Dam detection method and system based on unmanned aerial vehicle

Abstract

The invention relates to a dam detection method and system based on an unmanned aerial vehicle, and relates to the field of safety detection technology, and the method comprises the steps: obtaining an area detection time point; determining a detection interval duration according to the region detection time point and the current time point; constructing a historical interval to obtain crack interval duration; determining an effective interval duration according to all crack interval durations in a single local area, and defining the local area with the detection interval duration greater than the effective interval duration as a required detection area; determining area detection points on the demand detection area, constructing a simulation detection path according to each area detection point and the starting point, and determining a demand flight path according to the simulation detection path; and determining the required flight path with the minimum numerical value, defining the corresponding simulation detection path as a use detection path, and controlling the unmanned aerial vehicle to move along the use detection path at the starting point to obtain a dam detection image. The method and the device have the effect of improving the detection efficiency of dam detection.

Description

Technical Field

[0001] The present application relates to the field of safety detection technology, and in particular to a dam detection method and system based on drones. Background Art

[0002] A dam is a building or structure used to prevent water from entering or leaving the dam. It protects the lives and property of people in the surrounding areas. By building sluice gates, channels and other facilities on the dam, water flow can be regulated to achieve farmland irrigation and the rational use of water resources. Dam leakage refers to the phenomenon that moisture or water in the dam or levee structure penetrates or seeps out through cracks, holes or other infiltration channels. Dam leakage may have an adverse effect on the stability of the dam. Therefore, the detection of dam cracks is an important link. After completing the detection of the dam, the cracks can be repaired in time when no leakage occurs to avoid losses.

[0003] At present, the inspection of dams is often carried out using drones equipped with image capture equipment. That is, the area to be inspected is generated based on the dam's own parameters, and the staff determines the flight path of the drone in the inspection area, so that the drone can acquire images of various parts of the dam during flight inspection, and thus the cracks in the dam can be determined based on the images.

[0004] In the above-mentioned related technologies, accurate detection is required for each location when crack detection is performed. However, during the actual use of the dam, due to the influence of water erosion, the probability of cracks occurring at each location is different. Therefore, when using the above-mentioned scheme for detection, there will be many invalid detections, resulting in low detection efficiency and inconvenience for daily general inspections. There is still room for improvement. Summary of the invention

[0005] In order to improve the detection efficiency of dam inspection, the present application provides a dam inspection method and system based on drones.

[0006] In the first aspect, the present application provides a dam detection method based on a drone, which adopts the following technical solution: A dam inspection method based on drone, comprising: Obtaining regional detection time points of each preset local area on the dam; Determine the detection interval duration based on the regional detection time point and the current time point; Constructing a historical interval with the current time point as the rear end point and a width of the preset historical duration on the preset time axis, and obtaining the crack interval duration of each local area in the historical interval; In a single local area, the effective interval duration is calculated based on all crack interval durations, and the local area where the detection interval duration is longer than the effective interval duration is defined as the required detection area; Determine area detection points on the demand detection area, randomly sort each area detection point and connect it in series with a preset starting point to construct a simulated detection path, and determine the demand flight distance according to the simulated detection path; Determine the demand flight distance with the smallest value according to the preset sorting rule, define the simulated detection path corresponding to the demand flight distance as the used detection path, and control the drone to move along the used detection path at the starting point to obtain the dam detection image.

[0007] Optionally, the step of calculating the effective interval duration according to all crack interval durations in a single local area includes: Obtain the crack determination time point corresponding to the crack interval duration; Determine the data separation duration according to the crack determination time point and the current time point, and define the data separation duration with the smallest value as the reference separation duration; Calculate according to the data separation duration and the reference separation duration to determine the extension ratio; Determine the relative reliability coefficient corresponding to the extension ratio according to the preset reliable matching relationship, and calculate according to each relative reliability coefficient to determine the reliable calculation weight corresponding to each crack interval duration; Calculate according to each crack interval duration and the corresponding reliable calculation weight to determine the simulated interval duration; Judge whether the simulated interval duration is greater than the preset fixed interval duration; If the simulated interval duration is not greater than the fixed interval duration, determine the simulated interval duration as the effective interval duration; If the simulated interval duration is greater than the fixed interval duration, determine the fixed interval duration as the effective interval duration.

[0008] Optionally, after the demand flight distance is determined, the dam detection method based on the drone further includes: Define the local area that is not the demand detection area as the waiting area; Determine the detection coverage area according to the current simulated detection path and the preset unit coverage area; Define the area in the single waiting area that is within the detection coverage area as the attached area, determine the attached ratio according to the attached area and the waiting area, and define the waiting area with the attached ratio greater than the preset effective detection ratio as the effective area; Calculate according to the detection interval duration of the effective area and the effective interval duration to determine the duration interval ratio; Determine the simulated reduction distance corresponding to the duration interval ratio according to the preset alternative matching relationship, and sum up all the simulated reduction distances to determine the overall reduction distance; Calculate the difference based on the overall reduced distance and the required flight distance to update the required flight distance.

[0009] Optionally, after updating the required flight distance, the dam detection method based on the drone further includes: Determine whether there are at least two simulated detection paths with the same and minimum required flight distances; If there are not at least two simulated detection paths with the same and minimum required flight distances, define the simulated detection path corresponding to the minimum required flight distance as the used detection path; If there are at least two simulated detection paths with the same and minimum required flight distances, define the simulated detection path corresponding to the minimum required flight distance as the alternative detection path; In the alternative detection path, define the effective area with a time interval ratio greater than the preset boundary interval ratio as the high-value area; Count according to the high-value area to determine the high-value quantity, determine the high-value quantity with the largest value according to the sorting rule, and define the alternative detection path corresponding to the high-value quantity as the used detection path.

[0010] Optionally, after obtaining the dam detection image, the dam detection method based on the drone further includes: Perform feature recognition on the dam detection image to determine the dam crack feature; Determine whether there is an intersection between the dam crack feature and the contour line of the local area corresponding to the dam detection image; If the dam crack feature does not have an intersection with the contour line of the local area corresponding to the dam detection image, output a detection completion signal; If the dam crack feature has an intersection with the contour line of the local area corresponding to the dam detection image, define the local area on the remaining contour line with this intersection as the area to be measured; Define the area to be measured that is not the required detection area and the effective area as the secondary area, and construct a secondary operation path according to the secondary area, and control the drone to move along the secondary operation path to obtain the dam detection image again.

[0011] Optionally, after determining the secondary area, the dam detection method based on the drone further includes: Define the effective area with the dam crack feature as the crack area; Define the time interval ratio corresponding to the crack area as the crack presence ratio; Randomly select a crack presence ratio as the central presence ratio, and construct a ratio proximity range according to the central presence ratio and the preset proximity ratio; Count according to the proportion of cracks within a similar proportion range to determine the internal existence quantity, and count according to all the crack proportion to determine the overall existence quantity; Calculate based on the internal existence quantity and the overall existence quantity to determine the internal quantity proportion; Determine whether there is a situation where the internal quantity proportion is greater than the preset aggregation quantity proportion; If there is no situation where the internal quantity proportion is greater than the aggregation quantity proportion, then maintain the currently determined secondary area; If there is a situation where the internal quantity proportion is greater than the aggregation quantity proportion, then calculate the average value of the crack existence proportion within the proportion range corresponding to the largest internal quantity proportion to determine the advance average proportion; Determine the local area where the advance average proportion is greater than the corresponding time interval proportion and is not the area to be detected and the effective area as the secondary area.

[0012] Optionally, after the detection completion signal is output, the dam detection method based on the drone further includes: Obtain the number of detection areas and the number of abnormal areas; Calculate based on the number of detection areas and the number of abnormal areas to determine the abnormal detection proportion; Determine the detection evaluation coefficient corresponding to the abnormal detection proportion according to the preset evaluation matching relationship.

[0013] In a second aspect, the present application provides a dam detection system based on a drone, adopting the following technical solution: A dam detection system based on a drone, including: An acquisition module, configured to acquire the area detection time points of each preset local area on the dam; A processing module, connected to the acquisition module, for storing and processing information; The processing module determines the detection interval duration according to the area detection time point and the current time point; The processing module constructs a historical interval on the preset time axis with the current time point as the rear end point and a width of the preset historical duration, and obtains the crack interval durations of each local area in the historical interval; The processing module calculates according to all the crack interval durations under a single local area to determine the effective interval duration, and defines the local area where the detection interval duration is greater than the effective interval duration as the area to be detected; The processing module determines the area detection points on the area to be detected, randomly sorts the area detection points and connects them in series with the preset starting point to construct a simulated detection path, and determines the required flight distance according to the simulated detection path; The processing module determines the required flight distance with the smallest value according to a preset sorting rule, defines the simulation detection path corresponding to the required flight distance as the detection path to be used, and controls the drone to move along the detection path to be used at the starting point to obtain the dam detection image.

[0014] In summary, the present application includes at least one of the following beneficial technical effects: When using a drone to detect cracks in a dam on a daily basis, it is possible to detect areas with a higher likelihood of cracks at the current time point based on the crack frequencies that have occurred in each dam under historical conditions, thereby reducing the areas of ineffective detection and improving the detection efficiency for dam detection; During the process of selecting the drone path, fully consider the detection conditions of the areas passed by, so as to set a suitable path for the drone to fly, thereby achieving a better overall detection effect; Determine the second flight path based on the crack conditions determined by the first flight, thereby minimizing the situation of ineffective detection by the drone and improving the overall detection efficiency. Description of the Drawings

[0015] Figure 1 is a flowchart of a dam detection method based on a drone.

[0016] Figure 2 is a module flowchart of a dam detection method based on a drone. Detailed Embodiments

[0017] In order to make the objectives, technical solutions, and advantages of the present application clearer, the following is a further detailed description of the present application in conjunction with Figure 1 - Figure 2 and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] The following further describes the embodiments of the present application in detail with reference to the accompanying drawings of the specification.

[0019] The embodiments of the present application disclose a dam detection method based on a drone. Referring to Figure 1 , the method flow of the dam detection method based on a drone includes the following steps: Step S100: Obtain the regional detection time points of each preset local area on the dam.

[0020] The local area is an area for detecting cracks delimited by the staff according to the dam. When the drone hovers directly above the center point of the area, the area image can be exactly and completely collected. The specific division of the local area is determined by the staff in advance and will not be elaborated here; the regional detection time point is the time point when the local area was last detected for cracks using a drone.

[0021] Step S101: Determine the detection interval duration according to the regional detection time point and the current time point.

[0022] The detection interval duration is the duration elapsed since the last detection of the local area, that is, the time interval between the regional detection time point and the current time point.

[0023] Step S102: Construct a historical interval on the preset time axis with the current time point as the rear end point and a width of the preset historical duration, and obtain the crack interval durations of each local area in the historical interval.

[0024] The time axis is an axis formed by combining each time point. This axis points from the elapsed time points to the time points that have not yet been reached. Among them, the elapsed time points are on the left side of this time axis, and the left side is defined as the front end of the time axis; the historical duration is the duration set by the staff to obtain data on the historical crack conditions of the dam. By constructing the historical interval, it is convenient to obtain the data within the historical duration; the crack interval duration is the time interval between the time points of two adjacent crack occurrences detected in the local area within the historical interval.

[0025] Step S103: Calculate according to all the crack interval durations under a single local area to determine the effective interval duration, and define the local area where the detection interval duration is greater than the effective interval duration as the area requiring detection.

[0026] The effective interval duration is regarded as the interval duration of the general crack occurrence situation in this local area. It can be obtained by calculating the average value of all crack interval durations, or by the method from Step S200 to Step S2052; when the detection interval duration is greater than the effective interval duration, it indicates that the possibility of cracks in this area is relatively large. Therefore, it is defined as the area requiring detection to distinguish different local areas for subsequent analysis.

[0027] Step S104: Determine the regional detection points on the area requiring detection, randomly sort each regional detection point and connect it in series with the preset starting point to construct a simulated detection path, and determine the required flight distance according to the simulated detection path.

[0028] The area detection point is the position point where the overall image of the area to be detected can be obtained after the drone hovers. Generally, this area detection point is the center point of the area to be detected; the starting point is the position point when the drone has not started operating, which can be obtained by installing a positioning device on the drone. The simulated detection path is the path obtained by sequentially connecting all the sorted area detection points with the starting point as the first endpoint. That is, when the drone moves according to the simulated detection path, it can obtain the images of each area to be detected; the required flight distance is the path that the drone needs to move when flying and operating according to the simulated detection path.

[0029] Step S105: Determine the required flight distance with the smallest value according to the preset sorting rule, and define the simulated detection path corresponding to this required flight distance as the used detection path, and control the drone to move along the used detection path at the starting point to obtain the dam detection image.

[0030] The sorting rule is a method set by the staff to sort the numerical values, such as the bubble sort method. Through the sorting rule, the required flight distance with the smallest value can be determined. That is, when the drone moves and operates according to the simulated detection path corresponding to this required flight distance, the overall detection efficiency is the highest. At this time, it is defined as the used detection path to control the drone operation; among them, the dam detection image is the image of the dam surface collected by the drone during the movement.

[0031] The steps of calculating the effective interval duration according to all the crack interval durations in a single local area include: Step S200: Obtain the crack determination time point corresponding to the crack interval duration.

[0032] The crack determination time point is the time point of discovering the crack at the back end when determining the crack interval duration.

[0033] Step S201: Determine the data separation duration according to the crack determination time point and the current time point, and define the data separation duration with the smallest value as the reference separation duration.

[0034] The data separation duration is the time interval between the crack determination time point and the current time point. Defining the reference separation duration is to distinguish different data separation durations for subsequent analysis.

[0035] Step S202: Calculate according to the data separation duration and the reference separation duration to determine the extension ratio.

[0036] The extension ratio is the ratio obtained by dividing the data separation duration by the reference separation duration.

[0037] Step S203: Determine the relative reliability coefficient corresponding to the extension ratio according to the preset reliable matching relationship, and calculate according to each relative reliability coefficient to determine the reliable calculation weight corresponding to each crack interval duration.

[0038] The relative reliability coefficient is a parameter reflecting data reliability. The larger the delay ratio, the longer the time interval between the corresponding data acquisition time point and the current time point, and the lower the corresponding relative reliability coefficient at this time. The reliable matching relationship between the two is set by the staff in advance according to the actual situation; by summing up each relative reliability coefficient and then dividing each relative reliability coefficient by the sum value, the corresponding reliable calculation weight can be obtained.

[0039] Step S204: Calculate according to each crack interval duration and the corresponding reliable calculation weight to determine the simulated interval duration.

[0040] By multiplying each crack interval duration by the corresponding reliable calculation weight and then summing up, a more appropriate simulated interval duration that can reflect the interval duration of crack occurrence in this local area can be obtained.

[0041] Step S205: Determine whether the simulated interval duration is greater than the preset fixed interval duration.

[0042] The fixed interval duration is the longest duration that a single local area can be not detected as set by the staff. The purpose of the judgment is to know whether the determined simulated interval duration is too long.

[0043] Step S2051: If the simulated interval duration is not greater than the fixed interval duration, then determine the simulated interval duration as the effective interval duration.

[0044] When the simulated interval duration is not greater than the fixed interval duration, it means that the determined simulated interval duration at this time is the value for subsequent analysis. At this time, it can be determined as the effective interval duration.

[0045] Step S2052: If the simulated interval duration is greater than the fixed interval duration, then determine the fixed interval duration as the effective interval duration.

[0046] When the simulated interval duration is greater than the fixed interval duration, it means that the determined fixed interval duration at this time is the value for subsequent analysis. At this time, it can be determined as the effective interval duration.

[0047] After the required flight path is determined, the dam detection method based on the unmanned aerial vehicle further includes: Step S300: Define the local areas that are not the required detection areas as waiting areas.

[0048] Defining the waiting areas to identify the local areas that do not need to be detected currently is convenient for subsequent analysis.

[0049] Step S301: determining a detection coverage area according to a current simulated detection path and a preset unit coverage area.

[0050] The unit coverage area is the area that the drone can cover and detect when it is at one point, and the detection coverage area is the area that the drone can detect after moving along the simulated detection path.

[0051] Step S302: define the area within the detection coverage area of ​​a single waiting area as an incidental area, determine the incidental proportion based on the incidental area and the waiting area, and define the waiting area whose incidental proportion is greater than the preset effective detection proportion as an effective area.

[0052] The incidental area is defined to determine the area in a single waiting area that overlaps with the detection coverage area, that is, to determine the area where the drone is detected; the incidental ratio is the ratio of the incidental area in a single waiting area, which is determined by dividing the area of ​​the incidental area by the area of ​​the waiting area; the effective detection ratio is the minimum incidental ratio set by the staff to determine that the waiting area is well detected. When the incidental ratio is greater than the effective detection ratio, it means that the drone has conducted a relatively complete detection of the waiting area during the flight. Therefore, it is defined as a valid area to distinguish between different waiting areas for subsequent analysis.

[0053] Step S303: Calculate the duration interval ratio according to the detection interval duration of the effective area and the effective interval duration.

[0054] The duration interval ratio is the ratio of the detection interval duration determined in the effective area to the effective interval duration. The larger the value is, the more worthy the local area is to be detected.

[0055] Step S304: Determine the simulated distance reduction corresponding to the time interval ratio according to the preset candidate matching relationship, and perform sum calculation based on all the simulated distance reductions to determine the overall distance reduction.

[0056] The simulated distance reduction is a parameter value that objectively reflects whether the effective area is worthy of being detected. The larger the parameter value is, the more worthy the effective area is to be detected, and the corresponding time interval accounts for a larger proportion. The candidate matching relationship between the two is determined by the staff through multiple tests in advance; the overall distance reduction is the sum of all determined simulated distance reductions.

[0057] Step S305: Calculate the difference between the overall reduction distance and the required flight distance to update the required flight distance.

[0058] The updated demand flight distance can be achieved by subtracting the overall reduction distance from the demand flight distance, facilitating the subsequent selection of a more appropriate path for the UAV to fly.

[0059] After the demand flight distance is updated, the dam detection method based on the UAV further includes: Step S400: Determine whether there are at least two simulated detection paths with the same and minimum demand flight distance.

[0060] The purpose of the determination is to find out whether there are multiple simulated detection paths that meet the requirements, facilitating the determination of the only detection path for use.

[0061] Step S4001: If there are not at least two simulated detection paths with the same and minimum demand flight distance, define the simulated detection path corresponding to the minimum demand flight distance as the detection path for use.

[0062] When there are not at least two simulated detection paths with the same and minimum demand flight distance, it means there is only one simulated detection path that meets the requirements. At this time, it can be determined as the detection path for use.

[0063] Step S4002: If there are at least two simulated detection paths with the same and minimum demand flight distance, define the simulated detection path corresponding to the minimum demand flight distance as the alternative detection path.

[0064] When there are at least two simulated detection paths with the same and minimum demand flight distance, it means there are multiple simulated detection paths that meet the requirements. At this time, they are defined as alternative detection paths to distinguish different simulated detection paths, facilitating the subsequent determination of the only detection path for use.

[0065] Step S401: Define the effective area with a duration interval ratio greater than the preset boundary interval ratio in the alternative detection path as the high-value area.

[0066] The boundary interval ratio is the minimum duration interval ratio required when the staff sets that the value of detecting the effective area is relatively high. By defining the high-value area, the areas that are quite worthy of incidental detection are determined, facilitating subsequent analysis.

[0067] Step S402: Count according to the high-value area to determine the high-value quantity, determine the high-value quantity with the largest value according to the sorting rule, and define the alternative detection path corresponding to the high-value quantity as the detection path for use.

[0068] The high-value quantity is the total quantity of the determined high-value areas. Determine the high-value quantity with the largest numerical value to indicate that the corresponding alternative detection path is the most suitable for the UAV to perform mobile operations at this time, and then determine it as the detection path to be used.

[0069] After obtaining the dam detection image, the dam detection method based on the UAV further includes: Step S500: Perform feature recognition on the dam detection image to determine the dam crack features.

[0070] The dam crack features are the features of the obvious cracks on the dam that need to be repaired. The feature recognition method can construct a recognition database through sample learning in advance, and then input the current dam detection image into the recognition database to identify and determine the crack features.

[0071] Step S501: Determine whether the dam crack features intersect with the contour line of the local area corresponding to the dam detection image.

[0072] The purpose of the determination is to know whether the currently determined crack features may extend to other local areas.

[0073] Step S5011: If the dam crack features do not intersect with the contour line of the local area corresponding to the dam detection image, output a detection completion signal.

[0074] When the dam crack features do not intersect with the contour line of the local area corresponding to the dam detection image, it means that all the determined crack features do not extend to other local areas, that is, the current detection is completed. At this time, output a detection completion signal to identify this situation.

[0075] Step S5012: If the dam crack features intersect with the contour line of the local area corresponding to the dam detection image, define the local area with this intersection on the remaining contour line as the area to be detected.

[0076] When the dam crack features intersect with the contour line of the local area corresponding to the dam detection image, it means that the crack features extend to other local areas. At this time, define it as the area to be detected to distinguish different local areas and facilitate subsequent analysis.

[0077] Step S502: Define the area to be detected that is not the area to be detected and the effective area as the secondary area, and construct a secondary operation path according to the secondary area, and control the UAV to move along the secondary operation path to obtain the dam detection image again.

[0078] When the area to be measured is neither the area to be detected nor the valid area, it indicates that the area to be measured needs to be detected. At this time, it is defined as a secondary area for identification to facilitate subsequent analysis. The secondary operation path is the flight path along which the UAV can detect each secondary area. The determination method of this path is the same as the above-mentioned detection path and will not be elaborated here. By controlling the UAV to move along the secondary operation path, the crack conditions in the remaining local areas can be continuously detected. If secondary areas are still determined subsequently, the detection can continue until a detection completion signal is output.

[0079] After the secondary area is determined, the dam detection method based on the UAV further includes: Step S600: Define the valid area with dam crack features as the crack area.

[0080] Define the crack area to identify the valid area with dam crack features for subsequent analysis.

[0081] Step S601: Define the ratio of the time interval corresponding to the crack area to the total time interval as the crack presence ratio.

[0082] Define the crack presence ratio to distinguish different ratios of time intervals for subsequent analysis.

[0083] Step S602: Randomly select a crack presence ratio as the central presence ratio, and construct a similar ratio range based on the central presence ratio and a preset similar ratio.

[0084] The similar ratio is the maximum difference allowed when the two crack presence ratio values set by the staff are relatively close. The similar ratio range is the range allowed for the other crack presence ratios that are relatively close to the central presence ratio. This range is determined by adding and subtracting the similar ratio from the central presence ratio respectively.

[0085] Step S603: Count the crack presence ratios within the similar ratio range to determine the internal presence quantity, and count all the crack presence ratios to determine the overall presence quantity.

[0086] The internal presence quantity is the total amount of data of the crack presence ratios within the determined similar ratio range, and the overall presence quantity is the total amount of data of the determined crack presence ratios.

[0087] Step S604: Calculate based on the internal presence quantity and the overall presence quantity to determine the internal quantity ratio.

[0088] The internal quantity ratio is the ratio value of the data volume of the crack presence ratios within the similar ratio range to the data volume of all the crack presence ratios, and is determined by dividing the internal presence quantity by the overall presence quantity.

[0089] Step S605: Determine whether there is a situation where the internal quantity ratio is greater than the preset aggregation quantity ratio.

[0090] The aggregation quantity ratio is the minimum internal quantity ratio required to reach when the identified large number of cracks exists and is aggregated as set by the staff. The purpose of the determination is to know the specific situation of the currently detected effective area.

[0091] Step S6051: If there is no situation where the internal quantity ratio is greater than the aggregation quantity ratio, then maintain the currently determined secondary area.

[0092] When there is no situation where the internal quantity ratio is greater than the aggregation quantity ratio, it indicates that there is no situation of collective equal-proportion change in the effective area, that is, it is impossible to determine the area to be detected subsequently based on the situation of the first detection. At this time, just normally maintain the determined secondary area.

[0093] Step S6052: If there is a situation where the internal quantity ratio is greater than the aggregation quantity ratio, then calculate the average value of the crack existence ratios within the range of similar ratios corresponding to the internal quantity ratio with the largest value to determine the advance average ratio.

[0094] When there is a situation where the internal quantity ratio is greater than the aggregation quantity ratio, it indicates that there is a situation where a large number of crack existence ratios are aggregated, that is, there is an equal-proportion advance situation in the occurrence of cracks in the dam. At this time, the advance average ratio can be obtained through average value calculation.

[0095] Step S606: Determine the local area where the advance average ratio is greater than the corresponding time interval ratio and is not the area to be detected and the effective area as the secondary area.

[0096] When the advance average ratio is greater than the corresponding time interval ratio, it indicates that there is also a relatively large crack risk in this local area. Therefore, it is also determined as the secondary area for detection, so that the area with cracks can be covered as much as possible during the second detection.

[0097] After the detection completion signal is output, the dam detection method based on the unmanned aerial vehicle further includes: Step S700: Obtain the number of detected areas and the number of abnormal areas.

[0098] The number of detected areas is the number of local areas that have been completely detected during the overall detection process of the unmanned aerial vehicle, and the number of abnormal areas is the number of areas with crack characteristics detected.

[0099] Step S701: Calculate based on the number of detected areas and the number of abnormal areas to determine the abnormal detection ratio.

[0100] The abnormal detection ratio is the ratio of the area with crack features detected, which is determined by dividing the number of abnormal areas by the number of detection areas.

[0101] Step S702: Determine the detection evaluation coefficient corresponding to the abnormal detection ratio according to the preset evaluation matching relationship.

[0102] The detection evaluation coefficient is a parameter value reflecting the detection effect of the current detection. The larger this value, the more valuable the current detection. Among them, when the abnormal detection ratio is larger, it means the necessity of detection at this time is stronger, that is, the detection evaluation coefficient is larger. Subsequently, the staff can determine the detection situation through the detection evaluation coefficient.

[0103] Refer to Figure 2 , based on the same inventive concept, an embodiment of the present invention provides a dam detection system based on a drone, including: An acquisition module, configured to acquire the area detection time points of each preset local area on the dam; A processing module, connected to the acquisition module, for storing and processing information; The processing module determines the detection interval duration according to the area detection time point and the current time point; The processing module constructs a historical interval on the preset time axis with the current time point as the rear end point and a width of the preset historical duration, and obtains the crack interval durations of each local area in the historical interval; The processing module calculates according to all the crack interval durations under a single local area to determine the effective interval duration, and defines the local area where the detection interval duration is greater than the effective interval duration as the area to be detected; The processing module determines the area detection points on the area to be detected, randomly sorts the area detection points and connects them in series with the preset starting point to construct a simulated detection path, and determines the required flight distance according to the simulated detection path; The processing module determines the required flight distance with the smallest value according to the preset sorting rule, and defines the simulated detection path corresponding to the required flight distance as the used detection path, and controls the drone to move along the used detection path at the starting point to obtain the dam detection image; An effective interval duration determination module, configured to determine a suitable effective interval duration for data analysis; A required flight distance update module, which updates the required flight distance according to the detection situation during the flight of the drone; A simulated detection path screening module, configured to screen multiple simulated detection paths that meet the requirements; A secondary detection control module, configured to determine the path of the secondary detection of the drone to control the operation of the drone; The secondary area adding module adds the secondary area according to the specific conditions of each local area; The detection result evaluation module is used to evaluate the overall detection situation of the dam.

[0104] Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above division of each functional module is used as an example. In actual applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working processes of the system, device, and unit described above, reference can be made to the corresponding processes in the foregoing method embodiments, and details are not described herein again.

Claims

1. A dam detection method based on drone, characterized in that: include: Obtaining regional detection time points of each preset local area on the dam; Determine the detection interval duration based on the regional detection time point and the current time point; Constructing a historical interval with the current time point as the rear end point and a width of the preset historical duration on the preset time axis, and obtaining the crack interval duration of each local area in the historical interval; In a single local area, the effective interval duration is calculated based on all crack interval durations, and the local area where the detection interval duration is longer than the effective interval duration is defined as the required detection area; Determine regional detection points on the required detection area, and randomly sort the regional detection points and connect them in series with a preset starting point to construct a simulated detection path, and determine the required flight distance based on the simulated detection path; The required flight distance with the minimum value is determined according to the preset sorting rules, and the simulated detection path corresponding to the required flight distance is defined as the used detection path, and the UAV is controlled to move along the used detection path at the starting point to obtain the dam detection image.

2. The dam detection method based on drone according to claim 1 is characterized in that: At The steps of calculating the effective interval duration based on all crack interval durations in a single local area include: Obtain the crack determination time point corresponding to the crack interval duration; Determine the data interval length according to the crack determination time point and the current time point, and define the data interval length with the smallest value as the benchmark interval length; The extension ratio is determined by calculating the data interval length and the benchmark interval length; The relative reliability coefficient corresponding to the extension ratio is determined according to the preset reliable matching relationship, and the reliable calculation weight corresponding to each crack interval duration is determined according to each relative reliability coefficient; The simulation interval is determined by calculating the interval length of each crack and the corresponding reliable calculation weight; Determine whether the simulation interval duration is greater than the preset fixed interval duration; If the simulation interval duration is not greater than the fixed interval duration, the simulation interval duration is determined as the effective interval duration; If the simulation interval duration is greater than the fixed interval duration, the fixed interval duration is determined as the effective interval duration.

3. The dam detection method based on drone according to claim 1 is characterized in that: After the required flight distance is determined, the dam inspection method based on drones also includes: A local area that is not a required detection area is defined as a waiting area; Determine the detection coverage area according to the current simulated detection path and the preset unit coverage area; The area within the detection coverage area of ​​a single waiting area is defined as an incidental area, and the incidental ratio is determined according to the incidental area and the waiting area, and the waiting area whose incidental ratio is greater than the preset effective detection ratio is defined as an effective area; The time interval ratio is determined by calculating the detection interval duration of the effective area and the effective interval duration; The simulated distance reduction corresponding to the time interval ratio is determined according to the preset candidate matching relationship, and the overall distance reduction is determined by summing up all the simulated distance reductions; The required flight distance is updated by performing a difference calculation based on the overall reduction distance and the required flight distance.

4. The dam detection method based on drone according to claim 3 is characterized in that: After the required flight route is updated, the dam inspection method based on drones also includes: Determine whether there are at least two simulated detection paths with the same required flight distance and the minimum required flight distance; If there are not at least two simulated detection paths with the same required flight distance and the minimum required flight distance, the simulated detection path corresponding to the minimum required flight distance is defined as the used detection path; If there are at least two simulated detection paths with the same required flight distance and the smallest required flight distance, the simulated detection path corresponding to the smallest required flight distance is defined as the alternative detection path; In the alternative detection path, a valid area where the time interval ratio is greater than the preset boundary interval ratio is defined as a high-value area; The high-value areas are counted to determine the high-value quantity, and the high-value quantity with the largest value is determined according to the sorting rule, and the alternative detection path corresponding to the high-value quantity is defined as the detection path to be used.

5. The dam detection method based on drone according to claim 4 is characterized in that: After the dam inspection image is obtained, the dam inspection method based on drone also includes: Perform feature recognition based on the dam inspection image to determine the dam crack characteristics; Determine whether the dam crack feature has an intersection with the contour line of the local area corresponding to the dam detection image; If the crack feature of the dam does not intersect with the contour line of the local area corresponding to the dam detection image, a detection completion signal is output; If there is an intersection between the dam crack feature and the contour line of the local area corresponding to the dam detection image, the local area with the intersection on the remaining contour lines is defined as the area to be detected; The area to be tested that is not a required detection area or a valid area is defined as a secondary area, and a secondary operation path is constructed based on the secondary area, and the UAV is controlled to move along the secondary operation path to obtain the dam detection image again.

6. The dam detection method based on drone according to claim 5 is characterized in that: After the secondary area is determined, the drone-based dam inspection method also includes: The effective area where the embankment crack characteristics exist is defined as the crack area; The proportion of time intervals corresponding to the crack area is defined as the crack existence proportion; A crack existence ratio is randomly selected as the center existence ratio, and a similar ratio range is constructed based on the center existence ratio and the preset similar ratio; Count the proportions of cracks in a similar range to determine the number of cracks inside, and count all the proportions of cracks to determine the overall number of cracks. Calculate the internal quantity and the overall quantity to determine the internal quantity percentage; Determine whether the internal quantity ratio is greater than the preset aggregation quantity ratio; If there is no situation where the proportion of internal quantity is greater than the proportion of aggregate quantity, the currently determined secondary area is maintained; If the proportion of internal quantity is greater than the proportion of aggregate quantity, the proportion of cracks in the range close to the proportion of the largest internal quantity is averaged to determine the advance average proportion; The local area whose average advance proportion is greater than the corresponding time interval proportion and is not a demand detection area or a valid area is determined as a secondary area.

7. The dam detection method based on drone according to claim 6 is characterized in that: After the detection signal is output, the dam detection method based on drone also includes: Get the number of detected areas and the number of abnormal areas; Calculate the number of detection areas and the number of abnormal areas to determine the abnormal detection ratio; The detection evaluation coefficient corresponding to the abnormal detection ratio is determined according to the preset evaluation matching relationship.

8. A dam inspection system based on drones, characterized in that: include: An acquisition module is used to acquire the regional detection time points of each preset local area on the dam; A processing module, connected to the acquisition module, for storing and processing information; The processing module determines the detection interval duration according to the regional detection time point and the current time point; The processing module constructs a historical interval with the current time point as the rear end point and a width of the preset historical duration on the preset time axis, and obtains the crack interval duration of each local area in the historical interval; The processing module calculates the effective interval duration according to all crack interval durations in a single local area, and defines the local area where the detection interval duration is greater than the effective interval duration as the required detection area; The processing module determines regional detection points on the required detection area, and randomly sorts the regional detection points and connects them in series with a preset starting point to construct a simulated detection path, and determines the required flight distance according to the simulated detection path; The processing module determines the required flight distance with the minimum value according to a preset sorting rule, defines the simulated detection path corresponding to the required flight distance as the used detection path, and controls the UAV to move along the used detection path at the starting point to obtain the dam detection image.

Images