A data processing method, device and medium for pipeline anomaly detection

Abstract

The application relates to the technical field of pipeline anomaly detection, in particular to a data processing method and device for pipeline anomaly detection and a medium. The method comprises the following steps: for any candidate inspection path, obtaining the light output power of each inspection position according to the anomaly probability of each inspection position and a preset first mapping relationship; determining the priority value of each candidate inspection path according to the light output power of each inspection position on each candidate inspection path, and determining the candidate inspection path with the maximum priority value as a target inspection path; for the image collected by any inspection position on the target inspection path, correcting the image collected by the inspection position according to the light output power of the inspection position; and judging whether pipeline anomaly occurs at each inspection position on the target inspection path according to the corrected image of each inspection position on the target inspection path. The application can balance the power consumption of the inspection robot and the accuracy of the pipeline anomaly detection result.

Description

Technical Field

[0001] This invention relates to the field of pipeline anomaly detection technology, and in particular to a data processing method, equipment and medium for pipeline anomaly detection. Background Technology

[0002] Pipelines may exhibit abnormalities such as leaks and cracks due to factors like corrosion, external damage, and material aging. If these are not detected and addressed promptly, they could lead to serious safety accidents. Therefore, efficient and accurate anomaly detection in pipelines is of paramount importance.

[0003] Currently, pipeline anomaly detection technology based on inspection robots has been widely used, but the following problems exist in practical applications: If the light output power of the inspection robot is set to a fixed value, and the light output power is set too high, the light will consume a lot of power, which may lead to insufficient power in long inspection path scenarios, and will also affect the signal strength of the inspection robot; if the light output power is set too low, although the light consumes less power, in the case of the dark environment of the pipeline, the acquired images may not be sufficient to meet the needs of anomaly detection, resulting in inaccurate anomaly detection results. Summary of the Invention

[0004] The purpose of this invention is to provide a data processing method, device, and medium for pipeline anomaly detection, in order to solve the problems of high power consumption of the lights or inaccurate anomaly detection results when the light output power of the inspection robot is set to a fixed value in the prior art.

[0005] According to a first aspect of the present invention, a data processing method for pipeline anomaly detection is provided, the method comprising the following steps:

[0006] S100, obtain the set of candidate inspection paths for the target pipeline route; the set of candidate inspection paths includes several candidate inspection paths.

[0007] S200, for any candidate inspection path, obtain the light output power of each inspection position on the candidate inspection path according to the anomaly probability of each inspection position on the candidate inspection path and the preset first mapping relationship; the first mapping relationship includes the mapping relationship between the anomaly probability and the light output power; the light output power of any inspection position on any candidate inspection path is positively correlated with the anomaly probability of the inspection position.

[0008] S300 determines the priority value of each candidate inspection path based on the light output power of each inspection position on each candidate inspection path, and determines the candidate inspection path with the highest priority value as the target inspection path.

[0009] S400: For any image acquired at any inspection location on the target inspection path, the image acquired at that inspection location is corrected according to the light output power at that inspection location to obtain the corrected image of that inspection location.

[0010] S500 determines whether there is a pipeline abnormality at each inspection location on the target inspection path based on the corrected image of each inspection location on the target inspection path.

[0011] According to a second aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the above-described data processing method for pipeline anomaly detection.

[0012] According to a third aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data processing method for pipeline anomaly detection.

[0013] The present invention has at least the following beneficial effects:

[0014] In this invention, the light output power at each inspection location is related to the anomaly probability of that location. For inspection locations with a high anomaly probability, increasing the light output power enhances the illumination intensity, resulting in clearer and more detailed images, which is beneficial for accurate subsequent judgment of pipeline anomalies. For inspection locations with a low anomaly probability, reasonably controlling the light output power reduces power consumption and avoids energy waste. Thus, this invention balances the power consumption of the inspection robot with the accuracy of pipeline anomaly detection results. Furthermore, this invention corrects the images acquired at each inspection location based on the light output power, effectively improving image quality for anomaly judgment and reducing image distortion caused by improper lighting conditions. Using the corrected images for pipeline anomaly judgment allows for more accurate identification of pipeline anomalies, improving detection accuracy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of a data processing method for pipeline anomaly detection provided in Embodiment 1 of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] In the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance, nor are they used to describe a specific order or sequence.

[0019] Example 1:

[0020] According to the present invention, a data processing method for pipeline anomaly detection is provided, such as... Figure 1 As shown, the method includes the following steps:

[0021] S100, obtain the set of candidate inspection paths for the target pipeline route; the set of candidate inspection paths includes several candidate inspection paths.

[0022] As a specific implementation method, the spatial distribution of pipelines is obtained through a pipeline geographic information system (GIS) or 3D modeling. This, combined with the mobility capabilities of the inspection robot (such as turning radius and climbing angle), generates multiple possible inspection paths. The paths must cover all preset inspection locations and meet the robot's passage requirements. Those skilled in the art will understand that the process of obtaining candidate inspection paths is prior art and will not be described in detail here.

[0023] This embodiment uses the physical layout of the pipeline and the robot's motion constraints to exhaustively or heuristically generate candidate paths, ensuring the feasibility of the inspection path and laying the foundation for subsequent selection of the optimal path.

[0024] S200, for any candidate inspection path, obtain the light output power of each inspection position on the candidate inspection path according to the anomaly probability of each inspection position on the candidate inspection path and the preset first mapping relationship; the first mapping relationship includes the mapping relationship between the anomaly probability and the light output power; the light output power of any inspection position on any candidate inspection path is positively correlated with the anomaly probability of the inspection position.

[0025] As a specific implementation, the inspection robot carries a lighting fixture during the inspection process. In pipeline environments where lighting is often insufficient, the lighting fixture carried by the inspection robot provides active illumination, ensuring that the robot's onboard visual sensors (such as cameras) can clearly collect environmental data. In this embodiment, the light output power is also the output power of the lighting fixture carried by the inspection robot.

[0026] As a specific implementation, for each inspection position on each candidate path, determine its abnormal probability (such as the probability of corrosion and leakage) according to historical abnormal data, real-time monitoring data or experience. Through a preset first mapping relationship, convert the abnormal probability into the light output power of the inspection robot during the inspection process. The higher the abnormal probability, the greater the light output power.

[0027] As a specific implementation, the preset mapping relationship is: P = min(P min ×(1 + k×c), P max ), where P min and P max are the preset minimum light output power and maximum light output power respectively, P is the light output power, c is the abnormal probability, 0 < c < 1, k is the influence coefficient of the abnormal probability on the light output power, k > 0; min( ) is to take the minimum value. Among them, P min can meet the lower requirements for the light output power during image acquisition, and P max can meet the higher requirements for the light output power during image acquisition. Optionally, P min , P max and k are empirical values.

[0028] In this embodiment, for the positions with high abnormal probability, enhance the illumination power, and for the positions with low abnormal probability, reduce the illumination power. Thus, this embodiment realizes the intelligent allocation of lighting resources, can focus on high-risk areas, and improve the abnormal recognition rate.

[0029] S300. Determine the priority value of each candidate inspection path according to the light output power of each inspection position on each candidate inspection path, and determine the candidate inspection path with the largest priority value as the target inspection path.

[0030] As a specific implementation, S300 includes:

[0031] S310. Modify the initial power decay curve of the specified candidate inspection path according to the light output power of each inspection position on the specified candidate inspection path to obtain the target power decay curve of the specified candidate inspection path; the abscissa of the initial power decay curve is the inspection distance, and the ordinate of the initial power decay curve is the power of the inspection robot; the specified candidate path is any candidate inspection path in the candidate inspection path set.

[0032] As a specific implementation, the initial power decay curve is the relationship between the power of the inspection robot and the inspection distance under the preset light output power. S310 includes:

[0033] S311. Divide the specified candidate inspection path into several segments according to the inspection positions.

[0034] As a specific implementation method, the start and end points of each inspection location are used as dividing points, thereby enabling the division of a specified candidate inspection path. It should be understood that, for non-full coverage inspection (selective location inspection) scenarios, any segment obtained by division is either the segment corresponding to the inspection location or the segment not corresponding to the inspection location (i.e., the segment between two inspection locations); for full coverage inspection scenarios, any segment obtained by division is the segment corresponding to the inspection location.

[0035] S312, for any segment, obtain the power consumption of the segment based on the corresponding light output power; the power consumption of any segment is the product of the power consumed by the inspection robot per unit length moved under the preset light output power and the rate of change of the light output power of the segment, and the rate of change of the light output power of any segment is the ratio of the corresponding light output power of the segment to the preset light output power.

[0036] In this embodiment, if a segment corresponds to an inspection location, the light output power corresponding to that segment can be obtained according to the first mapping relationship; if a segment does not correspond to an inspection location, optionally, the light output power corresponding to that segment is P. min Or 0 (if the light output power of the inspection robot in non-inspection positions is set to P) min Then, set the light output power of the segment corresponding to the non-inspection position to P. min If the light output power of the inspection robot in non-inspection positions is set to 0, then the light output power of the corresponding segment in the non-inspection position is set to 0.

[0037] In this embodiment, the initial power decay curve is a pre-constructed curve that can be constructed experimentally. The initial power decay curve represents the relationship between the inspection robot's power consumption and inspection distance under a preset light output power. It reflects the impact of inspection distance on power consumption when the light output power is fixed at the preset power during the inspection process. As the inspection distance increases, the power consumption gradually decreases. The power consumption per unit distance moved by the inspection robot under the preset light output power can be obtained from the initial power decay curve. Optionally, the preset light output power is an empirical value, and the preset light output power is greater than or equal to P. min And less than or equal to P max .

[0038] S313: Starting from the starting point of the specified candidate inspection path, obtain the remaining battery power of the inspection robot when it reaches each segment.

[0039] In one specific implementation, the remaining circuit when the inspection robot reaches any segment is the difference between the initial power consumption of the inspection robot when it starts inspection and the total power consumption of all segments before that segment.

[0040] S314, obtain the target power decay curve of the specified candidate inspection path based on the remaining power when the inspection robot arrives at each segment.

[0041] As a specific implementation method, a coordinate system is established, with the horizontal axis representing the inspection distance and the vertical axis representing the battery level. Based on the remaining battery level when the inspection robot reaches each segment, the corresponding discrete points are marked in the coordinate system. Then, the discrete points adjacent to the inspection distance are connected by straight lines to obtain the target battery attenuation curve of the specified candidate inspection path.

[0042] S320: Obtain the actual signal strength at each inspection position on the specified candidate inspection path based on the target power attenuation curve, signal strength attenuation curve, and standard signal strength at each inspection position on the specified candidate inspection path; the horizontal axis of the signal strength attenuation curve is the power of the inspection robot, and the vertical axis of the signal strength attenuation curve is the signal strength attenuation rate; the standard signal strength at any inspection position on the specified candidate inspection path is the actual signal strength at that inspection position when the inspection robot is fully charged.

[0043] In this embodiment, the standard signal strength at any inspection location on the candidate inspection path is known and can be obtained experimentally.

[0044] In this embodiment, the signal strength attenuation curve is a pre-constructed curve that can be constructed experimentally. The signal strength attenuation curve reflects the influence of the power level on the signal strength attenuation rate. The lower the power level, the higher the signal strength attenuation rate may be.

[0045] As one specific implementation, S320 includes:

[0046] S321, obtain the cumulative inspection distance of the target inspection location in the specified candidate inspection path according to the specified candidate inspection path; the target inspection location is any inspection location on the specified candidate inspection path.

[0047] In one specific implementation, if the target inspection location is a distance, then the cumulative inspection distance corresponding to the midpoint of the target inspection location is determined as the cumulative inspection distance of the target inspection location in the specified candidate inspection path.

[0048] S322: Obtain the remaining power of the inspection robot when it reaches the target inspection position based on the cumulative inspection distance of the target inspection position in the specified candidate inspection path and the target power decay curve of the specified candidate inspection path.

[0049] S323, obtain the signal strength attenuation rate of the inspection robot at the target inspection position based on the remaining power and signal strength attenuation curve when the inspection robot reaches the target inspection position.

[0050] S324: Obtain the actual signal strength at the target inspection location based on the signal strength attenuation rate of the inspection robot at the target inspection location and the standard signal strength at the target inspection location.

[0051] As a specific implementation, the actual signal strength at the target inspection location is the difference between the standard signal strength at the target inspection location and the signal strength attenuation of the inspection robot at the target inspection location. The signal strength attenuation of the inspection robot at the target inspection location is the product of the actual signal strength at the target inspection location and the signal strength attenuation rate of the inspection robot at the target inspection location.

[0052] Based on S321-S324, the signal strength of each inspection location during the actual inspection process (considering power consumption) can be calculated relatively accurately, providing key signal strength data for subsequent priority value calculation.

[0053] S330 obtains the priority value of the specified candidate inspection path based on the actual signal strength of each inspection position on the specified candidate inspection path.

[0054] In one specific implementation, S330 includes:

[0055] S331, obtain the average actual signal strength of the specified candidate inspection path based on the actual signal strength of each inspection position on the specified candidate inspection path.

[0056] S332, obtain the actual signal strength variance of the specified candidate inspection path based on the actual signal strength of each inspection position on the specified candidate inspection path.

[0057] S333, obtain the priority value of the specified candidate inspection path based on the mean and variance of the actual signal strength of the specified candidate inspection path; the priority value of the specified candidate inspection path is positively correlated with the mean of the actual signal strength, and negatively correlated with the variance of the actual signal strength.

[0058] In this embodiment, the priority value is positively correlated with the mean of the actual signal strength, i.e., the larger the mean of the actual signal strength, the higher the priority value; and negatively correlated with the variance of the actual signal strength, i.e., the larger the variance of the actual signal strength, the lower the priority value. The priority value can be obtained using a weighted method. For example, the mean of the actual signal strength and the variance of the actual signal strength are normalized respectively. Then, the product of the first weight and the normalized mean of the actual signal strength is taken as the first component. The difference between 1 and the normalized variance of the actual signal strength is obtained. The product of the second weight and the difference is taken as the second component. The sum of the first component and the second component is taken as the priority value. Here, the sum of the first weight and the second weight is 1, both the first weight and the second weight are greater than 0 and less than 1, and the first weight and the second weight are empirical values.

[0059] In this embodiment, the mean actual signal strength reflects the overall signal strength level along the inspection path. A higher mean indicates better overall signal strength, which is beneficial for data transmission and robot control during the inspection process. The variance of the actual signal strength reflects the fluctuation of signal strength. A smaller variance indicates more stable signal strength and a more reliable inspection process. By combining these two indicators, the signal quality of the inspection path can be comprehensively evaluated. Therefore, this embodiment can obtain a reasonable priority value for evaluating the merits of the inspection path, providing a quantitative basis for subsequently determining the target inspection path.

[0060] S400: For any image acquired at any inspection location on the target inspection path, the image acquired at that inspection location is corrected according to the light output power at that inspection location to obtain the corrected image of that inspection location.

[0061] As one specific implementation, S400 includes:

[0062] S410, obtain the brightness gain coefficient and brightness offset corresponding to the light output power at the specified inspection location based on the light output power at the specified inspection location and the first preset model; the specified inspection location is any inspection location on the target inspection path; the first preset model includes the relationship between light output power and brightness gain coefficient and brightness offset.

[0063] In this embodiment, the first preset model is a pre-built model. By inputting the light output power of the specified inspection position into the first preset model, the brightness gain coefficient and brightness offset corresponding to the light output power of the specified inspection position can be obtained.

[0064] As an optional implementation, the first preset model is a neural network model, which performs inference by learning the relationship between the brightness gain coefficient and brightness offset and the light output power in multiple sets of data. Optionally, the first preset model adopts an existing fully connected neural network model, with the corresponding input being the light output power and the corresponding output being the brightness gain coefficient and brightness offset. Any set of data includes a light output power and its corresponding brightness gain coefficient and brightness offset. The process of training the first preset model is existing technology and will not be described in detail here.

[0065] S420 performs grayscale processing on the image collected at the specified inspection location to obtain a grayscale image of the specified inspection location.

[0066] As a specific implementation method, the images collected at the designated inspection locations are RGB images; those skilled in the art will know that the process of converting RGB images into grayscale images is existing technology and will not be described in detail here.

[0067] S430 corrects the grayscale image of the specified inspection position based on the brightness gain coefficient and brightness offset corresponding to the light output power of the specified inspection position, and obtains the corrected image of the specified inspection position.

[0068] In one specific implementation, S430 includes: for any pixel in the grayscale image at a specified inspection position, the brightness of the corresponding pixel in the corrected image at the specified inspection position is L. out L out =max(L min ,min(L max ,a×L in +b)), where a is the brightness gain coefficient corresponding to the light output power at the specified inspection position, b is the brightness offset corresponding to the light output power at the specified inspection position, L in L represents the brightness of the pixel in the grayscale image at the specified inspection location. min and L max These are the preset minimum and maximum brightness values, where `max()` takes the maximum value and `min()` takes the minimum value. Therefore, this embodiment corrects the acquired image based on actual lighting conditions, resulting in more accurate anomaly detection based on the corrected image.

[0069] S500 determines whether there is a pipeline abnormality at each inspection location on the target inspection path based on the corrected image of each inspection location on the target inspection path.

[0070] Those skilled in the art will recognize that the process of determining whether a pipe has an abnormality based on a grayscale image is existing technology, such as using Canny edge detection to extract pipe cracks.

[0071] In this embodiment, the light output power at each inspection location is related to the anomaly probability of that location. For inspection locations with a high anomaly probability, the light output power is increased to improve the illumination intensity, thereby acquiring clearer and more detailed images, which is beneficial for accurate subsequent judgment of pipeline anomalies. For inspection locations with a low anomaly probability, the light output power is reasonably controlled to avoid energy waste. Moreover, this embodiment corrects the images acquired at each inspection location based on the light output power, which can effectively correct the images, improve the image quality for anomaly judgment, and reduce image distortion caused by improper lighting conditions. Judging pipeline anomalies based on the corrected images can more accurately identify pipeline anomalies and improve detection accuracy.

[0072] Example 2:

[0073] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it performs the following steps:

[0074] Obtain a set of candidate inspection paths for the target pipeline route; the set of candidate inspection paths includes several candidate inspection paths.

[0075] For any candidate inspection path, the light output power of each inspection position on the candidate inspection path is obtained according to the anomaly probability of each inspection position on the candidate inspection path and the preset first mapping relationship; the first mapping relationship includes the mapping relationship between the anomaly probability and the light output power; the light output power of any inspection position on any candidate inspection path is positively correlated with the anomaly probability of the inspection position.

[0076] The priority value of each candidate inspection path is determined based on the light output power of each inspection position on each candidate inspection path, and the candidate inspection path with the highest priority value is determined as the target inspection path.

[0077] For any image acquired at any inspection location along the target inspection path, the image acquired at that inspection location is corrected based on the light output power at that inspection location to obtain the corrected image for that inspection location.

[0078] Determine whether there is a pipeline abnormality at each inspection location on the target inspection path based on the corrected image of each inspection location on the target inspection path.

[0079] Example 3:

[0080] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps:

[0081] Obtain a set of candidate inspection paths for the target pipeline route; the set of candidate inspection paths includes several candidate inspection paths.

[0082] For any candidate inspection path, the light output power of each inspection position on the candidate inspection path is obtained according to the anomaly probability of each inspection position on the candidate inspection path and the preset first mapping relationship; the first mapping relationship includes the mapping relationship between the anomaly probability and the light output power; the light output power of any inspection position on any candidate inspection path is positively correlated with the anomaly probability of the inspection position.

[0083] The priority value of each candidate inspection path is determined based on the light output power of each inspection position on each candidate inspection path, and the candidate inspection path with the highest priority value is determined as the target inspection path.

[0084] For any image acquired at any inspection location along the target inspection path, the image acquired at that inspection location is corrected based on the light output power at that inspection location to obtain the corrected image for that inspection location.

[0085] Determine whether there is a pipeline abnormality at each inspection location on the target inspection path based on the corrected image of each inspection location on the target inspection path.

[0086] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0087] While specific embodiments of the invention have been described in detail by way of example, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of the invention. It should also be understood that various modifications can be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.

Claims

1. A data processing method for pipeline anomaly detection, characterized in that, The method comprises the following steps: S100, acquiring a candidate inspection path set of a target pipeline route; the candidate inspection path set comprises a plurality of candidate inspection paths; S200, for any candidate inspection path, acquiring a light output power of each inspection position on the candidate inspection path according to an abnormal probability of each inspection position on the candidate inspection path and a preset first mapping relationship; the first mapping relationship comprises a mapping relationship between the abnormal probability and the light output power; the light output power of any inspection position on any candidate inspection path is positively correlated with the abnormal probability of the inspection position; S300, determining a priority value of each candidate inspection path according to the light output power of each inspection position on each candidate inspection path, and determining a candidate inspection path with a maximum priority value as a target inspection path; S400, for an image collected at any inspection position on the target inspection path, correcting the image collected at the inspection position according to the light output power of the inspection position to obtain a corrected image of the inspection position; S500, judging whether a pipeline abnormality occurs at each inspection position on the target inspection path according to the corrected image of each inspection position on the target inspection path; S300 comprises: S310, correcting an initial power decay curve of a specified candidate inspection path according to the light output power of each inspection position on the specified candidate inspection path to obtain a target power decay curve of the specified candidate inspection path; the horizontal coordinate of the initial power decay curve is an inspection distance, and the vertical coordinate of the initial power decay curve is an electric quantity of an inspection robot; the specified candidate path is any candidate inspection path in the candidate inspection path set; S320, acquiring an actual signal strength of each inspection position on the specified candidate inspection path according to the target power decay curve of the specified candidate inspection path, a signal strength decay curve and a standard signal strength of each inspection position on the specified candidate inspection path; the horizontal coordinate of the signal strength decay curve is the electric quantity of the inspection robot, and the vertical coordinate of the signal strength decay curve is a signal strength decay rate; the standard signal strength of any inspection position on the specified candidate inspection path is an actual signal strength of the inspection robot at the inspection position in a full electric quantity state; S330, acquiring a priority value of the specified candidate inspection path according to the actual signal strength of each inspection position on the specified candidate inspection path.

2. The data processing method for pipeline anomaly detection according to claim 1, wherein, S400 comprises: S410, acquiring a luminance gain coefficient and a luminance offset corresponding to the light output power of a specified inspection position according to the light output power of the specified inspection position and a first preset model; the specified inspection position is any inspection position on the target inspection path; the first preset model comprises a relationship between the light output power and the luminance gain coefficient and the luminance offset; S420, performing grayscale processing on the image collected at the specified inspection position to obtain a grayscale image of the specified inspection position; S430, correcting the grayscale image of the specified inspection position according to the luminance gain coefficient and the luminance offset corresponding to the light output power of the specified inspection position to obtain a corrected image of the specified inspection position.

3. The data processing method for pipeline anomaly detection according to claim 2, wherein, S430 comprises: For any pixel point in the gray image of the designated inspection position, the luminance of the corresponding pixel point in the corrected image of the designated inspection position is L out , L out =max(L min ,min(L max ,a×L in +b)), a is a luminance gain coefficient corresponding to the light output power of the designated inspection position, b is a luminance offset corresponding to the light output power of the designated inspection position, L in is the luminance of the pixel point in the gray image of the designated inspection position, L min and L max are respectively a preset minimum luminance and a maximum luminance, max() is a maximum value, and min() is a minimum value.

4. The data processing method for pipeline anomaly detection according to claim 1, wherein S330 comprises: S331, obtaining a mean of actual signal strength of the specified candidate inspection path according to the actual signal strength of each inspection position on the specified candidate inspection path; S332, obtaining a variance of actual signal strength of the specified candidate inspection path according to the actual signal strength of each inspection position on the specified candidate inspection path; S333, obtaining a priority value of the specified candidate inspection path according to the mean of actual signal strength and the variance of actual signal strength of the specified candidate inspection path; The priority value of the specified candidate inspection path is positively correlated with the mean of actual signal strength, and the priority value of the specified candidate inspection path is negatively correlated with the variance of actual signal strength.

5. The data processing method for pipeline anomaly detection according to claim 1, wherein, The initial power decay curve is a relationship between the power of the inspection robot and the inspection distance under a preset light output power, and S310 comprises: S311, dividing the specified candidate inspection path into a plurality of segments according to the inspection positions; S312, for any segment, obtaining the power consumption of the segment according to the light output power corresponding to the segment; the power consumption of any segment is the product of the power consumed by the inspection robot per unit length of movement and the light output power variation rate of the segment under the preset light output power, and the light output power variation rate of any segment is the ratio of the light output power corresponding to the segment to the preset light output power; S313, sequentially obtaining the remaining power of the inspection robot when reaching each segment from the starting point of the specified candidate inspection path; S314, obtaining the target power decay curve of the specified candidate inspection path according to the remaining power of the inspection robot when reaching each segment.

6. The data processing method for pipeline anomaly detection according to claim 1, wherein, S320 comprises: S321, obtaining the cumulative inspection distance of the target inspection position in the specified candidate inspection path according to the specified candidate inspection path; the target inspection position is any inspection position on the specified candidate inspection path; S322, obtaining the remaining power of the inspection robot when reaching the target inspection position according to the cumulative inspection distance of the target inspection position in the specified candidate inspection path and the target power decay curve of the specified candidate inspection path; S323, obtaining the signal strength decay rate of the inspection robot at the target inspection position according to the remaining power of the inspection robot when reaching the target inspection position and the signal strength decay curve; S324, obtaining the actual signal strength of the target inspection position according to the signal strength decay rate of the inspection robot at the target inspection position and the standard signal strength of the target inspection position.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the data processing method for pipeline anomaly detection according to any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, the computer-readable storage medium comprising: The computer program is executed by the processor to implement the data processing method for pipeline anomaly detection according to any one of claims 1 to 6.

Images