A power transmission line point cloud difference inspection method, device, equipment and medium

Abstract

The embodiment of the specification discloses a power transmission line point cloud difference inspection method, device, equipment and medium, comprising: determining two specified times of radar point cloud data difference detection when a monitoring device inspects a power transmission line; determining a first radar point cloud data group and a second radar point cloud data group corresponding to the two specified times respectively; partitioning according to depth information, and dividing radar point cloud data with depth information less than a first preset value in the first radar point cloud data group and the second radar point cloud data group to a first partition; in the first partition, establishing a corresponding relationship between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group, and determining the difference detection radar point cloud data in the first partition; inputting the difference detection radar point cloud data into a pre-trained difference identification model to obtain a difference identification result, and processing hidden dangers according to the difference analysis result.

Description

Technical Field

[0001] This specification relates to the field of computer technology, and in particular to a method, apparatus, equipment and medium for inspecting point cloud differences in power transmission lines. Background Technology

[0002] Workers observe, inspect, and measure various components of transmission lines using their eyes, binoculars, and other tools and instruments. The purpose is to understand the operational status of the transmission lines and promptly identify equipment defects and threats to line safety. Because overhead lines are widely distributed and operate outdoors for extended periods, they are frequently affected by changes in the surrounding environment and nature. Furthermore, since distribution lines have more and more complex equipment than transmission lines, inspections should not only focus on the transmission lines themselves but also on inspecting specialized equipment.

[0003] In existing technologies, the inspection of power transmission lines mostly utilizes drone image point cloud inspection or portable LiDAR equipment. Drone image point cloud inspection enables 3D modeling and three-dimensional inspection of power transmission line corridors, reducing the inspection workload of power personnel and improving inspection efficiency, thus gaining widespread adoption. Portable LiDAR equipment can be used in no-fly zones or scenarios where drone scanning is not suitable, serving as a supplement to drone scanning. However, both drone inspection and portable LiDAR inspection equipment suffer from high inspection costs and low radar point cloud data update frequency, resulting in a poor user experience when inspecting power transmission lines. Summary of the Invention

[0004] This specification provides one or more embodiments of a method, apparatus, equipment, and medium for inspecting point cloud differences in transmission lines, which are used to solve the following technical problems:

[0005] Drone inspection and portable radar inspection equipment suffer from high inspection costs and low radar point cloud data update frequency, resulting in a poor user experience when inspecting power transmission lines.

[0006] One or more embodiments of this specification employ the following technical solutions:

[0007] This specification provides one or more embodiments of a method for inspecting point cloud differences in transmission lines, the method comprising:

[0008] When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data.

[0009] The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0010] Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value are divided into the first partition.

[0011] In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition is determined.

[0012] The differential detection radar point cloud data is input into a pre-trained differential identification model to obtain differential identification results, and potential hazards are handled based on the differential analysis results.

[0013] This specification provides one or more embodiments of a power transmission line point cloud difference inspection device, the device comprising:

[0014] The time determination unit determines two designated times for differential detection of radar point cloud data when the monitoring equipment is inspecting power transmission lines;

[0015] The data group determination unit determines the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times respectively. Both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0016] The unit is divided into partitions based on depth information, and radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than a first preset value are divided into the first partition.

[0017] The detection data determination unit establishes a correspondence between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group in the first partition, removes radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group, and determines the differentiated detection radar point cloud data in the first partition.

[0018] The hazard identification and processing unit inputs the differential detection radar point cloud data into a pre-trained differential identification model to obtain differential identification results, and performs hazard processing based on the differential analysis results.

[0019] This specification provides one or more embodiments of a power transmission line point cloud difference inspection device, the device comprising:

[0020] At least one processor; and,

[0021] A memory communicatively connected to the at least one processor; wherein,

[0022] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0023] When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data.

[0024] The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0025] Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value are divided into the first partition.

[0026] In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition is determined.

[0027] The differential detection radar point cloud data is input into a pre-trained differential identification model to obtain differential identification results, and potential hazards are handled based on the differential analysis results.

[0028] This specification provides one or more embodiments of a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured as follows:

[0029] When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data.

[0030] The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0031] Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value are divided into the first partition.

[0032] In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition is determined.

[0033] The differential detection radar point cloud data is input into a pre-trained differential identification model to obtain differential identification results, and potential hazards are handled based on the differential analysis results.

[0034] The above-mentioned technical solutions adopted in the embodiments of this specification can achieve the following beneficial effects: The embodiments of this specification comprehensively consider the influence of micro-meteorological factors on radar ranging, introduce a radar temperature control system, realize data acquisition in a more ideal temperature zone, and improve the consistency of radar point cloud data acquisition; In view of the problem of many false alarms in radar point cloud data difference analysis, a partitioning and judgment method for acquired radar point cloud data is adopted. By using the average reflectance of the partitions, factors with large differences in the same scene between two sets of data are eliminated, reducing the influence of rain, snow and frost. By dividing the confidence level of the partitions and adopting different judgment strategies, a more accurate point cloud data difference judgment is achieved, reducing the false alarm rate, improving inspection efficiency, and having considerable practical value. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0036] Figure 1 A flowchart illustrating a point cloud difference inspection method for transmission lines provided in one or more embodiments of this specification;

[0037] Figure 2 A flowchart of the inspection process for a fixed-point lidar device provided in one or more embodiments of this specification;

[0038] Figure 3 Flowchart of the differentiated point cloud determination method provided in one or more embodiments of this specification;

[0039] Figure 4 A schematic diagram of the structure of a point cloud difference inspection device for transmission lines provided in one or more embodiments of this specification;

[0040] Figure 5This is a structural schematic diagram of a power transmission line point cloud difference inspection device provided for one or more embodiments of this specification. Detailed Implementation

[0041] This specification provides a method, apparatus, equipment, and medium for inspecting point cloud differences in transmission lines.

[0042] Drone image point cloud inspection enables 3D modeling and three-dimensional inspection of power transmission line corridors, reducing the inspection workload of power personnel and improving inspection efficiency, leading to its widespread adoption. Portable LiDAR equipment can be used in no-fly zones or scenarios where drone scanning is not suitable, serving as a supplement to drone scanning. Intelligent power transmission line inspection devices based on fixed-point LiDAR combine the characteristics of imagery and LiDAR to achieve 3D ranging of potential hazards in the corridor, quantifying the distance from the hazard target to the conductor, and improving inspection efficiency. With technological maturity and reduced radar costs, it will be widely adopted in the future. However, drone inspection and portable radar inspection equipment suffer from high inspection costs and low point cloud data update frequency, hindering real-time monitoring of corridors.

[0043] Meanwhile, the power transmission line monitoring device based on fixed-point lidar does not comprehensively consider the combined impact of micro-meteorological information such as temperature, humidity, light, and weather on the radar data collected, which may lead to misjudgments in hazard warnings.

[0044] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0045] Figure 1 This is a flowchart illustrating a method for inspecting point cloud differences in transmission lines, provided in one or more embodiments of this specification. This process can be executed by a transmission line inspection system, which can freely set the inspection frequency, greatly improving the problem of low update frequency of radar point cloud data. Some input parameters or intermediate results in the process can be manually adjusted to help improve accuracy.

[0046] The method flow steps of the embodiments in this specification are as follows:

[0047] S102, when the monitoring equipment is conducting power transmission line inspections, determine two designated times for differential detection of radar point cloud data.

[0048] In the embodiments of this specification, the monitoring device can be a fixed-point lidar device used to obtain radar point cloud data. The ranging principle of a fixed-point lidar device is to continuously send light pulses to the target, and then use a sensor to receive the light returning from the object. The distance to the target is obtained by detecting the flight (round trip) time of the light pulses. A TOF camera simultaneously obtains the depth (distance) information of the image.

[0049] It should be noted that the two specified times are two time points used for radar point cloud data differentiation detection. That is, the radar point cloud data at the two specified times are compared to identify differences, and the results of the comparison are used to determine whether there are hidden dangers or faults in the transmission line.

[0050] In this embodiment of the specification, when determining two designated times for radar point cloud data differential detection, the two designated times can be determined according to a pre-set radar point cloud data differential detection time interval. The radar point cloud data differential detection time interval can be set according to actual conditions. If the transmission line is located in an area prone to accidents, the radar point cloud data differential detection time interval can be adjusted to adapt to the current environment of the transmission line.

[0051] S104, determine the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times respectively, both of which are generated by the monitoring device.

[0052] Radar point cloud data is significantly affected by micrometeorological conditions. Ignoring micrometeorological data may lead to inaccurate radar point cloud data, affecting the final inspection results. Micrometeorology is unpredictable and its changes cannot be analyzed manually, so real-time monitoring of weather conditions is necessary. For example, in freezing weather, transmission lines often experience icing, resulting in line breakage and tower collapse; strong winds can cause lines to tilt and break, or deform components and towers, leading to power outages.

[0053] Based on the above problems, in determining the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times in this embodiment, micro-meteorological data can be acquired first according to a preset time interval. This time interval can also be set according to the actual situation. The acquired micro-meteorological data is used to subsequently determine the micro-meteorological data corresponding to the two specified times. Then, it is determined whether the micro-meteorological data corresponding to the two specified times meets the preset acquisition conditions. The preset acquisition conditions can also be set according to the actual situation to meet the micro-meteorological conditions for subsequent differential detection of radar point cloud data. For example, acquisition conditions can be set in terms of light intensity, temperature, humidity, wind speed, rainfall, air pressure, etc. After that, if it is determined that the micro-meteorological data corresponding to the two specified times meets the preset acquisition conditions, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times can be determined. If it is determined that the micro-meteorological data corresponding to the two specified times does not meet the preset acquisition conditions, the specified time can be reset to ensure that the subsequent radar point cloud data is accurate and reliable.

[0054] When acquiring micro-meteorological data, a micro-meteorological monitoring system can be used. This system can be designed for monitoring the meteorological environment of transmission lines at specific locations. Meteorological sensors (light intensity sensors, temperature sensors, humidity sensors, wind speed sensors, wind direction sensors, rainfall sensors, and air pressure sensors) are installed on the transmission line towers to detect the weather.

[0055] Furthermore, besides the impact of micro-meteorological data on radar point cloud data, the temperature of the monitoring equipment itself also affects the acquisition of radar point cloud data. If it is determined that the micro-meteorological data corresponding to the two specified times meet the preset acquisition conditions, it can be determined whether the temperature of the monitoring equipment is within a preset temperature range. If it is determined that the temperature of the monitoring equipment is within the preset temperature range, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined. In the above-mentioned process of controlling the temperature of the monitoring equipment itself, a radar temperature control system can be used.

[0056] S106, partition the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value and divide them into the first partition.

[0057] In this embodiment of the specification, after partitioning according to depth information, the radar point cloud data in the first radar point cloud data and the second radar data in which the depth information is not less than a first preset value are divided into a second partition, and the radar point cloud data in the first radar point cloud data group and the second radar data group in the second partition are removed.

[0058] In the embodiments of this specification, the depth information is the distance at which the monitoring device collects radar point cloud data. If the depth information is higher than a certain value, the accuracy of the collected radar point cloud data will be affected. Based on this, the embodiments of this specification partition the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group based on the depth information. The radar point cloud data with a depth information less than a first preset value is assigned to the first partition, and the radar point cloud data in this partition has a certain degree of reliability. The radar point cloud data with a depth information not less than the first preset value is assigned to the second partition, and the radar point cloud data in this partition does not have a certain degree of reliability. The radar point cloud data in this partition is then discarded.

[0059] This embodiment describes differential detection of radar point cloud data sets from two different times, essentially radar point cloud data sets at the same location but at different times. The two radar point cloud data sets have a data correspondence. It's possible that a radar point cloud data set from the first radar point cloud data set is assigned to the second partition, making the data unreliable. This same radar point cloud data set also exists in the second radar point cloud data set, but it might be located in the first partition of the second radar point cloud data set. Therefore, the radar point cloud data in the first partition needs to be processed to prevent mismatched radar point cloud data from appearing during subsequent differential detection. This problem can be addressed by executing step S108.

[0060] S108, in the first partition, establish a correspondence between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group, remove the radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group, and determine the differentiated detection radar point cloud data in the first partition.

[0061] For the radar point cloud data in the first partition, it is also necessary to pay attention to the reflectivity of the radar point cloud data. If the depth information of some radar point cloud data falls between the second and first preset values, radar point cloud data with a reflectivity not less than the third preset value is also unreliable and needs to be discarded to ensure the reliability of the collected radar point cloud data. The first, second, and third preset values ​​can all be adjusted according to actual conditions. The first preset value is greater than the second preset value, and the third preset value is unrelated to either the first or second preset values.

[0062] Based on the above problems, the embodiments of this specification can divide the first partition into a first sub-partition and a second sub-partition. When determining the differentiated detection radar point cloud data in the first partition, partitioning is performed based on depth information. Radar point cloud data in both the first and second radar data groups with depth information less than a second preset value are assigned to the first sub-partition, where the second preset value is less than the first preset value. Radar point cloud data in both the first and second radar data groups with depth information greater than the first preset value are assigned to the second sub-partition. The reflectivity of the radar point cloud data in the second sub-partition is determined, and radar point cloud data with reflectivity less than a third preset value are filtered out. Radar point cloud data with reflectivity not less than the third preset value in the second sub-partition are removed, as are radar point cloud data in the first and second radar data groups that cannot establish a correspondence. This determines the differentiated detection radar point cloud data in the first partition. The removal of radar point cloud data in the second sub-partition and the removal of radar point cloud data in the first and second radar data groups that cannot establish a correspondence are the same as described above and will not be repeated here.

[0063] It should be noted that the radar point cloud data removed in the embodiments of this specification consists of a small number of discrete points, which does not affect the subsequent differential identification between the first radar point cloud data group and the second radar data group.

[0064] S110, input the differential detection radar point cloud data into the pre-trained differential recognition model to obtain the differential recognition result, and handle the hidden dangers based on the differential analysis result.

[0065] In this embodiment of the specification, when handling potential hazards based on the differential analysis results, if a potential hazard exists in the differential identification results within the first sub-partition, a corresponding alarm message is output. If a potential hazard exists in the differential identification results within the second sub-partition, confirmation is performed using differential identification results from multiple sets of radar point cloud data. If all differential identification results from the multiple sets of radar point cloud data indicate a potential hazard, a corresponding alarm message is output. The reliability of the radar point cloud data in the second sub-partition is lower than that in the first sub-partition; therefore, confirmation is performed using differential identification results from multiple sets of radar point cloud data.

[0066] When the differential identification result in this specification's embodiments indicates no hidden danger, the hidden danger handling method can be to perform no handling; when the differential identification result indicates the existence of a hidden danger, during the hidden danger handling process, different warning levels can be set according to the distance between the hidden danger and the transmission line. Furthermore, when determining the warning level, the distance to the hidden danger and the type of hidden danger can also be combined. Corresponding to the above scheme, this specification's embodiments also provide more specific implementation methods, see [link to documentation]. Figure 2The flowchart shown is for the inspection of fixed-point lidar equipment, used to determine the reliability of the collected lidar point cloud data. The specific working method is as follows:

[0067] At the start of the inspection, the equipment periodically reads micro-meteorological data according to a set time interval. At this point, the equipment functions as a micro-meteorological monitoring system. Then, when the fixed-point lidar equipment begins its inspection, it queries the currently collected micro-meteorological data and determines whether the current data meets the collection conditions. If the current micro-meteorological data meets the set collection conditions, the collected radar point cloud data is highly reliable, and the fixed-point lidar equipment temperature reaches the preset temperature range; in this case, radar point cloud data is collected. If the current micro-meteorological data does not meet the set collection conditions, the collected radar point cloud data is of low reliability, and radar point cloud data is not collected.

[0068] After collecting the radar point cloud data, differential detection was performed on the radar point cloud data. (See [link to documentation]). Figure 3 The flowchart shown illustrates the differential point cloud determination method, used to determine the differences between radar point cloud data from different times, and then to handle potential hazards. The specific working method is as follows:

[0069] The process begins by setting the radar point cloud data difference detection time interval and two specified time periods. Then, the monitoring equipment collects radar point cloud data containing micro-meteorological information according to the set time. If unsuitable collection conditions occur, the equipment automatically adjusts the collection time according to set rules. Next, the two sets of collected radar point cloud data are divided into three partitions based on depth information. Radar point cloud data in partition III is not sent to the difference identification model. The average reflectance difference of the radar point cloud data in partition II is judged. If the average reflectance difference is less than a set value, the radar point cloud data in partition II is sent to the difference identification model, and the radar point cloud data in partition I is also sent to the difference identification model for difference analysis. If the average reflectance difference of the radar point cloud data in partition II is not less than the set value, the difference analysis process exits. Afterwards, once differences are identified in the radar point cloud data of the corresponding partition, the data is returned to the corresponding image. The identified difference area is input into the hazard model to identify whether a hazard exists, eliminating interference from mobile devices. Hazards can include large vehicles, etc. After assessment, if a potential hazard is found in partition I, it will be addressed; if a potential hazard is found in partition II, it will be confirmed through further analysis of multiple sets of data; radar point cloud data in partition III will be directly filtered out without assessment. In this specification, both the differentiated identification model and the hazard model in the embodiments can be pre-trained neural network models.

[0070] It should be noted that the embodiments in this specification are mainly used to address changes in terrain and trees, and generally have a long cycle, so the radar parameters should be kept as consistent as possible. On the device side: When the lidar starts scanning, it reads the current temperature of the fixed-point lidar device. If the temperature is lower than the set value, the fixed-point lidar device is heated to reach the set temperature, then radar point cloud data scanning is initiated, and radar point cloud data with meteorological information is returned.

[0071] Meanwhile, the embodiments in this specification select confidence intervals based on depth information: three confidence intervals, i.e., three partitions, are set. In the first interval, the data is less affected by temperature, light intensity, and humidity, and the data reliability is high; in the second interval, the data reliability is moderate, and the influence of parameters such as light intensity, temperature, and humidity needs to be considered; in the third interval, the confidence level is low, the data is unreliable, and it is directly filtered out.

[0072] Furthermore, the discrete points acquired by the fixed-point lidar equipment in the embodiments of this specification can be verified in a two-dimensional image, filtering out noise. It should be noted that the embodiments of this specification comprehensively consider the influence of micro-meteorological factors on radar ranging, introducing a radar temperature control system to achieve data acquisition within a more ideal temperature range, improving the consistency of radar point cloud data acquisition. Addressing the issue of numerous false alarms during radar point cloud data difference analysis, a partitioning method for the acquired radar point cloud data is adopted. By using the average reflectance of each partition to eliminate factors with large differences in the same scene between two sets of data, the influence of rain, snow, and frost is reduced. Through partitioned confidence levels and the adoption of different judgment strategies, relatively accurate point cloud data difference determination is achieved, reducing the false alarm rate and improving inspection efficiency, demonstrating significant practical value.

[0073] Figure 4 This is a schematic diagram of the structure of a point cloud difference inspection device for transmission lines provided in one or more embodiments of this specification. The device includes: a time determination unit 402, a data group determination unit 404, a division unit 406, a detection data determination unit 408, and a hidden danger identification and processing unit 410.

[0074] The time determination unit 402 determines two designated times for differential detection of radar point cloud data when the monitoring equipment is inspecting power transmission lines;

[0075] The data group determination unit 404 determines the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times respectively. Both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0076] The partitioning unit 406 partitions the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value into the first partition.

[0077] In the first partition, the detection data determination unit 408 establishes a correspondence between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group, removes radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group, and determines the differentiated detection radar point cloud data in the first partition.

[0078] The hazard identification and processing unit 410 inputs the differential detection radar point cloud data into a pre-trained differential identification model to obtain differential identification results, and performs hazard processing based on the differential analysis results.

[0079] Figure 5 This specification provides a schematic diagram of the structure of a power transmission line point cloud difference inspection device according to one or more embodiments. The device includes:

[0080] At least one processor; and,

[0081] A memory communicatively connected to the at least one processor; wherein,

[0082] The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to:

[0083] When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data.

[0084] The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0085] Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value are divided into the first partition.

[0086] In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition is determined.

[0087] The differential detection radar point cloud data is input into a pre-trained differential identification model to obtain differential identification results, and potential hazards are handled based on the differential analysis results.

[0088] This specification provides one or more embodiments of a non-volatile computer storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured as follows:

[0089] When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data.

[0090] The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device.

[0091] Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar point cloud data group whose depth information is less than the first preset value are divided into the first partition.

[0092] In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition is determined.

[0093] The differential detection radar point cloud data is input into a pre-trained differential identification model to obtain differential identification results, and potential hazards are handled based on the differential analysis results.

[0094] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0095] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0096] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for inspecting point cloud differences in transmission lines, characterized in that, The method includes: When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data. The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device. Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value are divided into the first partition. In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition are determined. The differential detection radar point cloud data is input into a pre-trained differential recognition model to obtain differential recognition results, and potential hazards are handled based on the differential analysis results. The first partition includes a first subpartition and a second subpartition; The determination of the differentiated detection radar point cloud data in the first partition specifically includes: Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the second preset value are divided into the first sub-partition, where the second preset value is less than the first preset value. The radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value but greater than the second preset value are divided into the second sub-partition; Determine the reflectivity of radar point cloud data in the second sub-partition, filter out radar point cloud data with reflectivity less than a third set value, remove radar point cloud data with reflectivity not less than the third set value in the second sub-partition, and remove radar point cloud data in the first radar point cloud data group and the second radar data group that cannot establish a corresponding relationship, and determine the differentiated detection radar point cloud data in the first sub-partition. The determination of the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times specifically includes: Micro-meteorological data are acquired according to a pre-set time interval; Determine whether the micrometeorological data corresponding to the two specified times meet the preset collection conditions; If it is determined that the micro-meteorological data corresponding to the two specified times meet the preset collection conditions, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively. The process of handling potential hazards based on the differential analysis results specifically includes: If there are potential risks in the differential identification results within the first sub-partition, output the corresponding alarm information; If there are potential risks in the differential identification results within the second sub-partition, the differential identification results obtained from multiple sets of radar point cloud data are used for confirmation. If all the differential identification results from the multiple sets of radar point cloud data indicate potential risks, the corresponding alarm information is output.

2. The method according to claim 1, characterized in that, After partitioning based on depth information, the method further includes: The radar point cloud data in the first radar point cloud data and the second radar data in which the depth information is not less than the first preset value are divided into the second partition, and the radar point cloud data in the first radar point cloud data group and the second radar data group in the second partition are removed.

3. The method according to claim 1, characterized in that, The determination of the two specified time points for differential detection of radar point cloud data specifically includes: Based on the pre-set differential detection time interval for radar point cloud data, two designated times for differential detection of radar point cloud data are determined.

4. The method according to claim 1, characterized in that, If it is determined that the micrometeorological data corresponding to the two specified times meet the preset collection conditions, the method further includes: Determine whether the temperature of the monitoring device is within a preset temperature range; If it is determined that the temperature of the monitoring device is within the preset temperature range, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively.

5. A point cloud difference inspection device for transmission lines, characterized in that, The device includes: The time determination unit determines two designated times for differential detection of radar point cloud data when the monitoring equipment is inspecting power transmission lines; The data group determination unit determines the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times respectively. Both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device. The unit is divided into partitions based on depth information, and radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than a first preset value are divided into the first partition. The detection data determination unit establishes a correspondence between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group in the first partition, removes radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar data group, and determines the differentiated detection radar point cloud data in the first partition. The hazard identification and processing unit inputs the differential detection radar point cloud data into a pre-trained differential identification model to obtain differential identification results, and performs hazard processing based on the differential analysis results. The first partition includes a first subpartition and a second subpartition; The determination of the differentiated detection radar point cloud data in the first partition specifically includes: Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the second preset value are divided into the first sub-partition, where the second preset value is less than the first preset value. The radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value but greater than the second preset value are divided into the second sub-partition; Determine the reflectivity of radar point cloud data in the second sub-partition, filter out radar point cloud data with reflectivity less than a third set value, remove radar point cloud data with reflectivity not less than the third set value in the second sub-partition, and remove radar point cloud data in the first radar point cloud data group and the second radar data group that cannot establish a corresponding relationship, and determine the differentiated detection radar point cloud data in the first sub-partition. The determination of the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times specifically includes: Micro-meteorological data are acquired according to a pre-set time interval; Determine whether the micrometeorological data corresponding to the two specified times meet the preset collection conditions; If it is determined that the micro-meteorological data corresponding to the two specified times meet the preset collection conditions, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively. The process of handling potential hazards based on the differential analysis results specifically includes: If there are potential risks in the differential identification results within the first sub-partition, output the corresponding alarm information; If there are potential risks in the differential identification results within the second sub-partition, the differential identification results obtained from multiple sets of radar point cloud data are used for confirmation. If all the differential identification results from the multiple sets of radar point cloud data indicate potential risks, the corresponding alarm information is output.

6. A point cloud difference inspection device for transmission lines, characterized in that, The device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data. The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device. Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value are divided into the first partition. In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition are determined. The differential detection radar point cloud data is input into a pre-trained differential recognition model to obtain differential recognition results, and potential hazards are handled based on the differential analysis results. The first partition includes a first subpartition and a second subpartition; The determination of the differentiated detection radar point cloud data in the first partition specifically includes: Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the second preset value are divided into the first sub-partition, where the second preset value is less than the first preset value. The radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value but greater than the second preset value are divided into the second sub-partition; Determine the reflectivity of radar point cloud data in the second sub-partition, filter out radar point cloud data with reflectivity less than a third set value, remove radar point cloud data with reflectivity not less than the third set value in the second sub-partition, and remove radar point cloud data in the first radar point cloud data group and the second radar data group that cannot establish a corresponding relationship, and determine the differentiated detection radar point cloud data in the first sub-partition. The determination of the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times specifically includes: Micro-meteorological data are acquired according to a pre-set time interval; Determine whether the micrometeorological data corresponding to the two specified times meet the preset collection conditions; If it is determined that the micro-meteorological data corresponding to the two specified times meet the preset collection conditions, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively. The process of handling potential hazards based on the differential analysis results specifically includes: If there are potential risks in the differential identification results within the first sub-partition, output the corresponding alarm information; If there are potential risks in the differential identification results within the second sub-partition, the differential identification results obtained from multiple sets of radar point cloud data are used for confirmation. If all the differential identification results from the multiple sets of radar point cloud data indicate potential risks, the corresponding alarm information is output.

7. A non-volatile computer storage medium, characterized in that, The computer-executable instructions are stored thereon and are configured as follows: When monitoring equipment is used to inspect power transmission lines, two designated times are determined for differential detection of radar point cloud data. The first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively, and both the first radar point cloud data group and the second radar point cloud data group are generated by the monitoring device. Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value are divided into the first partition. In the first partition, a correspondence is established between the radar point cloud data of the first radar point cloud data group and the radar point cloud data of the second radar point cloud data group. Radar point cloud data in the first partition that cannot establish a correspondence between the first radar point cloud data group and the second radar point cloud data group are removed, and the differentiated detection radar point cloud data in the first partition are determined. The differential detection radar point cloud data is input into a pre-trained differential recognition model to obtain differential recognition results, and potential hazards are handled based on the differential analysis results. The first partition includes a first subpartition and a second subpartition; The determination of the differentiated detection radar point cloud data in the first partition specifically includes: Based on depth information, the radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the second preset value are divided into the first sub-partition, where the second preset value is less than the first preset value. The radar point cloud data in the first radar point cloud data group and the second radar data group whose depth information is less than the first preset value but greater than the second preset value are divided into the second sub-partition; Determine the reflectivity of radar point cloud data in the second sub-partition, filter out radar point cloud data with reflectivity less than a third set value, remove radar point cloud data with reflectivity not less than the third set value in the second sub-partition, and remove radar point cloud data in the first radar point cloud data group and the second radar data group that cannot establish a corresponding relationship, and determine the differentiated detection radar point cloud data in the first sub-partition. The determination of the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times specifically includes: Micro-meteorological data are acquired according to a pre-set time interval; Determine whether the micrometeorological data corresponding to the two specified times meet the preset collection conditions; If it is determined that the micro-meteorological data corresponding to the two specified times meet the preset collection conditions, the first radar point cloud data group and the second radar point cloud data group corresponding to the two specified times are determined respectively. The process of handling potential hazards based on the differential analysis results specifically includes: If there are potential risks in the differential identification results within the first sub-partition, output the corresponding alarm information; If there are potential risks in the differential identification results within the second sub-partition, the differential identification results obtained from multiple sets of radar point cloud data are used for confirmation. If all the differential identification results from the multiple sets of radar point cloud data indicate potential risks, the corresponding alarm information is output.

Images