Cable loosening identification method based on data processing

Abstract

The present invention provides a cable loosening identification method based on data processing. The method responds to a loosening identification request, determines a comprehensive pre-inspection path according to the cable properties of a target cable, performs image acquisition on the upper surface of the target cable based on the comprehensive pre-inspection path, and obtains a plurality of pre-inspection acquisition images; performs multi-dimensional anomaly recognition on the pre-inspection acquisition image according to a loosening anomaly recognition strategy, and obtains an abnormal acquisition image corresponding to a corresponding upward-looking hovering point; sequentially counts downward-looking hovering points corresponding to the lower surface of the target cable based on the abnormal acquisition image, obtains an abnormal focusing path, and acquires an abnormal acquisition image corresponding to the abnormal focusing path; calculates the abnormality ratio in the abnormal acquisition image, obtains a loosening analysis value corresponding to the target cable, and determines a loosening identification result of the target cable based on the loosening analysis value.

Description

Technical Field

[0001] The invention relates to data processing technology, and in particular to a cable loosening identification method based on data processing. Background Art

[0002] In large-scale engineering structures, such as bridges, high-rise buildings, and utility pole installations, cables are key load-bearing and stabilizing components, and their safety and stability are of vital importance. However, as the use time increases, the cables may be damaged due to material aging, environmental corrosion, stress concentration, etc., which in turn threatens the safety of the overall structure. Therefore, timely and accurate identification of the working status of the cables is of great significance for preventing accidents and ensuring structural safety.

[0003] At present, traditional cable loosening detection methods mostly rely on manual inspections. This method is not only inefficient, but also easily affected by human factors, making it difficult to ensure the accuracy and consistency of the detection results.

[0004] Therefore, how to customize and generate corresponding detection paths according to the target cables that actually need to be detected, improve detection efficiency, and realize automatic and precise identification of cables has become an urgent problem to be solved. Summary of the invention

[0005] The present invention provides a method for identifying loose cables based on data processing. The present invention can generate corresponding detection paths according to the target cables that need to be detected, thereby improving detection efficiency and realizing automatic and accurate identification of cables.

[0006] A first aspect of the present invention provides a method for identifying loose cables based on data processing, comprising:

[0007] In response to the looseness identification request, a comprehensive pre-inspection path is determined according to the cable properties of the target cable, and an image of the upper surface of the target cable is captured based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection captured images;

[0008] Perform multi-dimensional anomaly recognition on the pre-checked captured image according to the loose anomaly recognition strategy to obtain an anomaly captured image corresponding to the corresponding upward-looking hovering point;

[0009] Based on the abnormal acquisition image, sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable to obtain an abnormal focusing path, and acquiring an abnormal acquisition image corresponding to the abnormal focusing path;

[0010] The abnormality ratio in the abnormal acquisition image is calculated to obtain a looseness analysis value corresponding to the target cable, and a looseness identification result of the target cable is determined based on the looseness analysis value.

[0011] Optionally, in a possible implementation of the first aspect, in response to the looseness identification request, a comprehensive pre-inspection path is determined according to the cable properties of the target cable, and an image is captured of the upper surface of the target cable based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection captured images, including:

[0012] In response to the looseness identification request, a cable collection area in a collection interface corresponding to the collection device and a collection length corresponding to the cable collection area are obtained, and cable properties of the target cable are retrieved, wherein the cable properties include cable length, cable height and cable angle;

[0013] The number of acquisitions is obtained according to the ratio of the cable length to the acquisition length, and when it is determined that the number of acquisitions is an integer, the number of acquisitions is doubled to obtain the number of divisions;

[0014] The cable height is evenly divided based on the number of divisions to obtain a plurality of height acquisition segments, the height distance of the height acquisition segments is acquired, and the height distance is doubled to obtain a decreasing distance;

[0015] According to the difference between the cable height and the height distance, an initial height is obtained, and the initial height is successively decreased based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple acquisition heights;

[0016] Based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain a plurality of upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point, wherein the upward-looking hovering point has a corresponding downward-looking hovering point;

[0017] Counting the upward-looking hovering points in sequence along the target cable based on a first preset direction to obtain a comprehensive preview path corresponding to the target cable;

[0018] Based on the comprehensive preview path, images of the upper surface of the target cable are sequentially captured to obtain preview capture images corresponding to each upward-looking hovering point in the comprehensive preview path.

[0019] Optionally, in a possible implementation of the first aspect, the method further includes:

[0020] When it is determined that the acquisition times is not an integer, extracting an integer value from the acquisition times, and obtaining an evenly divided length according to the product of the integer value and the acquisition length;

[0021] Obtaining a remaining length according to a difference between the cable length and the average length, and determining a remaining height based on the remaining length and the cable angle;

[0022] Based on the difference between the cable height and the remaining height, an average height is obtained, and the integer value is doubled to obtain the number of divisions;

[0023] Evenly dividing the average height based on the number of divisions to obtain a plurality of height acquisition segments, acquiring height distances of the height acquisition segments, and doubling the height distances to obtain decreasing distances;

[0024] According to the difference between the cable height and the height distance, an initial height is obtained, and the initial height is successively decreased based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple acquisition heights;

[0025] Based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain a plurality of upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point;

[0026] Determine an upper positioning area in the cable collection area based on the remaining length, taking the upper boundary line of the cable collection area as a starting line;

[0027] Determine an outer identification area of ​​an upper surface and an inner identification area of ​​a lower surface of the target cable based on the remaining length and a second preset direction;

[0028] When the upper positioning area corresponding to the acquisition device overlaps with the outer identification area, the current hovering point is obtained as the upward-looking hovering point;

[0029] When the upper positioning area corresponding to the acquisition device overlaps with the inner identification area, the current hovering point is obtained as the downward-looking hovering point, and the upward-looking hovering point has a corresponding downward-looking hovering point;

[0030] The upward-looking hovering points are counted sequentially along the target cable based on a first preset direction to obtain a comprehensive preview path corresponding to the target cable.

[0031] Optionally, in a possible implementation manner of the first aspect, performing multi-dimensional abnormality recognition on the pre-check acquisition image according to the loose abnormality recognition strategy to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point includes:

[0032] According to the corrosion anomaly recognition strategy, corrosion anomaly recognition is performed on the pre-inspection acquisition image to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point;

[0033] Based on the defect and abnormality recognition strategy, the pre-inspection acquisition image is subjected to defect and abnormality recognition to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point;

[0034] The pre-checked captured image is subjected to burr anomaly recognition according to the burr anomaly recognition strategy to obtain an abnormal captured image corresponding to the corresponding upward-looking hovering point.

[0035] Optionally, in a possible implementation manner of the first aspect, performing corrosion anomaly recognition on the pre-inspection acquired image according to the corrosion anomaly recognition strategy to obtain an abnormal acquired image corresponding to a corresponding upward-looking hovering point includes:

[0036] Retrieving a corrosion pixel value, and determining a corrosion pixel point in the pre-check acquisition image based on the corrosion pixel value;

[0037] The number of corroded pixel points in the pre-check acquisition image is counted to obtain the corrosion number, and when it is determined that the corrosion number is greater than the preset corrosion number, the pre-check acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

[0038] Optionally, in a possible implementation manner of the first aspect, the performing defect and abnormality recognition on the pre-check acquisition image based on the defect and abnormality recognition strategy to obtain the abnormal acquisition image corresponding to the corresponding upward-looking hovering point includes:

[0039] Acquire the image center point of the pre-check acquisition image, construct a coordinate system based on the image center point, and obtain the coordinates of the pixel points corresponding to the pre-check acquisition image;

[0040] Binarization is performed on the corresponding pre-check acquisition image to obtain a cable recognition image, the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area are obtained as the surface horizontal coordinate, and the horizontal coordinate between the maximum horizontal coordinate and the minimum horizontal coordinate is used as the internal horizontal coordinate;

[0041] Constructing a surface vertical line perpendicular to the horizontal axis of the coordinate system according to the surface horizontal coordinate, and when determining that there are non-adjacent black pixels on the surface vertical line in the cable recognition image, sequentially counting the coordinates of the black pixels on each of the surface vertical lines to obtain a defective coordinate set;

[0042] Extracting coordinates corresponding to two adjacent black pixel points whose ordinates are discontinuous in the defect coordinate set as defect judgment coordinates;

[0043] Obtaining a defect judgment value according to the absolute value of the vertical coordinate difference in the defect judgment coordinate, and obtaining the number of the defect judgment values ​​to obtain the defect number;

[0044] When it is determined that the defect judgment value is greater than a preset missing value, or the defect quantity is greater than a preset defect quantity, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image;

[0045] An internal vertical line perpendicular to the horizontal axis of the coordinate system is constructed according to the internal horizontal coordinate. When it is determined that there are non-adjacent black pixels on the internal vertical line in the cable recognition image, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

[0046] Optionally, in a possible implementation manner of the first aspect, performing burr abnormality recognition on the pre-check acquisition image according to the burr abnormality recognition strategy to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point includes:

[0047] Acquire the image center point of the pre-check acquisition image, construct a coordinate system based on the image center point, and obtain the coordinates of the pixel points corresponding to the pre-check acquisition image;

[0048] Binarization is performed on the corresponding pre-check acquisition image to obtain a cable recognition image, and a standard width is obtained according to the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area;

[0049] Based on the same vertical coordinate, the maximum horizontal coordinate corresponding to the black pixel point in the cable recognition image is obtained as the first horizontal coordinate, and the minimum horizontal coordinate is obtained as the second horizontal coordinate;

[0050] The actual width is obtained according to the absolute value of the difference between the first horizontal coordinate and the second horizontal coordinate, and when it is determined that the actual width is greater than the standard width, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

[0051] Optionally, in a possible implementation manner of the first aspect, sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable based on the abnormal acquired image to obtain the abnormal focusing path includes:

[0052] Based on the abnormal acquisition image, determine the corresponding upward-looking hovering point as the upward-looking abnormal point, and determine the corresponding downward-looking hovering point as the direct downward-looking acquisition point according to the upward-looking abnormal point;

[0053] Acquire a downward-looking hovering point adjacent to the direct downward-looking acquisition point as an indirect downward-looking acquisition point, and obtain a downward-looking acquisition point based on the direct downward-looking acquisition point and the indirect downward-looking acquisition point;

[0054] The downward-looking acquisition points corresponding to the lower surface of the target cable are counted in sequence according to the second preset direction to obtain an abnormal focusing path.

[0055] Optionally, in a possible implementation manner of the first aspect, calculating the abnormality ratio in the abnormal acquisition image to obtain a looseness analysis value corresponding to the target cable, and determining a looseness identification result of the target cable based on the looseness analysis value includes:

[0056] Counting the number of abnormal collected images to obtain the number of abnormal images, and obtaining the number of images collected by the collection device for the target cable as the number of collected images, and obtaining the proportion of abnormal images according to the proportion of the number of abnormal images and the number of collected images;

[0057] The cable collection area is updated to the abnormal collection image, and the number of pixel points in the cable collection area in the abnormal collection image is counted to obtain the total number of areas;

[0058] Counting the number of corroded pixel points in the abnormally collected image based on the corroded pixel value to obtain the number of regional corrosion, and obtaining the corrosion proportion according to the ratio of the number of regional corrosion to the total number of the region;

[0059] Binarization is performed on the abnormal acquisition image to obtain a cable identification image, and the number of white pixels in the cable acquisition area in the cable identification image is counted to obtain the number of regional defects, and the defect ratio is obtained according to the ratio of the number of regional defects to the total number of regions;

[0060] Determine the remaining area outside the cable collection area in the cable recognition image as a burr judgment area;

[0061] Counting the number of black pixels in the burr determination area in the cable recognition image to obtain the number of regional burrs, and obtaining the burr ratio according to the ratio of the number of regional burrs to the total number of regional burrs;

[0062] Calculate according to the abnormal image proportion, corrosion proportion, defect proportion and burr proportion to obtain a looseness analysis value corresponding to the target cable;

[0063] When it is determined that the looseness analysis value is greater than the preset looseness value, a looseness identification result corresponding to the target cable is generated and sent to the maintenance end.

[0064] Optionally, in a possible implementation manner of the first aspect, the calculating according to the abnormal image proportion, the corrosion proportion, the defect proportion and the burr proportion to obtain a looseness analysis value corresponding to the target cable includes:

[0065] The looseness analysis value is obtained by the following formula:

[0066]

[0067] Among them, A is the looseness analysis value, n is the number of abnormal images, N is the number of collected images, k n is the image weight value, c is the number of regional corrosion, M is the total number of regions, k c is the corrosion weight value, d is the number of regional defects, k d is the defect weight value, f is the number of regional burrs, kf is the glitch weight value.

[0068] According to a second aspect of an embodiment of the present invention, an electronic device is provided, comprising: a memory, a processor and a computer program, wherein the computer program is stored in the memory, and the processor runs the computer program to execute the first aspect of the present invention and various methods that may be involved in the first aspect.

[0069] According to a third aspect of an embodiment of the present invention, a storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the first aspect of the present invention and various methods that may be involved in the first aspect.

[0070] The beneficial effects of the present invention are as follows:

[0071] 1. The present invention can generate corresponding detection paths according to the target cables that need to be detected, improve detection efficiency, and realize automatic and precise identification of cables. First, the present invention can determine a comprehensive pre-inspection path according to the cable attribute information of the target cable, so as to collect images of the upper surface of the target cable, so as to facilitate the subsequent multi-dimensional abnormality identification of the obtained pre-inspection collection images, thereby realizing abnormal detection of the upper surface of the target cable. Secondly, the present invention can perform dimensional abnormality identification on each pre-inspection collection image, so as to obtain an abnormal collection image corresponding to the corresponding upward hovering point, so as to facilitate the subsequent determination of the downward hovering point corresponding to the lower surface that needs to be imaged according to the upward hovering point corresponding to the abnormal collection image, thereby obtaining an abnormal focusing path, realizing automatic inspection and shooting, reducing manual detection time, and improving detection efficiency. Finally, the present invention can calculate the abnormal proportion of the abnormal collection image, obtain the looseness analysis value corresponding to the target cable, and thereby determine the looseness identification result corresponding to the target cable, realizing automatic detection while improving the accuracy of the detection result.

[0072] 2. The present invention can be customized according to the target cable to obtain a comprehensive pre-inspection path, realize rapid collection by the acquisition device, reduce the time cost of manual inspection, and thus improve the detection efficiency of the target cable. Among them, the present invention can obtain the cable length, cable height and cable angle of the target cable, and then determine the number of acquisitions and the number of segments according to the acquisition length, so as to segment the cable, thereby facilitating the determination of the corresponding shooting position of the acquisition device, and thus making the acquired image clearer, facilitating the identification of abnormal conditions of the target cable, and improving the accuracy of the detection results. In addition, the present invention can count the upward hovering points in sequence along the target cable from the top to the bottom, so as to obtain a comprehensive pre-inspection path for the acquisition device to collect the target cable, realize rapid collection by the acquisition device, reduce the time cost of manual inspection, and thus improve the detection efficiency of the target cable.

[0073] 3. The present invention can identify corrosion anomalies, defect anomalies and burr anomalies on the collected pre-inspection images, thereby realizing automatic identification of abnormal collected images, improving detection efficiency and ensuring the accuracy of detection results. Among them, the present invention can perform binarization processing on the pre-inspection collected images to obtain a cable recognition image, and then count the number of pixel points of the corresponding pixel values ​​to obtain the proportion of each abnormal problem, and then refer to the multi-dimensional data for calculation to obtain the looseness analysis value to ensure the accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A flow chart of a cable loosening identification method based on data processing provided by the present invention;

[0075] Figure 2 A schematic diagram of a collection interface provided by the present invention;

[0076] Figure 3 A schematic diagram of an upward-looking hovering point and a downward-looking hovering point provided by the present invention;

[0077] Figure 4 A schematic diagram of a cable recognition image provided by the present invention;

[0078] Figure 5 A schematic diagram of the hardware structure of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0079] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0080] like Figure 1 As shown, the present invention provides a flow chart of a cable loosening identification method based on data processing, and the cable loosening identification method based on data processing includes:

[0081] S1, responding to a looseness identification request, determining a comprehensive pre-inspection path according to cable properties of a target cable, and performing image acquisition on an upper surface of the target cable based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection acquisition images.

[0082] It should be noted that when a request for loose cable identification is received, a comprehensive pre-inspection path for image acquisition of the target cable can be determined based on the various cable attribute data of the target cable, such as the length of the cable, the cutting angle, etc., so that the drone can perform image acquisition and identification of the target cable based on the obtained comprehensive pre-inspection path, and then, in order to facilitate the subsequent identification of the collected pre-inspection image, thereby determining the abnormal information of the cable, and taking timely measures to ensure safety.

[0083] It can be understood that the looseness identification request is the demand information for requesting the identification of the cable, the target cable is the cable that needs to be detected and identified, the cable attribute is the attribute information corresponding to the target cable, such as the length, position, the cutting angle between the cable and the ground, etc., the comprehensive pre-inspection path is the path for detecting and identifying the target cable, and the pre-inspection acquisition image is the image acquired of the target cable according to the comprehensive pre-inspection path.

[0084] Through the above implementation, the present invention can determine a comprehensive pre-check path, thereby obtaining a pre-check acquisition image, so as to facilitate subsequent recognition of the pre-check acquisition image, thereby determining the working state of the target cable.

[0085] In some embodiments, the specific implementation of step S1 (the step of responding to the looseness identification request, determining a comprehensive pre-inspection path according to the cable properties of the target cable, and performing image acquisition on the upper surface of the target cable based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection acquisition images) includes:

[0086] S11, in response to a looseness identification request, obtaining a cable collection area in a collection interface corresponding to a collection device and a collection length corresponding to the cable collection area, and retrieving cable properties of a target cable, wherein the cable properties include cable length, cable height and cable angle.

[0087] It should be noted that the cable is usually in an inclined state. Therefore, when using a drone to shoot, the position may be offset during the movement, so that the image with the cable cannot be captured. Therefore, in order to enable the drone to clearly capture the surface image of the cable, the captured image is recognized and confirmed in real time, so that Figure 2 As shown, the cable collection area can be set in the collection interface, so that when the image of the cable taken deviates from the cable collection area, the shooting angle of the drone can be automatically adjusted to ensure that the cable in the collected image is in the corresponding cable collection area, thereby facilitating the subsequent cable anomaly identification.

[0088] It can be understood that the acquisition device is a device for acquiring images of the target cable, for example, it can be a drone, the acquisition interface is an interface for acquiring images, the cable acquisition area is an area for acquiring cable images, the acquisition length is the length of the cable in the acquired image, for example, it can be 1 meter, and the cable properties include cable length, cable height and cable angle.

[0089] Among them, the cable length is the actual length of the cable, for example, it can be 20 meters, the cable angle is the inclination angle between the cable and the ground, and the cable height is the height distance of the cable in space. Since the cable is in an inclined state, the length and height of the cable are different. For example, when the cable length is 20 meters and the cable angle is 30°, the corresponding cable height is 20×sin30°=10 meters.

[0090] Through the above implementation, the present invention can obtain the acquisition length, as well as the cable length, cable height and cable angle, which is convenient for subsequent segmentation processing of the cable, thereby facilitating image acquisition and obtaining a clear pre-inspection acquisition image for easy identification and detection.

[0091] S12, obtaining the number of acquisitions according to the ratio of the cable length to the acquisition length, and when it is determined that the number of acquisitions is an integer, doubling the number of acquisitions to obtain the number of divisions.

[0092] It should be noted that the length of the cable is usually long, up to more than ten meters or even dozens of meters. Therefore, when the cable is detected for abnormalities, it is necessary to capture images of the cable so as to identify the captured images. In order to make the captured images clearer and easier to identify and view, the cable can be captured in segments so that each captured image can clearly show the image information of the cable surface. For example, when the cable is 20 meters long, a 2-meter-long cable image can be captured each time it is shot. Therefore, the cable can be divided into 10 parts, and the image information of the entire cable can be obtained by shooting 10 times.

[0093] It can be understood that the number of acquisition times is the number of times the cable is photographed and images are collected, that is, it can be obtained by the ratio of the cable length and the acquisition length. For example, when the cable length is 20 meters and the acquisition length is 2 meters, the number of acquisition times can be obtained as 20 / 2=10 times, and the number of segmentations is twice the value of the number of acquisitions. For example, when the number of acquisitions is 10, the corresponding number of segmentations is 10×2=20.

[0094] It is not difficult to understand that in order to obtain a clear cable image and determine the location point for photographing and collecting the target cable, the corresponding cable needs to be segmented, and because when photographing each section of the cable, the collection device needs to be located at the center position of each section of the cable, the target cable needs to be segmented twice.

[0095] For example, when the cable length is 20 meters and the acquisition length is 2 meters, the target cable can be divided into 20 sections. At the same time, in order to determine the acquisition position point of each section of the cable, the center point of each section of the cable can be determined, and then the positions of 10 center points can be obtained. Therefore, the number of segmentation points for position segmentation of the entire target cable is 10+10=20.

[0096] S13, evenly dividing the cable height based on the number of divisions to obtain a plurality of height acquisition segments, acquiring height distances of the height acquisition segments, and doubling the height distances to obtain decreasing distances.

[0097] It can be understood that after the cable height is evenly divided according to the number of divisions, the moving height of the acquisition device for image acquisition can be determined later. Since the acquisition device moves from one acquisition point to another when shooting, that is, it moves the distance of two height acquisition segments, the height distance can be doubled to obtain a decreasing distance.

[0098] Among them, the height collection section is the collection distance section divided on the height, the height distance is the distance of the height collection section, and the decreasing distance is the distance that the collection device moves and decreases.

[0099] For example, when the number of segments is 20 and the cable height is 10 meters, 20 height acquisition segments with a length of 0.5 meters can be obtained, thereby obtaining a height distance of 0.5 meters, and then doubling the height distance to obtain a decreasing distance of 0.5×2=1 meter.

[0100] Through the above implementation, the present invention can determine the height acquisition segment, the height distance and the corresponding decreasing distance, so as to subsequently determine the point where the acquisition device performs hovering shooting and perform image acquisition.

[0101] S14, obtaining an initial height according to the difference between the cable height and the height distance, and sequentially decreasing the initial height based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple collection heights.

[0102] It can be understood that the initial height is the height at which the acquisition device initially photographs the target cable, that is, it can be obtained by the difference between the cable height and the height distance. For example, when the cable height is 10 meters and the height distance is 0.5 meters, the initial acquisition position height of the drone can be obtained as 10-0.5=9.5 meters.

[0103] Among them, the remaining height is the distance remaining after the initial height is reduced by the decreasing distance in sequence, and the acquisition height is the hovering height of the acquisition device for image acquisition. For example, when the initial height is 9.5 meters and the decreasing distance is 1 meter, the remaining distance can be obtained as 9.5-1=8.5 meters. The decreasing distance is reduced by 1 meter again to obtain a remaining distance of 8.5-1=7.5 meters, ..., and the remaining distance is 1.5-1=0.5 meters. 0.5-1=-0.5<0, thereby obtaining multiple acquisition heights of 8.5 meters, 7.5 meters...1.5 meters, and 0.5 meters respectively.

[0104] Through the above implementation, the present invention can obtain the acquisition height, so as to subsequently control the acquisition device to move to the corresponding height to acquire the cable image.

[0105] S15, based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain multiple upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point, and the upward-looking hovering point has a corresponding downward-looking hovering point.

[0106] It can be understood that the positioning height is the height at which the acquisition device acquires images of the cable, including the initial height and the acquisition height.

[0107] It is not difficult to understand that a point on the target cable that is at the same height as the positioning height can be determined as the positioning point, and because the acquisition device needs to maintain a certain spacing distance from the target cable when acquiring the target cable, so that the acquisition device can complete the image acquisition of each section of the cable, a height can be set artificially in advance, so that when the preset height is added to the positioning point, the corresponding upward hovering point is obtained, and the preset height is subtracted from the positioning point to obtain the corresponding downward hovering point, thereby determining multiple upward hovering points and downward hovering points corresponding to the target cable, so as to obtain a comprehensive preview path for image acquisition.

[0108] Among them, the preset height is a pre-set height, which is used to determine the corresponding upward hovering point and downward hovering point. The upward hovering point is the point where it hovers above the cable for collection, and the downward hovering point is the point where it hovers below the cable for collection, and the upward hovering point has a corresponding downward hovering point.

[0109] For example: Figure 3 As shown, when the preset height is added to the positioning point 1, the corresponding upward-looking hovering point 1 is obtained, and the preset height is subtracted from the positioning point 1 to obtain the corresponding downward-looking hovering point 1, and then multiple upward-looking hovering points and downward-looking hovering points corresponding to the target cable are determined, among which it can be concluded that the upward-looking hovering point 1 corresponds to the downward-looking hovering point 1.

[0110] S16, counting the upward-looking hovering points in sequence along the target cable based on a first preset direction to obtain a comprehensive pre-inspection path corresponding to the target cable.

[0111] It can be understood that the first preset direction is a moving direction from top to bottom.

[0112] It is not difficult to understand that by counting the upward-looking hovering points in order from the top to the bottom along the target cable, a comprehensive pre-inspection path for the acquisition device to acquire the target cable can be obtained.

[0113] For example, Figure 3 As shown, in the direction from top to bottom, the corresponding upward-looking hovering points are counted in sequence to obtain a comprehensive preview path corresponding to the target cable.

[0114] S17, based on the comprehensive preview path, sequentially capture images of the upper surface of the target cable to obtain preview capture images corresponding to each upward-looking hovering point in the comprehensive preview path.

[0115] It can be understood that the acquisition device can be controlled to shoot the upper surface of the target cable according to the comprehensive pre-inspection path, such as shooting at the position of the upward-looking hovering point 1, and then moving to the position of the upward-looking hovering point 2 to shoot again... until the corresponding pre-inspection acquisition image is obtained after shooting at each upward-looking hovering point in the comprehensive pre-inspection path.

[0116] It is not difficult to understand that the shooting angle for image acquisition at the upward hovering point of the positioning point corresponding to the initial height can be manually set in advance through debugging, or it can be automatically adjusted according to the angle between the cable in the captured pre-inspection acquisition image and the cable acquisition area after the offset, so that the cable in the captured pre-inspection acquisition image is in the cable acquisition area, so as to facilitate subsequent abnormality detection.

[0117] Through the above implementation, the present invention can obtain a comprehensive pre-inspection path, realize rapid collection by the collection device, reduce the time cost of manual inspection, and thus improve the detection efficiency of the target cable.

[0118] It should be noted that, since the cable can be used in multiple scenarios and due to differences in related equipment, the length of the cable may also be different. When the length of the cable is not an integer multiple of the acquisition length, the corresponding number of acquisitions is also not an integer. In order to completely capture the target cable, in some embodiments, the following steps are further included:

[0119] A01, when it is determined that the acquisition times is not an integer, extract the integer value in the acquisition times, and obtain the evenly divided length according to the product of the integer value and the acquisition length.

[0120] It can be understood that when the ratio of the cable length to the collection length is calculated and the obtained value is not an integer, the integer value in the collection times can be extracted to obtain the average length, thereby dividing the cable length, and then processing and determining the hovering point position of the collection device separately, so as to realize automatic collection and detection of the target cable.

[0121] The integer value is the integer value of the number of collection times. For example, when the calculated number of collection times is 10.5, the corresponding integer value is 10.

[0122] Moreover, the average length is the length that can achieve average division, that is, the product of the integer value and the collection length. For example, when the integer value is 10 and the collection length is 2 meters, the average length can be obtained as 10×2=20 meters.

[0123] A02, obtaining a remaining length according to the difference between the cable length and the average length, and determining a remaining height based on the remaining length and the cable angle.

[0124] It can be understood that the remaining length is the difference between the cable length and the average length, and the remaining height is the height of the target cable determined according to the remaining length.

[0125] For example, when the cable length is 21 meters and the average length is 20 meters, the remaining length is 21-20=1 meter. When the cable angle is 30°, the remaining height is 1×sin30°=0.5 meter.

[0126] Through the above implementation, the present invention can obtain the remaining length and the remaining height, so as to facilitate the subsequent image acquisition of the cable of the corresponding length.

[0127] A03, based on the difference between the cable height and the remaining height, obtain the average height, double the integer value to obtain the number of divisions.

[0128] It can be understood that the average height is the difference between the cable height and the remaining height. For example, when the cable height is 10.5 meters and the remaining height is 0.5 meters, the average height can be 10 meters.

[0129] It is not difficult to understand that when the average height is processed to determine the hovering point of the corresponding acquisition device, the corresponding number of divisions is obtained in accordance with the doubling process of the above step S12, so as to divide the target cable according to the number of divisions.

[0130] A04, evenly divide the average height based on the number of divisions to obtain multiple height acquisition segments, obtain height distances of the height acquisition segments, and double the height distances to obtain decreasing distances.

[0131] It can be understood that after the average height is evenly divided according to the number of divisions, the moving height of the acquisition device for image acquisition can be determined subsequently. Since the acquisition device moves from one position point to another, that is, it moves the distance of two height acquisition segments, the height distance can be doubled to obtain a decreasing distance.

[0132] Among them, the height collection section is the collection distance section divided on the height, the height distance is the distance of the height collection section, and the decreasing distance is the distance that the collection device moves and decreases.

[0133] For example, when the number of segments is 20 and the cable height is 10 meters, 20 height acquisition segments with a length of 0.5 meters can be obtained, thereby obtaining a height distance of 0.5 meters, and then doubling the height distance to obtain a decreasing distance of 0.5×2=1 meter.

[0134] Through the above implementation, the present invention can determine the height acquisition segment, the height distance and the corresponding decreasing distance, so as to subsequently determine the point where the acquisition device performs hovering shooting and perform image acquisition.

[0135] A05, obtaining an initial height according to the difference between the cable height and the height distance, and successively decreasing the initial height based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple collection heights.

[0136] It can be understood that this implementation is consistent with step S14 in which the number of acquisitions is an integer, and will not be described in detail here.

[0137] A06, based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain multiple upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point.

[0138] It can be understood that the positioning height is the height at which the acquisition device acquires images of the cable, including the initial height and the acquisition height.

[0139] It is not difficult to understand that a point on the target cable that is at the same height as the positioning height can be determined as the positioning point, and because the acquisition device needs to maintain a certain spacing distance from the target cable when acquiring the target cable, so that the acquisition device can complete the image acquisition of each section of the cable, a height can be set artificially in advance, so that when the preset height is added to the positioning point, the corresponding upward hovering point is obtained, and the preset height is subtracted from the positioning point to obtain the corresponding downward hovering point, thereby determining multiple upward hovering points and downward hovering points corresponding to the target cable, so as to obtain a comprehensive preview path for image acquisition.

[0140] Among them, the preset height is a pre-set height, which is used to determine the corresponding upward hovering point and downward hovering point. The upward hovering point is the point where it hovers above the cable for collection, and the downward hovering point is the point where it hovers below the cable for collection, and the upward hovering point has a corresponding downward hovering point.

[0141] A07, taking the upper boundary line of the cable collection area as a starting line, determining an upper positioning area in the cable collection area based on the remaining length.

[0142] It should be noted that in order to ensure that the image of the remaining length of the cable is completely captured, the upper boundary line of the cable acquisition area can be used as the starting line to determine the upper positioning area in the cable acquisition area, so that the image of the remaining part of the cable can be captured based on the upper positioning area.

[0143] It can be understood that the starting line is the upper boundary line of the cable collection area, and the upper positioning area is an area with the upper boundary line of the cable collection area as the starting line and the length as the remaining length that can be collected.

[0144] It is not difficult to understand that the length of the cable collection area is the same as the collection length. Since the remaining length is less than the collection length, the upper positioning area in the cable collection area can be determined by the remaining length. For example, when the collection length in the cable collection area for capturing the cable image is 2 meters, the area corresponding to the remaining length that can be collected can be determined in the cable collection area based on the remaining length.

[0145] A08, determining an outer identification area on the upper surface and an inner identification area on the lower surface of the target cable based on the remaining length and the second preset direction.

[0146] It should be noted that the portion of the cable corresponding to the remaining length is close to the ground. In order to determine the shooting area of ​​the cable corresponding to the remaining length of the target cable, the portion of the cable with the remaining length from the ground can be determined from bottom to top according to the second preset direction, and the corresponding outer identification area and inner identification area can be obtained.

[0147] It can be understood that the second preset direction is a pre-set direction, which can be a direction from bottom to top. The outer identification area is the area corresponding to the upper surface where image acquisition and identification can be performed, and the inner identification area is the area corresponding to the lower surface where image acquisition and identification can be performed.

[0148] A09, when the upper positioning area corresponding to the acquisition device overlaps with the outer recognition area, the current hovering point is obtained as the upward-looking hovering point.

[0149] It can be understood that when the upper positioning area of ​​the control acquisition device overlaps with the outer identification area, it can be understood that the acquisition device can capture the image of the outer identification area corresponding to the upper surface of the cable corresponding to the remaining length at the current position and shooting angle, so that the current hovering point can be used as the upward-looking hovering point.

[0150] Through the above implementation, the present invention can obtain the upward hovering point of the cable corresponding to the remaining length, so as to determine the corresponding comprehensive pre-inspection path later, thereby performing image acquisition.

[0151] A10, when the upper positioning area corresponding to the acquisition device overlaps with the inner identification area, the current hovering point is obtained as the downward-looking hovering point, and the upward-looking hovering point has a corresponding downward-looking hovering point.

[0152] It can be understood that when the upper positioning area of ​​the control acquisition device overlaps with the inner identification area, it can be understood that the acquisition device can capture the image of the inner identification area corresponding to the lower surface of the cable corresponding to the remaining length at the current position and shooting angle, so that the current hovering point can be used as the downward-looking hovering point.

[0153] Furthermore, the upward hovering point for capturing images of the remaining length of the cable corresponds to the downward hovering point, so as to facilitate subsequent image capture of the lower surface of the cable.

[0154] It is worth mentioning that when an outer identification area and an inner identification area appear in the target cable, it means that the cable length cannot be divided into an even number. The drone acquisition device can automatically adjust the position and shooting angle so that the upper positioning area in the acquisition interface can overlap with the outer identification area or the inner identification area, thereby obtaining the remaining target cable segment for image acquisition. It can also be the position of the acquisition point that is manually preset. For example, the corresponding upward-looking hovering point and downward-looking hovering point are respectively on the outside and inside of the target cable on the ground, and the acquisition angle can be manually set actively so that the upper positioning area in the acquisition device overlaps with the outer identification area and the inner identification area of ​​the target cable, respectively. Therefore, the upward-looking hovering point and the downward-looking hovering point of the remaining length of the target cable can be determined for image acquisition.

[0155] A11, counting the upward-looking hovering points in sequence along the target cable based on a first preset direction to obtain a comprehensive preview path corresponding to the target cable.

[0156] It can be understood that by counting the upward-looking hovering points in sequence from top to bottom along the target cable, a comprehensive pre-inspection path for the acquisition device to acquire the target cable can be obtained, so as to facilitate subsequent image acquisition of the target cable according to the comprehensive pre-inspection path, realize automatic inspection of the target cable, and reduce personnel inspection time.

[0157] S2, performing multi-dimensional anomaly recognition on the pre-inspection captured image according to the loose anomaly recognition strategy to obtain an abnormal captured image corresponding to the corresponding upward-looking hovering point.

[0158] It should be noted that the cable may have various abnormal conditions, such as surface wear, corrosion and rust, or broken wires. Different types of abnormal problems are handled differently. Therefore, the pre-inspection image is identified for abnormalities in multiple dimensions to obtain the abnormal image corresponding to the corresponding upward hovering point, thereby realizing automatic detection and improving the accuracy of the detection results.

[0159] It can be understood that the pre-check captured image is identified in multiple dimensions to determine abnormal information of the target cable, thereby obtaining an abnormal captured image.

[0160] The abnormal acquisition image is an image in which the target cable in the acquisition image has abnormal information.

[0161] In some embodiments, the specific implementation of step S2 (performing multi-dimensional abnormality recognition on the pre-checked captured image according to the loose abnormality recognition strategy to obtain an abnormal captured image corresponding to the corresponding upward-looking hovering point) includes:

[0162] S21, performing corrosion anomaly recognition on the pre-inspection captured image according to a corrosion anomaly recognition strategy to obtain an abnormal captured image corresponding to a corresponding upward-looking hovering point.

[0163] It can be understood that through the corrosion anomaly recognition strategy, the corrosion of the pre-inspection acquisition image can be identified, so as to determine whether the corresponding target cable is corroded, thereby obtaining the abnormal acquisition image corresponding to the corresponding upward-looking hovering point and determining the abnormal corrosion condition of the corresponding target cable.

[0164] In some embodiments, the specific implementation of step S21 (the corrosion anomaly identification is performed on the pre-inspection acquisition image according to the corrosion anomaly identification strategy to obtain the abnormal acquisition image corresponding to the corresponding upward-looking hovering point) includes:

[0165] S211, retrieve the corrosion pixel value, and determine the corrosion pixel point in the pre-check acquisition image based on the corrosion pixel value.

[0166] It can be understood that the corrosion pixel value is a pixel value indicating corrosion, for example, it can be yellow, and the corrosion pixel point is a pixel point corresponding to the corrosion pixel value in the pre-check acquisition image.

[0167] It is not difficult to understand that the corresponding corroded pixel points can be determined in the pre-check acquisition image by corroding the pixel values, so as to obtain the corresponding abnormal acquisition image later.

[0168] S212, counting the number of corroded pixel points in the pre-check acquisition image to obtain a corrosion number, and when it is determined that the corrosion number is greater than a preset corrosion number, taking the pre-check acquisition image corresponding to the upward-looking hovering point as an abnormal acquisition image.

[0169] It can be understood that the corrosion number is the number of corroded pixel points in the pre-checked captured image, and the preset corrosion number is the number of pre-set corroded pixel points.

[0170] It is not difficult to understand that when the amount of corrosion in the pre-inspection acquisition image is less than the preset corrosion amount, it can be explained that the area of ​​corrosion and rust on the corresponding target cable is not large, and will not affect the normal function of the cable. However, when it is determined that the amount of corrosion is greater than the preset corrosion amount, it means that the corrosion problem at the corresponding position of the target cable is more serious, which may affect the normal use function of the cable. Therefore, in order to obtain accurate detection results, the pre-inspection acquisition image of the corresponding upward hovering point can be used as an abnormal acquisition image.

[0171] S22, performing defect and abnormality recognition on the pre-inspection acquired image based on a defect and abnormality recognition strategy to obtain an abnormal acquired image corresponding to the corresponding upward-looking hovering point.

[0172] It can be understood that defects and abnormalities are identified on each pre-inspection captured image to determine whether the corresponding target cable has defects or abnormalities, thereby obtaining an abnormal captured image corresponding to the corresponding upward-looking hovering point, which is convenient for subsequent analysis, thereby obtaining the looseness identification result of the target cable and improving the accuracy of the detection data.

[0173] In some embodiments, the specific implementation of step S22 (the defect and abnormality recognition of the pre-inspection acquisition image based on the defect and abnormality recognition strategy to obtain the abnormal acquisition image corresponding to the corresponding upward-looking hovering point) includes:

[0174] S221, acquiring the image center point of the pre-flight acquisition image, constructing a coordinate system based on the image center point, and obtaining the coordinates of the pixel points corresponding to the pre-flight acquisition image.

[0175] It can be understood that the center point of the image is the center point of the pre-check captured image.

[0176] It is not difficult to understand that the center point of the pre-inspection image can be obtained, and a coordinate system can be constructed based on the center point of the image to obtain the coordinates of each pixel point, thereby facilitating subsequent further processing of the corresponding image, thereby determining whether the target cable has any abnormal defects and completing automatic detection of the target cable.

[0177] S222, binarize the corresponding pre-check acquisition image to obtain a cable recognition image, obtain the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area as the surface horizontal coordinate, and use the horizontal coordinate between the maximum horizontal coordinate and the minimum horizontal coordinate as the internal horizontal coordinate.

[0178] It should be noted that the pixel values ​​of the corresponding pixels of the cable image in the pre-inspection acquisition image and the pixel values ​​of the pixel points of the other non-cable images are binarized, so that the pixel values ​​of the pixel points corresponding to the cable image are processed as black, and the pixel values ​​of the pixel points of the other area images are processed as white, so as to obtain a cable identification image with clear black and white pixel values, so as to facilitate the subsequent judgment of whether the target cable has defects according to the black and white pixel values. For example, when the pixel value of the cable image in the corresponding area should be black, but is identified as white, it can be said that the cable at the corresponding position is worn.

[0179] It can be understood that the cable identification image is an image obtained after the pre-check acquisition image is binarized.

[0180] Among them, the maximum abscissa is the maximum value of the abscissa, the minimum abscissa is the minimum value of the abscissa, the surface abscissa is the maximum and minimum abscissas corresponding to the cable collection area, and the internal abscissa is the abscissa between the maximum and minimum abscissas.

[0181] For example, when the maximum horizontal coordinate of the cable collection area is determined to be 5 and the minimum horizontal coordinate is -5 in the coordinate system, the surface horizontal coordinates are 5 and -5, and 4, 3, 2, 1, 0, -1, -2, -3, and -4 are obtained as the internal horizontal coordinates.

[0182] Through the above implementation, the present invention can determine the surface horizontal coordinate and the internal horizontal coordinate, so as to facilitate the subsequent judgment of whether the target cable has an abnormal problem of defect.

[0183] S223, constructing a surface vertical line perpendicular to the horizontal axis of the coordinate system according to the surface horizontal coordinate, and when it is determined that there are non-adjacent black pixels on the surface vertical line in the cable identification image, counting the coordinates of the black pixels on each of the surface vertical lines in turn to obtain a missing coordinate set.

[0184] It can be understood that the surface vertical line is a vertical line perpendicular to the horizontal axis passing through the coordinate point corresponding to the surface horizontal coordinate, the black pixel point is a pixel point with a black pixel value, and the defect coordinate set is a set of coordinates of the black pixel points on the surface vertical line.

[0185] It is not difficult to understand that when there are non-adjacent black pixels on the vertical line of the surface, it can be said that the target cable is missing at the corresponding position, and the cable defect may be caused by abnormal wear. Therefore, the coordinates of the corresponding black pixel points can be counted to obtain a defect coordinate set, so that the size of the defect area can be obtained based on the defect coordinate set in the future, thereby improving the accuracy of the detection results.

[0186] S224, extracting coordinates corresponding to two adjacent black pixel points with discontinuous vertical coordinates in the defect coordinate set as defect judgment coordinates.

[0187] It is understandable that in order to determine the length of the target cable defect, the coordinates corresponding to two adjacent black pixels with discontinuous ordinates in the defect coordinate set may be extracted, thereby facilitating the subsequent acquisition of the length of the defect area.

[0188] The defect judgment coordinates are the coordinates corresponding to two adjacent black pixel points with discontinuous vertical coordinates in the defect coordinate set. For example, when the defect coordinate set is {(5,0), (5,1), (5,4), (5,5), (5,6), (5,7), (5,8)}, since the vertical coordinates are not adjacent, the coordinates (5,1) and (5,4) can be used as the defect judgment coordinates.

[0189] S225, obtaining a defect judgment value according to the absolute value of the vertical coordinate difference in the defect judgment coordinate, and obtaining the number of the defect judgment values ​​to obtain the defect number.

[0190] It can be understood that the defect judgment value is the absolute value of the difference between the vertical coordinates in the defect judgment coordinates. For example, when the defect judgment coordinates are (5,1) and (5,4), the corresponding vertical coordinates are 1 and 4 respectively, and the defect judgment value can be obtained as 4-1=3.

[0191] Among them, the number of defects is the number of defect judgment values. In order to determine the number of defects in the cable in the corresponding image, the number of defect judgment values ​​can be counted to obtain the corresponding number of defects. It is not difficult to understand that the number of defect judgment values ​​corresponds to the number of defects in the actual cable.

[0192] S226, when it is determined that the defect judgment value is greater than a preset missing value, or the defect quantity is greater than a preset defect quantity, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

[0193] It can be understood that the preset missing value is a preset value for determining defects, which may be preset manually, and the preset defect number is a preset number of defects, for example, it may be 2.

[0194] It is not difficult to understand that when the defect judgment value is greater than the preset missing value, it means that the length of the cable defect exceeds the normal wear length and there is a safety hazard, or when the number of defects is greater than the preset number of defects, it can also mean that the number of defects in the current cable is abnormal. If it continues to be used without maintenance, dangerous accidents may occur. Therefore, the pre-inspection acquisition image of the corresponding upward hovering point can be used as an abnormal acquisition image to facilitate subsequent multi-dimensional evaluation calculations to obtain the detection results of the target cable, ensure the accuracy of the detection results, and improve the detection efficiency.

[0195] S227, constructing an internal vertical line perpendicular to the horizontal axis of the coordinate system according to the internal horizontal coordinate, and when it is determined that there are non-adjacent black pixels on the internal vertical line in the cable recognition image, taking the pre-inspection acquisition image corresponding to the upward-looking hovering point as an abnormal acquisition image.

[0196] It can be understood that the internal vertical line is a vertical line passing through the coordinate point corresponding to the internal horizontal coordinate and perpendicular to the horizontal axis.

[0197] It is not difficult to understand that when non-adjacent black pixels are identified on the internal vertical line, it can be explained that the lateral loss of the target cable has penetrated deep into the cable, which can indicate that the wear of the cable is abnormally serious, so the pre-inspection acquisition image of the corresponding upward hovering point can be directly used as the abnormal acquisition image.

[0198] S23, performing burr anomaly recognition on the pre-inspection acquired image according to the burr anomaly recognition strategy to obtain an abnormal acquired image corresponding to the corresponding upward-looking hovering point.

[0199] It is understandable that when the iron wire wrapped around the surface of the target cable is broken, the wire at the break will stick out, causing burrs to appear on the cable. Therefore, the pre-inspection image can be used to identify burr anomalies to determine whether the target cable has broken iron wires, so that timely processing can be carried out to ensure the safety of the cable.

[0200] In some embodiments, the specific implementation of step S23 (the burr abnormality recognition is performed on the pre-checked acquired image according to the burr abnormality recognition strategy to obtain the abnormal acquired image corresponding to the corresponding upward-looking hovering point) includes:

[0201] S231, acquiring the image center point of the pre-flight acquisition image, constructing a coordinate system based on the image center point, and obtaining the coordinates of the pixel points corresponding to the pre-flight acquisition image.

[0202] It can be understood that the center point of the image is the center point of the pre-check image, and the image center point can be used as the origin of the coordinate system to establish a coordinate system, thereby facilitating the acquisition of coordinates of corresponding pixels in the pre-check image, and further facilitating the identification of the pre-check image.

[0203] S232, binarizing the corresponding pre-check acquisition image to obtain a cable recognition image, and obtaining a standard width according to the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area.

[0204] It can be understood that, consistent with the implementation of the above step S222, the pre-check acquisition image is binarized to obtain a binarized cable identification image, so as to facilitate subsequent determination of whether the corresponding cable has burr abnormalities.

[0205] The standard width is the lateral width corresponding to the cable collection area, which can be obtained by the difference between the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable collection area. For example, when the maximum horizontal coordinate of the cable collection area is 5 and the minimum horizontal coordinate is -5, the standard width can be obtained as 5-(-5)=10.

[0206] S233, based on the same vertical coordinate, obtaining the maximum horizontal coordinate corresponding to the black pixel points in the cable recognition image as the first horizontal coordinate, and the minimum horizontal coordinate as the second horizontal coordinate.

[0207] It can be understood that the first horizontal coordinate is the maximum horizontal coordinate corresponding to the black pixel points with the same vertical coordinate in the Lasso recognition image, and the second horizontal coordinate is the minimum horizontal coordinate corresponding to the black pixel points with the same vertical coordinate in the Lasso recognition image.

[0208] Through the above implementation, the present invention can obtain the first horizontal coordinate and the second horizontal coordinate of each pixel point corresponding to the same vertical coordinate, so as to obtain the actual width subsequently, thereby comparing it with the standard width to determine the burr condition of the target cable.

[0209] S234, obtaining an actual width according to the absolute value of the difference between the first horizontal coordinate and the second horizontal coordinate, and when it is determined that the actual width is greater than the standard width, taking the pre-inspection acquisition image corresponding to the upward-looking hovering point as an abnormal acquisition image.

[0210] It can be understood that the actual width is the absolute value of the difference between the first horizontal coordinate and the second horizontal coordinate. For example, when the first horizontal coordinate is 8 and the second horizontal coordinate is -5, the actual width is 8-(-5)=13.

[0211] It is not difficult to understand that when the actual width is greater than the standard width, it can indicate that there is a prominent burr problem on the cable, so that the pre-inspection acquisition image of the corresponding upward-looking hovering point can be used as an abnormal acquisition image to facilitate subsequent image acquisition of the corresponding downward-looking hovering point to achieve complete detection of the target cable.

[0212] S3, based on the abnormal acquisition image, sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable to obtain an abnormal focusing path, and acquiring an abnormal acquisition image corresponding to the abnormal focusing path.

[0213] It can be understood that the corresponding upward-looking hovering point position can be determined based on the abnormal acquisition image, and then the corresponding downward-looking hovering point can be obtained based on the upward-looking hovering point, thereby determining the abnormal focusing path for performing lower surface abnormality detection on the target cable, so as to perform image acquisition on the lower surface of the cable and complete the corresponding abnormality detection.

[0214] The abnormal focusing path is a path for collecting the lower surface image of the target cable.

[0215] In some embodiments, the specific implementation of step S3 (the step of sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable based on the abnormal acquisition image to obtain the abnormal focusing path) includes:

[0216] S31, based on the abnormal acquisition image, determining the corresponding upward-view hovering point as an upward-view abnormal point, and determining the corresponding downward-view hovering point as a direct downward-view acquisition point according to the upward-view abnormal point.

[0217] It can be understood that the upward-looking abnormal point is the upward-looking hovering point corresponding to the abnormal acquisition image, and the direct downward-looking acquisition point is the downward-looking hovering point corresponding to the upward-looking abnormal point.

[0218] For example, when the abnormal acquisition image is the pre-inspection acquisition image corresponding to the upward-looking hovering point 3 obtained through identification and judgment, the upward-looking abnormal point is the upward-looking hovering point 3, and the corresponding downward-looking hovering point 3 can be used as the direct downward-looking acquisition point.

[0219] Through the above implementation, the present invention can determine the direct downward-looking acquisition point, so as to facilitate the subsequent determination of the indirect downward-looking acquisition point based on the direct downward-looking acquisition point, thereby obtaining the corresponding abnormal focusing path, and then performing image acquisition on the lower surface of the target cable.

[0220] S32, acquiring a downward-looking hovering point adjacent to the direct downward-looking acquisition point as an indirect downward-looking acquisition point, and obtaining a downward-looking acquisition point based on the direct downward-looking acquisition point and the indirect downward-looking acquisition point.

[0221] It should be noted that when abnormalities are identified in the pre-inspection acquisition image of the upper surface of the target cable, if no abnormal problems are identified in the pre-inspection acquisition image of multiple continuous target cables, it can be said that the actual target cable at the corresponding position is not damaged. Therefore, when the image of the lower surface of the target cable is subsequently acquired, only the downward-looking hovering point at the corresponding position of the abnormal acquisition image where the abnormality is identified and the adjacent downward-looking hovering points can be used as downward-looking acquisition points, which can reduce the amount of data processing, and at the same time can reduce the inspection time and improve the detection efficiency.

[0222] It can be understood that the indirect downward-looking acquisition point is a downward-looking hovering point adjacent to the direct downward-looking acquisition point. For example, when the direct downward-looking acquisition point is downward-looking hovering point 3, downward-looking hovering point 2 and downward-looking hovering point 4 adjacent to downward-looking hovering point 3 can be used as indirect downward-looking acquisition points.

[0223] Among them, the downward-looking acquisition points include direct downward-looking acquisition points and indirect downward-looking acquisition points.

[0224] S33, sequentially counting downward-looking acquisition points corresponding to the lower surface of the target cable according to the second preset direction to obtain an abnormal focusing path.

[0225] It can be understood that the downward-looking acquisition points corresponding to the target cable are counted in sequence according to the second preset direction to obtain an abnormal focusing path. For example, the abnormal focusing path may be downward-looking acquisition point 4-downward-looking acquisition point 3-downward-looking acquisition point 2.

[0226] Through the above implementation, the present invention can determine the abnormal focusing path for image acquisition of the lower surface of the target cable, thereby facilitating image acquisition by the acquisition device, reducing manual inspection time, and improving the efficiency of cable abnormality detection.

[0227] S4, calculating the abnormality ratio in the abnormal acquisition image to obtain a looseness analysis value corresponding to the target cable, and determining a looseness identification result of the target cable based on the looseness analysis value.

[0228] It should be noted that the anomalies in all abnormal acquisition images are summarized and the proportion of anomalies in the target cable is calculated. When the proportion of anomalies is large, it will affect the target cable and cause it to loosen, thereby posing a safety hazard. When the proportion of anomalies is small and still within the normal range, it may not affect the stability of the target cable. Therefore, the calculation and analysis can be performed based on the proportion of anomalies to obtain the looseness identification result of the target cable.

[0229] It can be understood that the abnormality ratio is the proportion of abnormal problems in the target cable, the looseness analysis value is the result value of the analysis of the looseness of the target cable, and the looseness identification result is the result of identifying whether the target cable is loose.

[0230] In some embodiments, the specific implementation of step S4 (the calculation of the abnormal proportion in the abnormal acquisition image, obtaining a looseness analysis value corresponding to the target cable, and determining the looseness identification result of the target cable based on the looseness analysis value) includes:

[0231] S41, counting the number of abnormal collected images to obtain the number of abnormal images, and obtaining the number of images collected by the collection device for the target cable as the number of collected images, and obtaining the proportion of abnormal images according to the number of abnormal images and the proportion of the number of collected images.

[0232] It can be understood that the number of abnormal images is the number of abnormally collected images. For example, when 3 abnormally collected images are obtained after identification and judgment, it can be concluded that the number of abnormal images is 3.

[0233] In addition, the number of collected images is the number of images collected by the collection device for the target cable. For example, when the drone collects images of the target cable and takes a total of 24 images, the number of collected images is 24.

[0234] Among them, the abnormal image ratio is the ratio of the number of images with abnormalities to the total number of collected images, that is, the ratio of the number of abnormal images to the number of collected images. For example, when the number of collected images is 24 and the number of abnormal collected images is 3, the abnormal image ratio can be obtained as 3 / 24=0.125.

[0235] S42, updating the cable collection area to the abnormal collection image, counting the number of pixel points in the cable collection area in the abnormal collection image, and obtaining the total number of areas.

[0236] It should be noted that the abnormal acquisition image is obtained by acquiring images of the target cable through an acquisition interface pre-set in the acquisition device, so that the specifications and sizes of the abnormal acquisition image and the acquisition interface are consistent. Therefore, the cable acquisition area can be updated in the abnormal acquisition image to facilitate the subsequent determination of the number of pixels in the corresponding area.

[0237] It can be understood that updating the cable acquisition area in the abnormal acquisition image facilitates determining the cable image area in the abnormal acquisition image, thereby obtaining the total number of corresponding pixel points, so as to facilitate the subsequent determination of the corresponding proportion of each problem, thereby making the calculated looseness analysis value more accurate and improving the accuracy of the detection results.

[0238] The total number of regions is the total number of pixels in the cable acquisition area within the abnormal acquisition image.

[0239] It is not difficult to understand that the image can be recognized and processed by the existing OpenCV recognition technology to obtain the number of pixels.

[0240] S43, counting the number of corroded pixel points in the abnormally collected image based on the corroded pixel value to obtain the regional corrosion number, and obtaining the corrosion proportion according to the ratio of the regional corrosion number to the total number of the region.

[0241] It can be understood that the regional corrosion number is the number of corroded pixels in the abnormal acquisition image, and the corrosion ratio is the ratio of the regional corrosion number to the total number of regions.

[0242] It is not difficult to understand that the OpenCV recognition technology is used to count the number of pixels belonging to the corrosion pixel value in the abnormal acquisition image, so that it can be compared with the total number of areas, and then the corrosion ratio of the corrosion in the image can be obtained, which is convenient for the subsequent calculation of the accurate looseness analysis value corresponding to the target cable.

[0243] S44, binarizing the abnormal acquisition image to obtain a cable identification image, counting the number of white pixels in the cable acquisition area in the cable identification image to obtain the number of regional defects, and obtaining the defect ratio based on the ratio of the number of regional defects to the total number of regions.

[0244] It can be understood that when the abnormal acquisition image is binarized, that is, Figure 4 As shown, the pixel values ​​of the normal cable and the pixel values ​​of the corrosion on the cable are binarized into black pixel values, and the pixel values ​​of the normal cable and the corrosion on the cable are binarized into white pixel values. Therefore, when there are white pixels in the cable acquisition area, it can be said that the black pixel values ​​that should have corresponded to the cable image have become white, and the corresponding white pixels are the pixels of the defective area of ​​the target cable. Therefore, the number of white pixels is counted to obtain the number of regional defects, and then the defect ratio of the corresponding target cable can be obtained.

[0245] Among them, the number of regional defects is the number of white pixels in the cable collection area in the cable recognition image, and the defect ratio is the ratio of the number of regional defects to the total number of regions.

[0246] S45, determining the remaining area outside the cable collection area in the cable recognition image as a burr judgment area.

[0247] It is understandable that when the target cable has a burr problem, the corresponding pixel value of the target cable will appear outside the cable acquisition area, so that the remaining area outside the cable acquisition area in the cable recognition image can be used as the burr judgment area.

[0248] Through the above implementation, the present invention can determine the burr judgment area, so as to subsequently count the number of corresponding pixel points, thereby determining the burr ratio, so as to subsequently obtain accurate loose analysis values, thereby determining the loose identification result.

[0249] S46, counting the number of black pixels in the burr determination area in the cable recognition image to obtain the number of regional burrs, and obtaining the burr ratio according to the ratio of the number of regional burrs to the total number of regional burrs.

[0250] It can be understood that the number of regional burrs is the number of black pixels in the burr judgment area in the cable recognition image, and the burr ratio is the ratio of the number of regional burrs to the total number of regional burrs.

[0251] Through the above implementation, the present invention obtains the burr ratio, so as to calculate and obtain the looseness analysis value in the subsequent step, thereby determining the looseness identification result of the target cable.

[0252] S47, calculating according to the abnormal image proportion, corrosion proportion, defect proportion and burr proportion to obtain a looseness analysis value corresponding to the target cable.

[0253] It can be understood that by calculating the abnormal image ratio, corrosion ratio, defect ratio and burr ratio, the looseness analysis value corresponding to the target cable is obtained.

[0254] In some embodiments, the specific implementation of step S47 (the calculation is performed according to the abnormal image proportion, corrosion proportion, defect proportion and burr proportion to obtain the looseness analysis value corresponding to the target cable) includes:

[0255] S471, the looseness analysis value is obtained by the following formula,

[0256]

[0257] Among them, A is the looseness analysis value, n is the number of abnormal images, N is the number of collected images, k n is the image weight value, c is the number of regional corrosion, M is the total number of regions, k c is the corrosion weight value, d is the number of regional defects, k d is the defect weight value, f is the number of regional burrs, k f is the glitch weight value.

[0258] It can be understood that each proportion has a corresponding weight value. Since different types of abnormal problems have different degrees of influence on the loosening of the cable, the image weight value, corrosion weight value, defect weight value and burr weight value can be pre-set according to the actual degree of influence to calculate the loosening analysis value of the target cable.

[0259] It is not difficult to understand that when the number of abnormal images is larger, the corresponding abnormal image proportion is larger, which can indicate that the number of abnormal problems of the target cable is large, and when the number of regional corrosion, the number of regional defects and the number of regional burrs are larger, the corresponding corrosion proportion, defect proportion and burr proportion will also increase accordingly, and the obtained loosening analysis value will also increase accordingly, which can indicate that the abnormal problem of the target cable is serious and may cause loosening of the target cable. Conversely, when the number of regional corrosion, the number of regional defects and the number of regional burrs are smaller, the corresponding corrosion proportion, defect proportion and burr proportion will also decrease accordingly, and the obtained loosening analysis value will also decrease accordingly, which can indicate that the abnormal problem of the target cable is low and may not cause loosening of the target cable, thereby realizing automatic detection.

[0260] S48, when it is determined that the looseness analysis value is greater than the preset looseness value, a looseness identification result corresponding to the target cable is generated and sent to the maintenance end.

[0261] It can be understood that the preset looseness value is a reference value for determining whether the target cable is loose, and can be preset manually. The maintenance end is an information terminal of the maintenance personnel for the target cable, and can be a mobile phone, computer, etc.

[0262] It is not difficult to understand that when the calculated looseness analysis value is greater than the preset looseness value, it can be said that the current target cable may be loose, so that the corresponding looseness identification result can be sent to the maintenance end, so that the maintenance personnel can clearly view the identification result, and take maintenance measures in time to repair the target cable to ensure that the cable can be used safely.

[0263] See also Figure 5 , is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention, the electronic device 50 includes: a processor 51, a memory 52 and a computer program; wherein

[0264] The memory 52 is used to store the computer program, and the memory may also be a flash memory. The computer program is, for example, an application program, a functional module, etc. for implementing the above method.

[0265] The processor 51 is used to execute the computer program stored in the memory to implement each step performed by the device in the above method. For details, please refer to the relevant description in the above method embodiment.

[0266] Optionally, the memory 52 may be independent or integrated with the processor 51 .

[0267] When the memory 52 is a device independent of the processor 51, the device may further include:

[0268] The bus 53 is used to connect the memory 52 and the processor 51 .

[0269] The present invention also provides a readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided by the various embodiments described above.

[0270] Among them, the readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, abbreviated as: ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist in a communication device as discrete components. The readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0271] The present invention also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0272] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0273] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying loose cables based on data processing, characterized in that: include: In response to the looseness identification request, a comprehensive pre-inspection path is determined according to the cable properties of the target cable, and an image of the upper surface of the target cable is captured based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection captured images; Perform multi-dimensional anomaly recognition on the pre-checked captured image according to the loose anomaly recognition strategy to obtain an anomaly captured image corresponding to the corresponding upward-looking hovering point; Based on the abnormal acquisition image, sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable to obtain an abnormal focusing path, and acquiring an abnormal acquisition image corresponding to the abnormal focusing path; Calculating the abnormality ratio in the abnormal acquisition image to obtain a looseness analysis value corresponding to the target cable, and determining a looseness identification result of the target cable based on the looseness analysis value; The performing multi-dimensional anomaly recognition on the pre-checked captured image according to the loose anomaly recognition strategy to obtain an abnormal captured image corresponding to the corresponding upward-looking hovering point includes: According to the corrosion anomaly recognition strategy, corrosion anomaly recognition is performed on the pre-inspection acquisition image to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point; Based on the defect and abnormality recognition strategy, the pre-inspection acquisition image is subjected to defect and abnormality recognition to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point; Perform burr anomaly recognition on the pre-checked acquired image according to the burr anomaly recognition strategy to obtain an abnormal acquired image corresponding to the corresponding upward-looking hovering point; The calculating the abnormal proportion in the abnormal acquisition image to obtain a looseness analysis value corresponding to the target cable, and determining a looseness identification result of the target cable based on the looseness analysis value, includes: Counting the number of abnormal collected images to obtain the number of abnormal images, and obtaining the number of images collected by the collection device for the target cable as the number of collected images, and obtaining the proportion of abnormal images according to the proportion of the number of abnormal images and the number of collected images; The cable collection area is updated to the abnormal collection image, and the number of pixel points in the cable collection area in the abnormal collection image is counted to obtain the total number of areas; Counting the number of corroded pixel points in the abnormally collected image based on the corroded pixel value to obtain the number of regional corrosion, and obtaining the corrosion proportion according to the ratio of the number of regional corrosion to the total number of the region; Binarization is performed on the abnormal acquisition image to obtain a cable identification image, and the number of white pixels in the cable acquisition area in the cable identification image is counted to obtain the number of regional defects, and the defect ratio is obtained according to the ratio of the number of regional defects to the total number of regions; Determine the remaining area outside the cable collection area in the cable recognition image as a burr judgment area; Counting the number of black pixels in the burr determination area in the cable recognition image to obtain the number of regional burrs, and obtaining the burr ratio according to the ratio of the number of regional burrs to the total number of regional burrs; Calculate according to the abnormal image proportion, corrosion proportion, defect proportion and burr proportion to obtain a looseness analysis value corresponding to the target cable; When it is determined that the looseness analysis value is greater than the preset looseness value, a looseness identification result corresponding to the target cable is generated and sent to the maintenance end; The calculation is performed according to the abnormal image proportion, corrosion proportion, defect proportion and burr proportion to obtain a looseness analysis value corresponding to the target cable, including: The looseness analysis value is obtained by the following formula: in, To loosen the analysis value, is the number of abnormal images, is the number of collected images, is the image weight value, is the regional corrosion amount, is the total number of regions, is the corrosion weight value, is the number of regional defects, is the defect weight value, is the number of regional glitches, and is the glitches weight value.

2. The method according to claim 1, characterized in that: In response to the looseness identification request, a comprehensive pre-inspection path is determined according to the cable properties of the target cable, and an image of the upper surface of the target cable is captured based on the comprehensive pre-inspection path to obtain a plurality of pre-inspection captured images, including: In response to the looseness identification request, a cable collection area in a collection interface corresponding to the collection device and a collection length corresponding to the cable collection area are obtained, and cable properties of the target cable are retrieved, wherein the cable properties include cable length, cable height and cable angle; The number of acquisitions is obtained according to the ratio of the cable length to the acquisition length, and when it is determined that the number of acquisitions is an integer, the number of acquisitions is doubled to obtain the number of divisions; The cable height is evenly divided based on the number of divisions to obtain a plurality of height acquisition segments, the height distance of the height acquisition segments is acquired, and the height distance is doubled to obtain a decreasing distance; According to the difference between the cable height and the height distance, an initial height is obtained, and the initial height is successively decreased based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple acquisition heights; Based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain a plurality of upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point, wherein the upward-looking hovering point has a corresponding downward-looking hovering point; Counting the upward-looking hovering points in sequence along the target cable based on a first preset direction to obtain a comprehensive preview path corresponding to the target cable; Based on the comprehensive preview path, images of the upper surface of the target cable are sequentially captured to obtain preview capture images corresponding to each upward-looking hovering point in the comprehensive preview path.

3. The method according to claim 2, characterized in that Also includes: When it is determined that the acquisition times is not an integer, extracting an integer value from the acquisition times, and obtaining an evenly divided length according to the product of the integer value and the acquisition length; Obtaining a remaining length according to a difference between the cable length and the average length, and determining a remaining height based on the remaining length and the cable angle; Based on the difference between the cable height and the remaining height, an average height is obtained, and the integer value is doubled to obtain the number of divisions; Evenly dividing the average height based on the number of divisions to obtain a plurality of height acquisition segments, acquiring height distances of the height acquisition segments, and doubling the height distances to obtain decreasing distances; According to the difference between the cable height and the height distance, an initial height is obtained, and the initial height is successively decreased based on the decreasing distance to obtain a remaining distance, until the remaining distance is less than 0, to obtain multiple acquisition heights; Based on the acquisition height and the initial height as the positioning height, determine the positioning point on the target cable according to the positioning height, and obtain a plurality of upward-looking hovering points and downward-looking hovering points corresponding to the target cable based on the preset height and the positioning point; Determine an upper positioning area in the cable collection area based on the remaining length, taking the upper boundary line of the cable collection area as a starting line; Determine an outer identification area of ​​an upper surface and an inner identification area of ​​a lower surface of the target cable based on the remaining length and a second preset direction; When the upper positioning area corresponding to the acquisition device overlaps with the outer identification area, the current hovering point is obtained as the upward-looking hovering point; When the upper positioning area corresponding to the acquisition device overlaps with the inner identification area, the current hovering point is obtained as the downward-looking hovering point, and the upward-looking hovering point has a corresponding downward-looking hovering point; The upward-looking hovering points are counted sequentially along the target cable based on a first preset direction to obtain a comprehensive preview path corresponding to the target cable.

4. The method according to claim 1, characterized in that The step of performing corrosion anomaly recognition on the pre-inspection acquisition image according to the corrosion anomaly recognition strategy to obtain an abnormal acquisition image corresponding to the corresponding upward-looking hovering point includes: Retrieving a corrosion pixel value, and determining a corrosion pixel point in the pre-check acquisition image based on the corrosion pixel value; The number of corroded pixel points in the pre-check acquisition image is counted to obtain the corrosion number, and when it is determined that the corrosion number is greater than the preset corrosion number, the pre-check acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

5. The method according to claim 1, characterized in that The defect and abnormality recognition strategy is used to identify the defect and abnormality of the pre-inspection acquisition image to obtain the abnormal acquisition image corresponding to the corresponding upward-looking hovering point, including: Acquire the image center point of the pre-check acquisition image, construct a coordinate system based on the image center point, and obtain the coordinates of the pixel points corresponding to the pre-check acquisition image; Binarization is performed on the corresponding pre-check acquisition image to obtain a cable recognition image, the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area are obtained as the surface horizontal coordinate, and the horizontal coordinate between the maximum horizontal coordinate and the minimum horizontal coordinate is used as the internal horizontal coordinate; Constructing a surface vertical line perpendicular to the horizontal axis of the coordinate system according to the surface horizontal coordinate, and when determining that there are non-adjacent black pixels on the surface vertical line in the cable recognition image, sequentially counting the coordinates of the black pixels on each of the surface vertical lines to obtain a defective coordinate set; Extracting coordinates corresponding to two adjacent black pixel points whose ordinates are discontinuous in the defect coordinate set as defect judgment coordinates; Obtaining a defect judgment value according to the absolute value of the vertical coordinate difference in the defect judgment coordinate, and obtaining the number of the defect judgment values ​​to obtain the defect number; When it is determined that the defect judgment value is greater than a preset missing value, or the defect quantity is greater than a preset defect quantity, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image; An internal vertical line perpendicular to the horizontal axis of the coordinate system is constructed according to the internal horizontal coordinate. When it is determined that there are non-adjacent black pixels on the internal vertical line in the cable recognition image, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

6. The method according to claim 1, characterized in that The step of performing burr anomaly recognition on the pre-checked acquired image according to the burr anomaly recognition strategy to obtain an abnormal acquired image corresponding to the corresponding upward-looking hovering point includes: Acquire the image center point of the pre-check acquisition image, construct a coordinate system based on the image center point, and obtain the coordinates of the pixel points corresponding to the pre-check acquisition image; Binarization is performed on the corresponding pre-check acquisition image to obtain a cable recognition image, and a standard width is obtained according to the maximum horizontal coordinate and the minimum horizontal coordinate corresponding to the cable acquisition area; Based on the same vertical coordinate, the maximum horizontal coordinate corresponding to the black pixel point in the cable recognition image is obtained as the first horizontal coordinate, and the minimum horizontal coordinate is obtained as the second horizontal coordinate; The actual width is obtained according to the absolute value of the difference between the first horizontal coordinate and the second horizontal coordinate, and when it is determined that the actual width is greater than the standard width, the pre-inspection acquisition image corresponding to the upward-looking hovering point is used as an abnormal acquisition image.

7. The method according to claim 3, characterized in that The step of sequentially counting downward-looking hovering points corresponding to the lower surface of the target cable based on the abnormal acquisition image to obtain an abnormal focusing path includes: Based on the abnormal acquisition image, determine the corresponding upward-looking hovering point as the upward-looking abnormal point, and determine the corresponding downward-looking hovering point as the direct downward-looking acquisition point according to the upward-looking abnormal point; Acquire a downward-looking hovering point adjacent to the direct downward-looking acquisition point as an indirect downward-looking acquisition point, and obtain a downward-looking acquisition point based on the direct downward-looking acquisition point and the indirect downward-looking acquisition point; The downward-looking acquisition points corresponding to the lower surface of the target cable are counted in sequence according to the second preset direction to obtain an abnormal focusing path.

Images