the mutual association between all fault feature classes is pre-set;

[0133] Through the feature fusion fault recognition of the optical cable image by the scheme, not only the similarity between the fusion features and the actual fault features is considered, but also the rationality of the simultaneous occurrence of each fusion feature fault in the same scene is comprehensively considered. Through comprehensive consideration, the depth of feature analysis of the optical cable image is greatly improved, which can effectively ensure the overall matching of the optical cable image when the feature region is divided, and further ensure the high-precision matching between the features of the optical cable image. Fault recognition.

[0134] Further, based on the same inventive concept as the above-mentioned image-based accurate optical cable fault positioning method, the scheme also proposes an image-based accurate optical cable fault positioning system, which comprises:

[0135] The storage module is used to store the fault database;

[0136] Optical cable image acquisition module, used to acquire optical cable images;

[0137] The feature splitting module is electrically connected to the optical cable image acquisition module. The feature splitting module is used to split and fuse the optical cable image according to the image splitting algorithm to obtain several fused feature sets.

[0138] The matching calculation module is electrically connected to the storage module and the feature splitting module. The matching calculation module is used to calculate the fault matching degree of each fused feature set according to the matching degree algorithm.

[0139] The filtering module is electrically connected to the matching calculation module. The filtering module is used to filter out the fusion feature set corresponding to the maximum fault matching degree, and use the fusion feature set as the image region division feature group.

[0140] The region segmentation module is electrically connected to the filtering module. The region segmentation module is used to segment the optical cable image into fault regions based on the optical cable image region corresponding to each fusion feature in the image region segmentation feature group.

[0141] The feature extraction unit is used to obtain the total number M of identical fault features in the optical cable image;

[0142] Fusion condition unit, split condition unit is used to generate feature fusion constraints;

[0143] The feature fusion unit generates several feature fusion sets under the constraints of feature fusion, and extracts several identical fault features with the same boundary from the optical cable image according to the numbers in the feature fusion array, and sets them into a fused feature set.

[0144] The matching calculation module includes:

[0145] The similarity fitting unit is used to calculate the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, and select the fault feature with the highest similarity as the target feature of the fusion feature.

[0146] The association calculation unit is used to calculate the feature association degree based on the target feature of each fusion feature in the fusion feature set, and obtain the feature association degree of the fusion feature set.

[0147] The matching calculation unit is configured to calculate the fault matching degree of the fusion feature set based on the similarity between each fusion feature in the fusion feature set and the target feature thereof and the feature correlation degree of the fusion feature set.

[0148] The working process of the system is as follows:

[0149] Step one: the optical cable image acquisition module acquires the optical cable image.

[0150] Step two: the feature extraction unit obtains the total number M of the same fault features in the optical cable image.

[0151] Step three: the split condition unit generates the feature fusion restriction condition.

[0152] Step four: the feature fusion unit generates a plurality of feature fusion sets under the restriction of the feature fusion restriction condition, extracts a plurality of same fault features with the same boundary from the optical cable image according to the numbers in the feature fusion sets, and sets the same fault features as a fusion feature set.

[0153] Step five: the similarity fitting unit calculates the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, and selects the fault feature with the maximum similarity as the target feature of the fusion feature.

[0154] Step six: the correlation calculation unit calculates the feature correlation degree of the fusion feature set based on the target feature of each fusion feature in the fusion feature set.

[0155] Step seven: the matching calculation unit calculates the fault matching degree of the fusion feature set based on the similarity between each fusion feature in the fusion feature set and the target feature thereof and the feature correlation degree of the fusion feature set.

[0156] Step eight: the screening module screens out the fusion feature set corresponding to the maximum fault matching degree, and sets the fusion feature set as the image region division feature group.

[0157] Step nine: the region division module performs fault region division on the optical cable image based on the optical cable image region corresponding to each fusion feature in the image region division feature group.

[0158] The method according to the embodiment of the application can also be implemented by means of the architecture of the electronic device shown in the figure. Figure 7 The electronic device shown in the figure can be implemented by means of the method according to the embodiment of the application. Figure 7As shown, the electronic device 500 can include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to a network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, can store an image-based accurate identification optical cable fault positioning method provided by the present application. The electronic device 500 can also include a user interface 508. Of course, Figure 7 The architecture shown is only exemplary, and in implementing different devices, according to actual needs, some of the components shown can be omitted Figure 7 One or more components of the electronic device shown.

[0159] Figure 8 The computer readable storage medium structure diagram provided by one example of the present application is shown. As shown in the figure, Figure 8 As shown, the computer readable storage medium 600 according to one embodiment of the present application. The computer readable storage medium 600 stores computer readable instructions. When the computer readable instructions are run by the processor, the image-based accurate identification optical cable fault positioning method according to the embodiments of the present application described with reference to the above figures can be executed. The storage medium 600 includes, but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may, for example, include random access memory (RAM) and cache memory, etc. Non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0160] In summary, the present application has the advantages of ensuring high accuracy matching between optical cable image fault identification and features, and improving the accuracy of optical cable image fault identification.

[0161] The basic principles, main features and advantages of the present application are shown and described above. Those skilled in the art should understand that the present application is not limited by the above examples, and the above examples and descriptions in the specification are only the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection claimed by the present application is defined by the appended claims and their equivalents.

Claims

1. A method for optical cable fault location based on precise image recognition, characterized in that, include: Establish a fault database; Acquire fiber optic cable images; The optical cable image is split and fused according to the image splitting algorithm to obtain several fusion feature sets; According to the matching degree algorithm, calculate the fault matching degree for each fusion feature set; Filter out the fusion feature set corresponding to the maximum fault matching degree, and use the fusion feature set as the image region segmentation feature group; Based on the optical cable image region corresponding to each fusion feature in the image region segmentation feature group, the fault region is segmented in the optical cable image. The matching degree algorithm is specifically as follows: Calculate the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, and select the fault feature with the highest similarity as the target feature of the fusion feature. Based on the target features of each fusion feature in the fusion feature set, the feature correlation degree is calculated to obtain the feature correlation degree of the fusion feature set. Based on the similarity between each fused feature in the fused feature set and its target feature, and the feature correlation degree of the fused feature set, the fault matching degree of the fused feature set is calculated using the matching degree calculation formula. The specific formula for calculating the matching degree is as follows: In the formula, P represents the fault matching degree of the fused feature set, and C... hi Let G be the similarity between the i-th fused feature in the fused feature set and its target feature. l To achieve feature correlation in the fusion feature set; Based on the target features of each fused feature in the fused feature set, the feature correlation degree is calculated. Specifically, the feature correlation degree of the fused feature set includes: Determine the fault feature class corresponding to the target feature of each fused feature; The fusion features in the fusion feature set are combined in pairs to form several fusion feature groups. The fusion feature groups composed of fusion features with the same boundary are called adjacent fusion feature groups, and the fusion feature groups composed of fusion features without the same boundary are called interval fusion feature groups. Based on the established association logic, the association degree between the fault feature classes corresponding to the target features of the two fusion features in each fusion feature group is determined and denoted as the association degree of the fusion feature group. Based on the correlation degree of all fused feature groups, the feature correlation degree of the fused feature set is calculated using the feature correlation degree comprehensive formula. The specific formula for the feature correlation degree is as follows: In the formula, g k Let g be the correlation degree of the k-th adjacent fused feature group. u Let represent the correlation degree of the u-th interval fusion feature group, K represent the total number of adjacent fusion feature groups, and U represent the total number of interval fusion feature groups.

2. The optical cable fault location method based on precise image recognition according to claim 1, characterized in that, The image segmentation algorithm specifically includes: Obtain the total number M of identical fault features in the optical cable image; The positive integers in the interval 0-M are sequentially traversed as the feature fusion number n; Based on the total number M of identical fault features in the optical cable image and the number of feature fusions n, feature fusion constraints are generated. Under the constraints of feature fusion, several feature fusion sets are generated; Based on the numbers in the feature fusion array, extract several identical fault features with the same boundaries from the optical cable image and combine them into a fusion feature set. The expression for the feature fusion constraint is as follows: In the formula, z i Let N be the number of identical fault features included in the i-th fused feature, and N+ be a set of positive integers.

3. The optical cable fault location method based on precise image recognition according to claim 2, characterized in that, The total number M of identical fault features obtained in the optical cable image specifically includes: The original optical cable image was processed into a grayscale image using a weighted average method to obtain the optical cable grayscale image. Set a gray-scale similarity deviation threshold, and divide the continuous intervals in the gray-scale image of the optical cable where the gray-scale deviation is less than the gray-scale similarity deviation threshold into the same fault features. The total number of identical fault features in the grayscale image of the optical cable is counted and denoted as M. The weighted average formula is as follows: Gray(i,j)=α1R(i,j)+α2G(i,j)+α3B(i,j), In the formula, Gray(i,j) is the gray level of the pixel in the i-th row and j-th column of the original optical cable image, α1, α2, and α3 are the red, green, and blue color sensitivity weighting coefficients, and R(i,j), G(i,j), and B(i,j) are the red, green, and blue color components of the pixel in the i-th row and j-th column, respectively.

4. The optical cable fault location method based on precise image recognition according to claim 3, characterized in that, The step of extracting several identical fault features with the same boundary from the optical cable image according to the numbers in the feature fusion array and combining them into a fusion feature set specifically includes: S1. Select an element from the feature fusion set, denoted as Z; S2. Select any identical fault feature from the optical cable image as the baseline fault feature; S3. Set a fusion index j and assign an initial value of 1 to j; S4. Determine if j is equal to Z. If yes, use the baseline fault feature as the fusion feature and proceed to step S6. If no, increment the value of j by one and proceed to step S5. S5. Select the same fault feature that has the same boundary as the baseline fault feature, merge it with the baseline fault feature to form a new baseline fault feature, and return to step S4. S6. Determine whether all elements in the feature fusion set have been traversed. If yes, proceed to S7. If no, remove Z from the feature fusion set and return to step S1. S7. Combine all fused features into a fused feature set; S8. Repeat steps S1-S7 until no new fused feature set can be generated under the feature fusion set.

5. A fiber optic cable fault location system based on precise image recognition, characterized in that, The method for optical cable fault location based on accurate image recognition as described in any one of claims 1-4 includes: Storage module, the storage module being used to store the fault database; Optical cable image acquisition module, the optical cable image acquisition module being used to acquire optical cable images; The feature splitting module is electrically connected to the optical cable image acquisition module. The feature splitting module is used to split and fuse the optical cable image according to the image splitting algorithm to obtain several fused feature sets. A matching calculation module is electrically connected to the storage module and the feature splitting module. The matching calculation module is used to calculate the fault matching degree of each fused feature set according to the matching degree algorithm. A filtering module is electrically connected to the matching calculation module. The filtering module is used to filter out the fusion feature set corresponding to the maximum fault matching degree, and use the fusion feature set as an image region division feature group. The region segmentation module is electrically connected to the filtering module. The region segmentation module is used to segment the optical cable image into fault regions based on the optical cable image region corresponding to each fusion feature in the image region segmentation feature group.

6. The optical cable fault location system based on precise image recognition according to claim 5, characterized in that, The feature splitting module includes: The feature extraction unit is used to obtain the total number M of identical fault features in the optical cable image; A fusion condition unit, wherein the splitting condition unit is used to generate feature fusion constraint conditions; The feature fusion unit generates several feature fusion sets under the constraints of feature fusion constraints, and extracts several identical fault features with the same boundary from the optical cable image according to the numbers in the feature fusion array, and sets them into a fused feature set. The matching calculation module includes: The similarity fitting unit is used to calculate the similarity between each fusion feature in the fusion feature set and each fault feature in the fault database, and to select the fault feature with the highest similarity as the target feature of the fusion feature. The association calculation unit is used to calculate the feature association degree based on the target feature of each fusion feature in the fusion feature set, and obtain the feature association degree of the fusion feature set. The matching calculation unit is used to calculate the fault matching degree of the fusion feature set based on the similarity between each fusion feature in the fusion feature set and its target feature, and the feature correlation degree of the fusion feature set, using the matching degree calculation formula.

7. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that may be executed by the at least one processor, the instructions being executed by at least one processor to enable the at least one processor to perform the optical cable fault location method based on precise image recognition as described in any one of claims 1-4.

8. A computer-readable storage medium storing a computer-readable program, characterized in that, When the computer program is executed by the processor, it implements the optical cable fault location method based on accurate image recognition as described in any one of claims 1-4.

Images