Crack detection method for power equipment, computer equipment and storage medium

Abstract

The present invention provides a method, device, computer equipment and storage medium for detecting cracks in electric equipment. The method comprises: displaying original image data collected by a camera on the detection interface, adjusting the camera according to shooting parameters, loading a calibration frame on the detection interface, generating target image data when the calibration frame overlaps with a two-dimensional code, segmenting the semantics of each element in the electric equipment in the target image data as feature image data, calculating the difference between the feature image data of this cycle and the feature image data of the previous cycle, obtaining incremental cracks, determining that the incremental cracks are newly added cracks in the electric equipment or cracks enlarged on the original cracks, filling the newly added cracks with a new color, and filling the cracks enlarged on the original cracks with a color different from the original cracks, so as to ensure the accuracy of detecting incremental cracks with a small error, so as to accurately evaluate the status of the electric equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power, and in particular to a crack detection method for electric power equipment, a computer device and a storage medium. Background Art

[0002] Many power equipment such as transformers and power boxes are deployed outdoors. During use, due to the influence of temperature, humidity, foundation settlement and other factors, the power equipment will age significantly and cracks will easily appear in the casing, fixing columns and other places.

[0003] Cracks are a sign of the status of power equipment. Monitoring the status of cracks can assist in evaluating the status of power equipment. Currently, crack monitoring mainly relies on manual measurement and recording by technicians during inspections, which results in large errors. Summary of the invention

[0004] The present invention proposes a crack detection method, device, computer equipment and storage medium for electric power equipment to solve the problem of large errors in crack monitoring of electric power equipment.

[0005] In a first aspect, an embodiment of the present invention provides a crack detection method for electric power equipment, comprising:

[0006] In each cycle, the original image data collected by the camera on the power equipment is displayed on the detection interface, and the original image data has a QR code;

[0007] Convert the two-dimensional code into a code and an angle code of the electric power equipment, wherein in a non-first cycle, the code and the angle code are used to map shooting parameters;

[0008] adjusting the camera according to the shooting parameters, loading a calibration frame on the detection interface, and generating target image data when the calibration frame overlaps with the QR code;

[0009] Inputting the target image data into a preset U-shaped network, segmenting the semantics of each element in the power equipment in the target image data as feature image data;

[0010] Calculate the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle to obtain the incremental cracks;

[0011] Comparing the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks that are enlarged on the original cracks;

[0012] In the target image data of this cycle, newly added cracks are filled with a new color, and cracks that are enlarged on the original cracks are filled with a color different from the original cracks.

[0013] In a second aspect, an embodiment of the present invention further provides a crack detection device for electric power equipment, comprising:

[0014] A preview module is used to display the original image data collected by the camera on the power equipment on the detection interface in each cycle, wherein the original image data has a QR code;

[0015] A two-dimensional code conversion module, used for converting the two-dimensional code into a code and an angle code of the electric power equipment, wherein in a non-first cycle, the code and the angle code are jointly mapped to a shooting parameter;

[0016] A parameter shooting module, used for adjusting the camera according to the shooting parameters, loading a calibration frame on the detection interface, and generating target image data when the calibration frame overlaps with the QR code;

[0017] A semantic recognition module, used for inputting the target image data into a preset U-shaped network, and segmenting the semantics of each element in the power equipment in the target image data as feature image data;

[0018] An incremental crack detection module, used to calculate the difference between the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain incremental cracks;

[0019] A crack comparison module, used for comparing the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks enlarged on the original cracks;

[0020] The color filling module is used to fill the newly added cracks with new colors in the target image data of this cycle, and fill the cracks that are enlarged on the original cracks with a color different from the original cracks.

[0021] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising:

[0022] one or more processors;

[0023] a memory for storing one or more programs,

[0024] When the one or more programs are executed by the one or more processors, the one or more processors implement the crack detection method for electric power equipment as described in the first aspect.

[0025] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for detecting cracks in an electric power device as described in the first aspect is implemented.

[0026] In this embodiment, in each cycle, the original image data collected by the camera on the power equipment is displayed on the detection interface. The original image data contains a QR code, and the QR code is converted into the code and angle code of the power equipment. In a non-first cycle, the code and the angle code are jointly mapped to the shooting parameters, and the camera is adjusted according to the shooting parameters. A calibration frame is loaded on the detection interface, and target image data is generated when the calibration frame overlaps with the QR code. The target image data is input into a preset U-shaped network, and the semantics of each element in the power equipment are segmented in the target image data as feature image data. The difference between the feature image data of this cycle and the feature image data of the previous cycle is calculated to obtain incremental cracks, and the incremental cracks are compared with the incremental cracks of the previous cycle. The characteristic image data are compared to determine whether the incremental cracks are new cracks in the power equipment or cracks that are enlarged on the original cracks. In the target image data of this period, the new cracks are filled with a new color, and the cracks that are enlarged on the original cracks are filled with a color different from the original cracks. This embodiment fixes the shooting parameters when shooting, thereby achieving information unification between target image data of different periods, thereby avoiding information discarding caused by shooting image data of power equipment at different angles, and avoiding information loss caused by later alignment of image data using methods such as magnification, reduction, and coordinate system projection. Therefore, the accuracy of detecting incremental cracks can be guaranteed with small errors, so as to accurately evaluate the status of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flow chart of a crack detection method for electric power equipment provided in Embodiment 1 of the present invention;

[0028] Figure 2 is a flow chart of a crack detection method for electric power equipment provided by Embodiment 2 of the present invention;

[0029] Figure 3 A schematic diagram of the structure of a crack detection device for electric power equipment provided in Embodiment 3 of the present invention;

[0030] Figure 4 A schematic diagram of the structure of a computer device provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0031] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0032] Embodiment 1

[0033] Figure 1 This is a flow chart of a crack detection method for electric power equipment provided in Embodiment 1 of the present invention. This embodiment can be applied to the case where image data of electric power equipment is collected in a predetermined shooting mode and incremental cracks are calculated. The method can be executed by a crack detection device for electric power equipment. The crack detection device for electric power equipment can be implemented by software and / or hardware and can be configured in a computer device, for example, a mobile terminal (such as a mobile phone, a tablet computer, etc.), a smart wearable device (such as a smart watch, smart glasses, etc.), etc., and specifically includes the following steps:

[0034] Step 101: In each cycle, the original image data collected by the camera from the power equipment is displayed on the detection interface.

[0035] In this embodiment, the technician carries a computer device to inspect the power equipment in each inspection cycle, and the cycle can be regular or irregular.

[0036] Among them, the power equipment has been equipped with a QR code recorded in a physical object such as a nameplate. Therefore, when shooting, the original image data contains the QR code (that is, the pixel point with the content of the QR code).

[0037] The QR code is composed of the code of the power equipment (such as ID) and the angle code. The angle code indicates the angle at which cracks in the power equipment are photographed. Since some power equipment is large-scale, there may be multiple cracks. Different angles can be set for different cracks. That is, the code of the same power equipment corresponds to one or more angle codes, and the same angle can accommodate one or more cracks.

[0038] In a non-first cycle, the technician starts the detection program and displays a user interface (UI), which is recorded as the detection interface. The detection program calls the camera on the computer device to collect raw image data of the power equipment, and the raw image data is displayed in the form of a preview on the detection interface.

[0039] Step 102: Convert the QR code into the code and angle code of the power equipment.

[0040] In this embodiment, the QR code in the original image data is decoded according to the agreed decoding method to obtain a set of character strings, which are then disassembled according to the agreed combination method to obtain the code and angle code of the power equipment. For example, the length of the character string is 10 bits, the first 9 bits are the code of the power equipment, and the last bit is the angle code.

[0041] In this embodiment, in a non-first cycle, the code and the angle code jointly map the shooting parameters, wherein the shooting parameters are parameters recorded in a mode of shooting the electric power device at a specified angle.

[0042] After parsing the code and angle code, the computer device can search in the local cache or database whether there are shooting parameters that are mapped together with the code and angle code. If so, the shooting parameters are extracted from the local cache or database. If not, the computer device requests the shooting parameters that are mapped together with the code and angle code from the server, and stores the mapping relationship between the code, angle code and shooting parameters in the local cache or database for subsequent use.

[0043] Step 103: adjust the camera according to the shooting parameters, load the calibration frame on the detection interface, and generate target image data when the calibration frame overlaps the QR code.

[0044] The shooting parameters are parameters recorded when the image data of the cracks of the power equipment were previously collected at the same angle. The shooting parameters are parsed and, during the preview process, some shooting parameters related to the camera are sent to the camera driver. The driver adjusts the camera according to the shooting parameters and loads a calibration frame on the detection interface according to the shooting parameters. The user moves the computer device according to the calibration frame so that the calibration frame overlaps with the QR code. When the calibration frame overlaps with the QR code, it indicates that the current state is the same as the state of the previous image data collection of the cracks of the power equipment. The shooting operation can be performed to generate target image data.

[0045] In a specific implementation, the focal length, the size and coordinates of the calibration frame, and the focus can be read from the shooting parameters, wherein the focus represents the location of the original crack of the power equipment, that is, the crack is a crack previously detected on the power equipment, and the crack belongs to an area, and the average value, midpoint and other values ​​of the area can be calculated as the location of the focus.

[0046] During the preview process, the camera lens is adjusted according to the focal length, and the calibration frame is loaded according to the size on the coordinates of the detection interface. The QR code has positioning points and the calibration frame has calibration points. The positioning points are asymmetrical, which can reflect the direction of the QR code. Correspondingly, the calibration points are asymmetrical, which can reflect the direction of the calibration frame.

[0047] When the outline of the calibration frame is aligned with the outline of the two-dimensional code, and the calibration point is aligned with the positioning point, a focusing operation is performed on the focus point to generate target image data.

[0048] When the number of focal points is two or more, a focusing operation can be performed on each focus separately to generate two or more frames of target image data, and finally the two or more frames of target image data are merged into one frame of target image data that is clear at each focus.

[0049] It should be noted that focal length, calibration frame, and focus are all fixed shooting parameters and cannot be adjusted by the current user, while parameters such as white balance are dynamic shooting parameters and are automatically adjusted according to the actual shooting environment.

[0050] Then, a first touch operation for adjusting the focus is received on the detection interface, and the first touch operation is ignored so that the camera maintains the focus.

[0051] A second touch operation for adjusting the calibration frame is received on the detection interface, and the second touch operation is ignored so that the calibration frame maintains the coordinates and the size.

[0052] A third touch operation for adjusting the focus is received on the detection interface, and the third touch operation is ignored to maintain the focus point.

[0053] Of course, the above shooting parameters are only used as examples. When implementing the embodiment of the present invention, other shooting parameters can be set according to actual conditions, and the embodiment of the present invention does not limit this. In addition, in addition to the above shooting parameters, those skilled in the art can also use other shooting parameters according to actual needs, and the embodiment of the present invention does not limit this.

[0054] Step 104: input the target image data into a preset U-shaped network, and segment the semantics of each element in the power equipment in the target image data as feature image data.

[0055] In this embodiment, a U-net can be pre-trained, and U-net can be used to semantically segment the target image data. The target image data is input into a preset U-net, and the semantics of each element in the power equipment are segmented in the target image data as feature image data.

[0056] The segmentation target is the cracks on the power equipment.

[0057] The structure of U-Net is to encode (downsampling) first, then decode (upsampling), and regress to the classification of pixels of the same size as the target image data.

[0058] U-Net can combine low-level information with high-level information:

[0059] Bottom layer (deep layer) information: low-resolution information after multiple downsampling can provide contextual semantic information of the segmented target (crack) in the entire target image data, which can be understood as a feature reflecting the relationship between the target (crack) and its environment. This feature helps to determine the category of the object (so classification problems usually involve low-resolution / deep layer information, not multi-scale fusion).

[0060] High-level (shallow) information: High-resolution information that is directly passed from the encoder to the decoder at the same level through the concatenate operation, which can provide more refined features for segmentation, such as gradients.

[0061] U-Net combines low-resolution information (providing a basis for object category identification) and high-resolution information (providing a basis for accurate segmentation and positioning), and can be used to segment cracks on power equipment.

[0062] In addition, U-Net contains residual connections (skip connections), which are also called skip connections. Adding residual connections to U-Net can solve the problems of gradient disappearance and gradient explosion during training, and can effectively reduce the number of network layers, making training easier.

[0063] Step 105: Calculate the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle to obtain the incremental cracks.

[0064] In this embodiment, the characteristic image data of the previous cycle can be queried from the local database based on the code and angle code of the power equipment. If found, the characteristic image data of the previous cycle is loaded. If not found, the characteristic image data of the previous cycle can be requested from the server based on the code and angle code of the power equipment.

[0065] The characteristic image data of this cycle are compared with the characteristic image data of the previous cycle, and the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle is calculated. Since the power equipment changes little in a relatively short period of time, the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle can reflect the incremental cracks, that is, taking the cracks as the variable, the increase in the cracks between the characteristic image data of this cycle and the characteristic image data of the previous cycle.

[0066] In one embodiment of the present invention, step 105 may include the following steps:

[0067] Step 1051 : performing a differential operation on the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain differential image data.

[0068] In this embodiment, a differential operation is performed on the characteristic image data of the current cycle and the characteristic image data of the previous cycle, and the result of the differential operation is recorded as differential image data.

[0069] The so-called differential operation may refer to subtracting the pixel value of the characteristic image data of the previous cycle from the pixel value of the characteristic image data of the current cycle, so as to weaken the similar parts between the characteristic image data of the current cycle and the characteristic image data of the previous cycle, and highlight the changed parts between the characteristic image data of the current cycle and the characteristic image data of the previous cycle. In addition to the incremental cracks, the changed parts may also include other situations, such as paint peeling, stains, noise, etc.

[0070] Step 1052: Detect connected regions in the differential image data as candidate objects.

[0071] Considering that the cracks are connected internally, a connected component analysis can be performed on the differential image data to detect the connected component in the differential image data and mark the connected component as a candidate object.

[0072] Among them, the connected region generally refers to an image region (Region, Blob) composed of foreground pixels with the same pixel value and adjacent positions in the differential image data, that is, the cracks are regarded as foreground targets and extracted so that the connected region analysis method can be used in subsequent processing application scenarios.

[0073] Step 1053: Extract the geometric features of the candidate object.

[0074] Step 1054: If the feature meets the preset target condition, the candidate object is determined to be an incremental crack.

[0075] In this embodiment, the cracks may be analyzed in advance to find out the geometric characteristics of the cracks, so as to set target conditions for the geometric characteristics of the cracks. If the target conditions are met, the cracks may be considered as cracks.

[0076] Then, for the candidate object, the geometric features of the candidate object can be extracted according to the requirements of the target condition, and the geometric features of the candidate object can be compared with the preset target condition.

[0077] If the geometric features of the candidate object meet the target condition, the candidate object can be determined to be an incremental crack.

[0078] If the geometric features of the candidate object do not satisfy the target condition, it can be determined that the candidate object is not an incremental crack.

[0079] For example, the total number of pixels in the candidate object may be counted as a geometric feature.

[0080] Any two pixel points on the edge of the candidate object are connected as a first candidate line segment, and the first candidate line segment with the largest length is taken as the first target line segment.

[0081] In a direction perpendicular to the first target line segment, any two pixel points on the edge of the candidate object are connected to form a second candidate line segment, and the second candidate line segment with the largest length is taken as the second target line segment.

[0082] The ratio between the first target line segment and the second target line segment is calculated as a geometric feature.

[0083] That is, the first target line segment is the longest distance of the candidate object in any direction, the second target line segment is perpendicular to the first target line segment, and the ratio between the first target line segment and the second target line segment can reflect the shape of the candidate object.

[0084] In this example, the target condition is that the total number of pixels is greater than or equal to a preset first threshold, and the ratio is greater than or equal to a preset second threshold.

[0085] Then, on the one hand, the total number of pixels is compared with a preset first threshold, and on the other hand, the ratio is compared with a preset second threshold.

[0086] If the total number of pixels is less than the first threshold, it means that the candidate object has fewer pixels and may be noise points.

[0087] If the total number of pixel points is greater than or equal to the first threshold, it means that the candidate object has more pixel points and may be an object with practical significance.

[0088] If the ratio is less than the second threshold, it means that the shape of the candidate object is not long and narrow and does not match the shape of the crack.

[0089] If the ratio is greater than or equal to the second threshold, it means that the shape of the candidate object is long and narrow, which matches the shape of the crack.

[0090] Then, if the total number is smaller than a preset first threshold, or the ratio is smaller than a preset second threshold, it is determined that the candidate object is not an incremental crack.

[0091] If the total number is greater than or equal to a preset first threshold, and the ratio is greater than or equal to a preset second threshold, the candidate object is determined to be an incremental crack.

[0092] Step 106: Compare the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks that are enlarged from the original cracks.

[0093] Since the feature image data of the previous cycle has calibrated the semantics of each element and detected the cracks of the previous cycle, the incremental cracks of this cycle can be compared with the feature image data of the previous cycle to determine whether the incremental cracks of this cycle are new cracks in the power equipment or cracks that are enlarged on the original cracks.

[0094] In a specific implementation, the original cracks of the power equipment (ie, the incremental cracks detected in the previous cycle) are queried in the feature image data of the previous cycle.

[0095] Calculate the distance between the incremental cracks of this cycle and the original cracks of the previous cycle.

[0096] If the distance is less than or equal to the preset third threshold, it means that the distance between the incremental cracks in this cycle and the original cracks in the previous cycle is small and is an error, and it is determined that the incremental cracks are cracks in the power equipment that are enlarged on the original cracks.

[0097] If the distance is greater than the preset third threshold, it means that the distance between the incremental cracks in this cycle and the original cracks in the previous cycle is large and is not an error, and it is determined that the incremental cracks are newly added cracks in the power equipment.

[0098] Step 107 , in the target image data of this cycle, newly added cracks are filled with a new color, and cracks enlarged on the original cracks are filled with a color different from the original cracks.

[0099] In the target image data of this cycle, the newly added cracks are filled with new colors so that the user can intuitively see the newly added cracks.

[0100] In the target image data of this cycle, the cracks that expand on the original cracks are filled with a color different from the original cracks, that is, for the same crack, each expansion of the crack is filled with a different color, allowing users to distinguish the development of the cracks by color, and thus determine the status of the power equipment.

[0101] In this embodiment, in each cycle, the original image data collected by the camera on the power equipment is displayed on the detection interface. The original image data contains a QR code, and the QR code is converted into the code and angle code of the power equipment. In a non-first cycle, the code and the angle code are jointly mapped to the shooting parameters, and the camera is adjusted according to the shooting parameters. A calibration frame is loaded on the detection interface, and target image data is generated when the calibration frame overlaps with the QR code. The target image data is input into a preset U-shaped network, and the semantics of each element in the power equipment are segmented in the target image data as feature image data. The difference between the feature image data of this cycle and the feature image data of the previous cycle is calculated to obtain incremental cracks, and the incremental cracks are compared with the incremental cracks of the previous cycle. The characteristic image data are compared to determine whether the incremental cracks are new cracks in the power equipment or cracks that are enlarged on the original cracks. In the target image data of this period, the new cracks are filled with a new color, and the cracks that are enlarged on the original cracks are filled with a color different from the original cracks. This embodiment fixes the shooting parameters when shooting, thereby achieving information unification between target image data of different periods, thereby avoiding information discarding caused by shooting image data of power equipment at different angles, and avoiding information loss caused by later alignment of image data using methods such as magnification, reduction, and coordinate system projection. Therefore, the accuracy of detecting incremental cracks can be guaranteed with small errors, so as to accurately evaluate the status of power equipment.

[0102] Embodiment 2

[0103] Figure 2 This is a flow chart of a crack detection method for power equipment provided in Embodiment 2 of the present invention. This embodiment is based on the above embodiment and further adds an operation of detecting cracks in the first cycle. The method specifically includes the following steps:

[0104] Step 201: In each cycle, the original image data collected by the camera from the power equipment is displayed on the detection interface.

[0105] In the first cycle, the technicians recorded the QR code on physical objects such as nameplates and assigned it to the power equipment. The technicians started the detection program and displayed the detection interface. The detection program called the camera on the computer device to collect raw image data of the power equipment, and the raw image data was displayed in the form of a preview on the detection interface.

[0106] Step 202: Convert the QR code into the code and angle code of the power equipment.

[0107] In this embodiment, the original image data contains a QR code (i.e., pixels whose content is a QR code). The QR code in the original image data is decoded according to an agreed decoding method to obtain a set of character strings. The character strings are disassembled according to an agreed combination method to obtain the code and angle code of the power equipment. For example, the length of the character string is 10 bits, the first 9 bits are the code of the power equipment, and the last bit is the angle code.

[0108] Step 203: In the first cycle, in response to a determination operation received on the detection interface, target image data is generated.

[0109] The user moves the computer device and adjusts the shooting parameters so that the cracks on the power equipment can be clearly seen on the detection interface. The cracks can be used for later observation. At this time, the confirmation operation is triggered on the detection interface, and the detection program executes the shooting operation to generate target image data.

[0110] Step 204: Generate shooting parameters for the target image data, and establish a relationship in which the code and the angle code are mapped to the shooting parameters.

[0111] In this embodiment, the parameters when the target image data is generated are recorded as shooting parameters, and a mapping relationship between the code, angle code and shooting parameters is established, recorded in a local database and uploaded to the server.

[0112] Exemplarily, there are positioning points in the QR code. In this example, on the one hand, the focal length of the camera used to generate the target image data can be recorded. On the other hand, the size and coordinates of the outline of the QR code can be measured as the size and coordinates of the calibration frame, and the positioning points of the QR code can be queried as the calibration points in the calibration frame. On the other hand, the position of the crack in the target image data can be queried as the focus.

[0113] Among them, for cracks in the target image data, the focus can be supplemented after the cracks are detected.

[0114] Step 205: In a non-first cycle, search for the code and the angle code to map the shooting parameters together.

[0115] Step 206: adjust the camera according to the shooting parameters, load the calibration frame on the detection interface, and generate target image data when the calibration frame overlaps the QR code.

[0116] Step 207: input the target image data into a preset U-shaped network, and segment the semantics of each element in the power equipment in the target image data as feature image data.

[0117] Step 208: Calculate the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle to obtain the incremental cracks.

[0118] Step 209: Compare the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks that are enlarged from the original cracks.

[0119] Step 210: In the target image data of this cycle, newly added cracks are filled with a new color, and cracks that are enlarged on the original cracks are filled with a color different from the original cracks.

[0120] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0121] Embodiment 3

[0122] Figure 3 A structural block diagram of a crack detection device for electric power equipment provided in Embodiment 3 of the present invention may specifically include the following modules:

[0123] The preview module 301 is used to display the original image data collected by the camera on the power equipment on the detection interface in each cycle, wherein the original image data has a QR code;

[0124] A two-dimensional code conversion module 302, used to convert the two-dimensional code into a code and an angle code of the electric power device, and in a non-first cycle, the code and the angle code are jointly mapped to shooting parameters;

[0125] A parameter shooting module 303, used for adjusting the camera according to the shooting parameters, loading a calibration frame on the detection interface, and generating target image data when the calibration frame overlaps with the QR code;

[0126] The semantic recognition module 304 is used to input the target image data into a preset U-shaped network and segment the semantics of each element in the power equipment in the target image data as feature image data;

[0127] The incremental crack detection module 305 is used to calculate the difference between the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain incremental cracks;

[0128] A crack comparison module 306 is used to compare the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks that are enlarged on the original cracks;

[0129] The color filling module 307 is used to fill the newly added cracks with new colors in the target image data of this cycle, and fill the cracks that are enlarged on the original cracks with a color different from the original cracks.

[0130] In one embodiment of the present invention, the two-dimensional code has a positioning point;

[0131] The shooting module 303 is also used for:

[0132] Reading the focal length, the size and coordinates of the calibration frame, and the focus from the shooting parameters, wherein the focus indicates the location of the original crack of the power equipment;

[0133] Adjusting the lens of the camera according to the focal length;

[0134] Loading the calibration frame according to the size on the coordinates of the detection interface, wherein the calibration frame has calibration points;

[0135] When the outline of the calibration frame is aligned with the outline of the two-dimensional code, and the calibration point is aligned with the positioning point, a focusing operation is performed on the focus to generate target image data.

[0136] In one embodiment of the present invention, the photographing module 303 is further used for:

[0137] Receiving a first touch operation for adjusting the focal length on the detection interface;

[0138] Ignore the first touch operation to enable the camera to maintain the focal length;

[0139] Receiving a second touch operation for adjusting the calibration frame on the detection interface;

[0140] Ignore the second touch operation so that the calibration frame maintains the coordinates and the size;

[0141] receiving a third touch operation for adjusting the focus on the detection interface;

[0142] The third touch operation is ignored to maintain the focus on the focal point.

[0143] In one embodiment of the present invention, the incremental crack detection module 305 is further used to:

[0144] Performing a difference operation on the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain differential image data;

[0145] detecting connected regions in the differential image data as candidate objects;

[0146] Extracting geometric features of the candidate object;

[0147] If the feature meets the preset target condition, the candidate object is determined to be an incremental crack.

[0148] In one embodiment of the present invention, the incremental crack detection module 305 is further used to:

[0149] Counting the total number of pixels in the candidate object;

[0150] Connecting any two pixel points on the edge of the candidate object as a first candidate line segment;

[0151] Taking the first candidate line segment with the largest length as the first target line segment;

[0152] In a direction perpendicular to the first target line segment, connecting any two pixel points on the edge of the candidate object to form a second candidate line segment;

[0153] Taking the second candidate line segment with the largest length as the second target line segment;

[0154] calculating a ratio between the first target line segment and the second target line segment;

[0155] If the feature satisfies a preset target condition, determining that the candidate object is an incremental crack includes:

[0156] If the total number is greater than or equal to a preset first threshold, and the ratio is greater than or equal to a preset second threshold, the candidate object is determined to be an incremental crack.

[0157] In one embodiment of the present invention, the crack comparison module 306 is further used to:

[0158] Searching the characteristic image data of the previous cycle for original cracks in the power equipment;

[0159] Calculating the distance between the incremental crack and the original crack;

[0160] If the distance is less than or equal to a preset third threshold, determining that the incremental crack is a crack in the power equipment that is enlarged on the original crack;

[0161] If the distance is greater than a preset third threshold, it is determined that the incremental crack is a newly added crack in the power equipment.

[0162] In one embodiment of the present invention, it also includes:

[0163] a custom shooting module, for generating target image data in response to a determination operation received on the detection interface in a first cycle;

[0164] The shooting parameter generating module is used to generate shooting parameters for the target image data and establish a relationship in which the code and the angle code are mapped to the shooting parameters.

[0165] In one embodiment of the present invention, the two-dimensional code has a positioning point; the shooting parameter generation module is further used to:

[0166] Recording the focal length of the camera when generating the target image data;

[0167] Measuring the size and coordinates of the outline of the two-dimensional code as the size and coordinates of the calibration frame;

[0168] Query the positioning point of the two-dimensional code as the calibration point in the calibration frame;

[0169] The position of the crack in the target image data is queried as the focus.

[0170] The crack detection device for electric equipment provided by the embodiment of the present invention can execute the crack detection method for electric equipment provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0171] Embodiment 4

[0172] Figure 4 A schematic diagram of the structure of a computer device provided in Embodiment 4 of the present invention. Figure 4 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 4 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0173] like Figure 4 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0174] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0175] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0176] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0177] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0178] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), may communicate with one or more devices that enable a user to interact with the computer device 12, and / or may communicate with any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the computer device 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the computer device 12, including, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0179] The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28, such as implementing the crack detection method for electric power equipment provided in the embodiment of the present invention.

[0180] Embodiment 5

[0181] Embodiment 5 of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the various processes of the above-mentioned crack detection method for power equipment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0182] Among them, computer-readable storage media may include, for example, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device.

[0183] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A crack detection method for power equipment, characterized in that: include: In each cycle, the original image data collected by the camera on the power equipment is displayed on the detection interface, and the original image data has a QR code; Convert the two-dimensional code into a code and an angle code of the electric power equipment, wherein in a non-first cycle, the code and the angle code are used to map shooting parameters; adjusting the camera according to the shooting parameters, loading a calibration frame on the detection interface, and generating target image data when the calibration frame overlaps with the QR code; Inputting the target image data into a preset U-shaped network, segmenting the semantics of each element in the power equipment in the target image data as feature image data; Calculate the difference between the characteristic image data of this cycle and the characteristic image data of the previous cycle to obtain the incremental cracks; Comparing the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks that are enlarged on the original cracks; In the target image data of this cycle, newly added cracks are filled with a new color, and cracks that are enlarged on the original cracks are filled with a color different from the original cracks.

2. The method according to claim 1, characterized in that The two-dimensional code has a positioning point; The step of adjusting the camera according to the shooting parameters, loading a calibration frame on the detection interface, and generating target image data when the calibration frame overlaps the two-dimensional code includes: Reading the focal length, the size and coordinates of the calibration frame, and the focus from the shooting parameters, wherein the focus indicates the location of the original crack of the power equipment; Adjusting the lens of the camera according to the focal length; Loading the calibration frame according to the size on the coordinates of the detection interface, wherein the calibration frame has calibration points; When the outline of the calibration frame is aligned with the outline of the two-dimensional code, and the calibration point is aligned with the positioning point, a focusing operation is performed on the focus to generate target image data.

3. The method according to claim 2, characterized in that The camera is adjusted according to the shooting parameters, a calibration frame is loaded on the detection interface, and target image data is generated when the calibration frame overlaps with the two-dimensional code, further comprising: Receiving a first touch operation for adjusting the focal length on the detection interface; Ignore the first touch operation to enable the camera to maintain the focal length; Receiving a second touch operation for adjusting the calibration frame on the detection interface; Ignore the second touch operation so that the calibration frame maintains the coordinates and the size; receiving a third touch operation for adjusting the focus on the detection interface; The third touch operation is ignored to maintain the focus on the focal point.

4. The method according to claim 1, characterized in that: The calculating the difference between the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain the incremental cracks includes: Performing a difference operation on the characteristic image data of the current cycle and the characteristic image data of the previous cycle to obtain differential image data; detecting connected regions in the differential image data as candidate objects; Extracting geometric features of the candidate object; If the feature meets the preset target condition, the candidate object is determined to be an incremental crack.

5. The method according to claim 4, characterized in that The step of extracting geometric features of the candidate object comprises: Counting the total number of pixels in the candidate object; Connecting any two pixel points on the edge of the candidate object as a first candidate line segment; Taking the first candidate line segment with the largest length as the first target line segment; In a direction perpendicular to the first target line segment, connecting any two pixel points on the edge of the candidate object to form a second candidate line segment; Taking the second candidate line segment with the largest length as the second target line segment; calculating a ratio between the first target line segment and the second target line segment; If the feature satisfies a preset target condition, determining that the candidate object is an incremental crack includes: If the total number is greater than or equal to a preset first threshold, and the ratio is greater than or equal to a preset second threshold, the candidate object is determined to be an incremental crack.

6. The method according to claim 1, characterized in that The step of comparing the incremental cracks with the characteristic image data of the previous cycle to determine whether the incremental cracks are newly added cracks in the power equipment or cracks enlarged from existing cracks includes: Searching the characteristic image data of the previous cycle for original cracks in the power equipment; Calculating the distance between the incremental crack and the original crack; If the distance is less than or equal to a preset third threshold, determining that the incremental crack is a crack in the power equipment that is enlarged on the original crack; If the distance is greater than a preset third threshold, it is determined that the incremental crack is a newly added crack in the power equipment.

7. The method according to any one of claims 1 to 6, characterized in that Also includes: In a first cycle, in response to a determination operation received on the detection interface, target image data is generated; Shooting parameters are generated for the target image data, and a relationship is established in which the code and the angle code are mapped to the shooting parameters.

8. The method according to claim 7, characterized in that The two-dimensional code has a positioning point; the generating of shooting parameters for the target image data includes: Recording the focal length of the camera when generating the target image data; Measuring the size and coordinates of the outline of the two-dimensional code as the size and coordinates of the calibration frame; Query the positioning point of the two-dimensional code as the calibration point in the calibration frame; The position of the crack in the target image data is queried as the focus.

9. A computer device, characterized in that: The computer device comprises: one or more processors; a memory for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the crack detection method for electric power equipment according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the crack detection method for electric power equipment according to any one of claims 1 to 8 is implemented.

Images