CN113741527B - Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles - Google Patents
Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles Download PDFInfo
- Publication number
- CN113741527B CN113741527B CN202111067758.3A CN202111067758A CN113741527B CN 113741527 B CN113741527 B CN 113741527B CN 202111067758 A CN202111067758 A CN 202111067758A CN 113741527 B CN113741527 B CN 113741527B
- Authority
- CN
- China
- Prior art keywords
- inspection
- oil well
- image
- area
- unmanned aerial
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 314
- 239000003129 oil well Substances 0.000 title claims abstract description 198
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000001914 filtration Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims description 22
- 230000006870 function Effects 0.000 claims description 17
- 230000009471 action Effects 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 13
- 230000009466 transformation Effects 0.000 claims description 13
- 238000000354 decomposition reaction Methods 0.000 claims description 12
- 238000012937 correction Methods 0.000 claims description 10
- 238000007621 cluster analysis Methods 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000010845 search algorithm Methods 0.000 claims description 8
- 230000009467 reduction Effects 0.000 claims description 7
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000003860 storage Methods 0.000 claims description 4
- 239000003208 petroleum Substances 0.000 abstract description 16
- 238000004519 manufacturing process Methods 0.000 abstract description 10
- 238000004458 analytical method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 210000001367 artery Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009954 braiding Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005266 casting Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011900 installation process Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/10—Simultaneous control of position or course in three dimensions
- G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
- G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the specification provides an oil well inspection method based on multiple unmanned aerial vehicles, which belongs to the technical field of petroleum production, and comprises the following steps: acquiring an oil well inspection task, wherein the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected; dividing an area to be inspected of an oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles for each inspection subarea; determining the optimal flight path of each unmanned aerial vehicle in the patrol sub-area; controlling the unmanned aerial vehicle to acquire images in the inspection subarea according to the optimal flight path, and acquiring an initial shooting image of the oil well; according to the position coordinates of the initial shooting image, filtering the overlapped area of each patrol sub-area, and obtaining the clear image of each patrol sub-area as an image to be analyzed; inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
Description
Technical Field
The specification relates to the technical field of petroleum production, in particular to an oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles.
Background
The petroleum industry is a high risk and high yielding industry, so the safe production of petroleum is the basis and premise for the production and development of the petroleum industry. The oil well equipment operates throughout the year on the petroleum exploitation site, and is easy to damage and corrode due to the loss of the mechanical equipment and the influence of natural factors such as wind, rain, lightning and the like, if the equipment is not found and eliminated in time, various faults can be developed, and the safety and stability of the oil well are greatly threatened. Meanwhile, due to the precious and high economic value of petroleum resources, illegal molecules have the phenomenon of stealing petroleum resources, and great economic loss is brought to the oil field. Therefore, it is very important to ensure the timing and fixed-point inspection of the oil well.
Because the well spacing of the production wells of the oil and gas field is dispersed, the span is large. And the environment is severe, and the road of the well site is possibly phagocytized at any time in the sand storm weather, so that the difficulty of well inspection and line inspection is high, and the workload of personnel and the risk of safe production are increased based on the modes of manual inspection and semi-manual inspection. In the prior art, unmanned aerial vehicle carries high-definition image equipment to carry out oil well inspection, need not personnel to operate closely for traditional mode of inspecting, comparatively safe and reliable. However, with the increasing of task complexity and environment differentiation, the requirements on the performance of the unmanned aerial vehicle are increasing, a great deal of time is required to be consumed when a single unmanned aerial vehicle performs inspection under a multi-oil-well scene, the instantaneity of inspection information is low, the execution task is single, the inspection efficiency of an oil well is low, and the safety condition of the oil well cannot be mastered in real time.
Based on the above, an unmanned aerial vehicle inspection method capable of improving the inspection efficiency and real-time performance of an oil well is needed.
Disclosure of Invention
One or more embodiments of the present disclosure provide a method, an apparatus, and a device for oil well inspection based on multiple unmanned aerial vehicles, which are used for solving the following technical problems: how to provide an unmanned aerial vehicle inspection method capable of improving the inspection efficiency and inspection accuracy of an oil well.
One or more embodiments of the present disclosure adopt the following technical solutions:
one or more embodiments of the present specification provide a multi-unmanned aerial vehicle-based oil well inspection method, including:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
Filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
Optionally, in one or more embodiments of the present disclosure, the dividing the area to be inspected of the oil well inspection task into a plurality of inspection sub-areas specifically includes:
acquiring historical inspection actions of all oil well equipment in the area to be inspected;
performing cluster analysis on the historical inspection actions based on a cluster algorithm to obtain a cluster center of the historical inspection actions within a preset distance threshold; wherein, the history inspection action includes: inspection of suspicious personnel, oil leakage inspection of oil well equipment, inspection of dangerous objects and inspection of combustible objects;
and dividing the region to be inspected according to the clustering range of the clustering center to obtain a plurality of inspection sub-regions.
Optionally, in one or more embodiments of the present disclosure, the determining, according to the determining, a preset optimal flight path in the multiple sub-areas specifically includes:
Determining initial oil well coordinates and target oil well coordinates of the unmanned aerial vehicle according to the coordinates of the oil wells in the inspection area and the importance level of the oil wells;
determining the flying height of the unmanned aerial vehicle based on the inspection task of the unmanned aerial vehicle;
intercepting a two-dimensional plan of the patrol sub-area according to the flying height;
determining a flight area of the unmanned aerial vehicle based on the two-dimensional plan;
determining an unmanned aerial vehicle passing path in the patrol sub-area according to the flight area of the unmanned aerial vehicle, the starting point oil well coordinates and the target oil well coordinates of the unmanned aerial vehicle;
and analyzing the unmanned aerial vehicle passing path based on a variable neighborhood search algorithm to obtain the optimal flight path of the patrol sub-area.
Optionally, in one or more embodiments of the present disclosure, after the determining the preset optimal flight path in the plurality of sub-areas, the method further includes:
selecting a plurality of shooting positions in the optimal flight path of each patrol sub-area;
and setting shooting angles and shooting parameters at each shooting position of the unmanned aerial vehicle according to the shooting positions and the range of the patrol sub-area, so that the shooting images cover preset fault patrol points of the oil well in the patrol sub-area.
Optionally, in one or more embodiments of the present disclosure, before the inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, the method further includes:
performing gray level correction and self-adaptive histogram equalization processing on the images to be analyzed of each inspection subarea to obtain corrected images of the images to be analyzed;
processing the corrected image based on a preset filtering mode, removing noise data of the corrected image, and obtaining a denoising image;
and taking the denoising image as an image to be analyzed meeting the requirements of each inspection subarea.
Optionally, in one or more embodiments of the present disclosure, the processing the corrected image based on a preset filtering manner, removing noise data of the corrected image, and obtaining a denoised image specifically includes:
determining parameters of wavelet transformation according to the estimated noise intensity of the corrected image; wherein the parameters include: wavelet basis function, decomposition layer number and threshold function;
decomposing the correction image according to the parameters of the wavelet transformation to obtain a high-frequency component and a low-frequency component of the correction image;
Projecting the high-frequency component into a low-frequency space by using a Gaussian random matrix to obtain a measured value after filtering a noise vector;
and carrying out wavelet inverse transformation on the low-frequency component and the measured value so as to reconstruct the denoising image.
Optionally, in one or more embodiments of the present disclosure, after the acquiring the clear image of each inspection sub-area as the image to be analyzed, the method further includes:
determining an optimal splicing mode among images in the patrol sub-areas based on a minimum-cut maximum flow principle and an overlapping area backed up by each patrol sub-area;
splicing the images to be analyzed of the patrol sub-area according to the optimal splicing mode to obtain an initial spliced image of the patrol sub-area;
and fusing the spliced images through a multi-resolution technology to obtain a panoramic image of the area to be inspected so as to realize task statistics of unmanned aerial vehicle inspection.
Optionally, in one or more embodiments of the present disclosure, before the inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, the method further includes:
acquiring a historical inspection image of an area to be inspected, and generating a sample library of the area to be inspected; wherein, the history inspection image includes: coordinates of the inspection oil well, fault points of the inspection oil well, potential safety hazard position labels and non-staff areas;
Dividing the historical inspection image according to the range of the inspection subarea to be used as an image sample;
extracting and decomposing a feature vector of the image sample, and performing dimension reduction decomposition on the feature vector to serve as a training sample;
training the training sample as input, and the potential safety hazard and position coordinates of the oil well as output pairs to obtain a recognition result;
and selecting the model with small recognition result error as the potential oil well hazard recognition network model.
One or more embodiments of the present specification provide an oil well inspection apparatus based on a multi-unmanned aerial vehicle, the apparatus comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
Determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
One or more embodiments of the present specification provide a non-volatile computer storage medium storing computer-executable instructions configured to:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
Determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
The above-mentioned at least one technical scheme that this description embodiment adopted can reach following beneficial effect:
based on many unmanned aerial vehicles patrol and examine in each subregion to patrol and examine the region, solve single unmanned aerial vehicle and patrol and examine the region and carry out the problem of time spent in the working at the oil well in a large scale. Through selecting the analysis path of a plurality of unmanned aerial vehicles, the unmanned aerial vehicles can cover the to-be-inspected area of the oil well in the shortest time, and the problem of the oil well inspection efficiency is solved. Meanwhile, through a preset oil well hidden danger identification network model, the image acquired by the multiple unmanned aerial vehicles is processed and analyzed, so that the identification of the oil well hidden danger can be realized, and the safety and reliability of an oil well inspection area are ensured.
Drawings
In order to more clearly illustrate the embodiments of the present description or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some of the embodiments described in the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. In the drawings:
FIG. 1 is a schematic flow diagram of a method for multi-unmanned aerial vehicle-based oil well inspection method according to one or more embodiments of the present disclosure;
FIG. 2 is a schematic illustration of a multi-unmanned aerial vehicle-based oil well inspection apparatus according to one or more embodiments of the present disclosure;
FIG. 3 is a non-volatile storage medium provided by one or more embodiments of the present description.
Detailed Description
The embodiment of the specification provides an oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles.
In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.
Petroleum is an energy artery, and ensuring safe operation of oil well equipment is an important responsibility of enterprises. The oil well equipment operates throughout the year on the petroleum exploitation site, and is easy to damage and corrode due to the loss of the mechanical equipment and the influence of natural environment factors such as wind, rain, lightning and the like, if the equipment is not found and eliminated in time, various faults can be developed, and the safety of the oil field is greatly threatened. Meanwhile, due to the precious and high economic value of petroleum resources, illegal molecules occur during the phenomenon of petroleum resource theft, and great economic loss is brought to an oil field, so that the inspection of the area of oil well equipment is very important.
In the traditional inspection mode, important operators for inspection are people whether the inspection is performed manually or semi-automatically. At present, part of oil fields are also distributed in areas such as gobi, salt marsh, high mountains, ravines, low-lying areas and the like, the geographic environment is complex, and the production and maintenance management of ground equipment after casting completely depend on manual inspection, so that the workload and difficulty of personnel are increased, the input cost is greatly increased, the normal production of the oil fields is influenced, and the requirements of digital oil field construction cannot be met. Therefore, unmanned aerial vehicle aerial photography inspection becomes a new oil well inspection mode in the prior art with characteristics such as low cost, high accuracy. However, as the complexity of tasks and the diversity of environments are continuously increased, the requirements on the performance of the unmanned aerial vehicle are also increasing, and the single unmanned aerial vehicle cannot feed back real-time information of a large-scale inspection task, so that the accuracy of inspection information acquired by the single unmanned aerial vehicle cannot meet the requirements. In addition, single unmanned aerial vehicle still has the shortcoming that the execution task is single, inefficiency, risk are big, consequently can't ensure the security of oil well equipment and oil well property.
In order to solve the technical problems, the specification provides an oil well inspection method based on multiple unmanned aerial vehicles. By dividing the oil well inspection tasks, different inspection tasks can cooperate through different multiple unmanned aerial vehicles, so that the accurate matching of the inspection tasks is realized, and the problem that the single unmanned aerial vehicle can not meet the multi-target inspection tasks due to single execution task is solved. Meanwhile, the problem of low efficiency of a single unmanned aerial vehicle in inspection is solved by adopting cooperative work of multiple unmanned aerial vehicles, and accuracy of data acquisition is improved. By planning the inspection path of multiple unmanned aerial vehicles, an optimal path of the inspection process of the unmanned aerial vehicles is obtained, the inspection time of the unmanned aerial vehicles is shortened, and the energy consumption in the inspection process is reduced, so that the inspection cost is reduced. By processing and analyzing the unmanned aerial vehicle inspection image, the real-time monitoring of potential safety hazards of the oil well is realized, and the safety of oil well equipment and property is ensured.
The embodiment of the present specification is executed by the internal unit of the server controlling the unmanned aerial vehicle inspection or the server controlling the unmanned aerial vehicle inspection.
The technical scheme provided in the specification is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, in one or more embodiments of the present disclosure, a flow diagram of a multi-unmanned-plane-based oil well inspection method is provided.
As shown in fig. 1, the method comprises the steps of:
step 101: acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected.
For oil and gas engineering, the production coordination in the initial oil exploration, design and subsequent large-drill platform site selection, oil and gas pipeline laying and installation processes all need high-precision and time-efficient images to support analysis planning. In the production process of petroleum, because oil well equipment operates throughout the year in petroleum exploitation sites, the damage and corrosion of the oil well equipment are easily caused by the loss of the mechanical equipment and the influence of natural factors such as weather, if the damage and corrosion are not found and eliminated in time, the safety and stability of the oil field can be threatened, in addition, because the economic value of petroleum is higher, the phenomenon of petroleum stealing frequently occurs, and great loss is caused to the economy of the oil field.
Therefore, in the inspection process of an oil well, various inspection works with different targets are required for the oil field based on various common potential safety hazards in the oil field and the oil field data which need to be collected and analyzed. And acquiring position coordinates of different inspection areas contained in the oil well inspection task, so that the targets of the inspection task are clearer.
Step 102: dividing the region to be inspected of the oil well inspection task into a plurality of inspection sub-regions, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection sub-region.
In one or more embodiments of the present disclosure, the dividing the area to be inspected of the oil well inspection task into a plurality of inspection sub-areas specifically includes:
acquiring historical inspection actions of all oil well equipment in the area to be inspected;
performing cluster analysis on the historical inspection actions based on a cluster algorithm to obtain a cluster center of the historical inspection actions within a preset distance threshold; wherein, the history inspection action includes: inspection of suspicious personnel, oil leakage inspection of oil well equipment, inspection of dangerous objects and inspection of combustible objects;
and dividing the region to be inspected according to the clustering range of the clustering center to obtain a plurality of inspection sub-regions.
Based on the oil well inspection tasks obtained in the step 101, the areas to be inspected of each oil well inspection task and the corresponding coordinate positions of the oil wells are obtained. Based on the to-be-inspected area contained in the inspection task, the historical inspection data of each oil well device in the to-be-inspected area can be searched and called through the database. After the historical inspection data is subjected to cluster analysis through a clustering algorithm, the clusters of various inspection tasks performed in the past time in the area to be inspected can be obtained, namely, the range in which each inspection task or each potential safety hazard occurs in the area to be inspected is obtained. For example: the inspection tasks of suspicious personnel are concentrated in the range near the oil tank of the oil well equipment, the oil leakage inspection of the oil well occurs near the turning node of each oil pipe line, and the inspection tasks of dangerous objects and combustible objects are in the threshold range of each oil well equipment. Based on the determination of the cluster center and the cluster range of each inspection task by cluster analysis, the division of inspection scenes of the area to be inspected is realized, and a plurality of inspection subareas with targeted inspection tasks in the area to be inspected are obtained.
The following description is needed: in one or more embodiments of the present disclosure, the performing cluster analysis on the obtained historical inspection actions based on the clustering algorithm specifically includes:
taking each inspection task as a sample of cluster analysis, and determining an inspection range threshold value of each inspection task;
selecting any one of the inspection tasks as a cluster center of the area to be inspected;
calculating the distance between each inspection task in the area to be inspected and the center of the cluster, and selecting the minimum distance as a marking value of the inspection task;
selecting the inspection task with the largest value from the marking values of the inspection tasks, and taking the inspection task as a newly-added cluster center if the inspection task exceeds the inspection range threshold;
iteratively obtaining all cluster centers of the area to be inspected;
dividing according to the preset range of each cluster center to obtain a plurality of inspection subareas of the area to be inspected.
Through the cluster analysis of the area to be inspected, each inspection task can be divided, so that the range of the cluster does not exceed the threshold range of the inspection task, the targeted inspection of multiple unmanned aerial vehicles on each task is ensured, the problem of confusion of the inspection task caused by large-scale inspection is avoided, meanwhile, all the inspection tasks are effectively divided, and the problems of omission of the inspection task and single execution task are avoided.
Step 103: and determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas.
In one or more embodiments of the present disclosure, the determining an optimal flight path of each of the plurality of inspection areas specifically includes:
determining initial oil well coordinates and target oil well coordinates of the unmanned aerial vehicle according to the coordinates of the oil wells in the inspection area and the importance level of the oil wells;
determining the flying height of the unmanned aerial vehicle based on the inspection task of the unmanned aerial vehicle;
intercepting a two-dimensional plan of the patrol sub-area according to the flying height;
determining a flight area of the unmanned aerial vehicle based on the two-dimensional plan;
determining an unmanned aerial vehicle passing path in the patrol sub-area according to the flight area of the unmanned aerial vehicle, the starting point oil well coordinates and the target oil well coordinates of the unmanned aerial vehicle;
and analyzing the unmanned aerial vehicle passing path based on a variable neighborhood search algorithm to obtain the optimal flight path of the patrol sub-area.
In one or more embodiments of the present disclosure, after the determining the optimal flight path of each drone in the plurality of inspection sub-areas, the method further includes:
selecting a plurality of shooting positions in the optimal flight path of each patrol sub-area;
And setting shooting angles and shooting parameters at each shooting position of the unmanned aerial vehicle according to the shooting positions and the range of the patrol sub-area, so that the shooting images cover preset fault patrol points of the oil well in the patrol sub-area.
After the unmanned aerial vehicle takes off for a period of time, the unmanned aerial vehicle enters a stage of stable flight. Based on the precision required by the inspection task, the flight speed and the flight height of the unmanned aerial vehicle are basically kept constant when the unmanned aerial vehicle stably flies. Therefore, based on the flying height of the unmanned aerial vehicle, the two-dimensional plan view of the patrol sub-area under the flying height is intercepted, and the two-dimensional plan view of the unmanned aerial vehicle at the flying height can be obtained. At the moment, the flight obstacle of the unmanned aerial vehicle can be identified by using geometric figures, so that a complex analysis process during path planning in a three-dimensional space is avoided. Through the analysis of the flight height two-dimensional plan, the unmanned aerial vehicle barrier-free flight area can be determined, and a plurality of passing paths of the unmanned aerial vehicle under the flight height can be obtained based on the starting point oil well coordinates and the target oil well coordinates of the unmanned aerial vehicle in the unmanned aerial vehicle flight area. And analyzing a plurality of communication paths through a variable neighborhood search algorithm to obtain the optimal flight path of the unmanned aerial vehicle in the inspection subarea, so that the unmanned aerial vehicle can realize the oil well inspection task in the shortest path, and the oil well inspection cost is saved.
After the flight path of the unmanned aerial vehicle is obtained, the plane position of the unmanned aerial vehicle flight is required to be determined according to the clustering center dividing the subareas and the flight height of the unmanned aerial vehicle. And taking the arithmetic average value of the positions of all the oil wells, which are required to be inspected, of the unmanned aerial vehicle in each inspection subarea as the center position, which is required to be photographed, of the unmanned aerial vehicle inspection. And selecting a plurality of shooting positions, setting shooting angles and shooting parameters of the shooting positions according to the range of the inspection subarea, and aligning from the boundary of the inspection subarea to the shot central position, so as to ensure that the shot image can cover all preset fault inspection points. Through the setting of unmanned aerial vehicle shooting position and shooting parameter, can cover whole scene when having guaranteed to shoot the attention object, avoided because treat the scene segmentation that the scope of patrolling and examining caused based on cluster analysis to lead to the incomplete problem of patrolling and examining.
The following description is needed: the nature of the varying neighborhood search algorithm is an improved local search algorithm. The neighborhood in the variable neighborhood search algorithm refers to the set of all solutions obtained for one operation on the current solution, so different neighborhood actions result in different neighbors. The variable neighborhood search algorithm generates corresponding neighborhoods by using different actions and searches in the neighborhoods alternately, so that the algorithm is balanced in evacuability and concentration.
Step 104: and controlling the unmanned aerial vehicle to acquire images in the inspection subarea according to the optimal flight path, and acquiring an initial shooting image of the oil well.
And (3) acquiring the optimal flight path of the unmanned aerial vehicle in each inspection subarea according to step 103, and shooting based on the optimal flight path and the set shooting points to obtain an initial shooting image capable of covering the inspection subarea.
Step 105: and determining the overlapping area of the initial shooting images of the unmanned aerial vehicle in each inspection area according to the position coordinates of the initial shooting images.
When unmanned aerial vehicle shoots according to the shooting position that sets up, because shooting angle needs to ensure that inspection fault point is shot, therefore the unavoidable problem that there is the initial shooting image that unmanned aerial vehicle was shot in a plurality of shooting positions that the shooting content part overlaps exists, or there is overlap region in the initial shooting image of a plurality of unmanned aerial vehicle exists. The image shot by the unmanned aerial vehicle contains the position information of each point in the image, and the overlapping area shot by the unmanned aerial vehicle in each inspection area can be determined through the position coordinates contained in the initial shot image.
Step 106: and filtering the overlapped area of each inspection subarea, and obtaining the clear image of each inspection subarea as an image to be analyzed.
In one or more embodiments of the present disclosure, after the acquiring the clear image of each inspection sub-area as the image to be analyzed, the method further includes:
determining an optimal splicing mode among images in the patrol sub-areas based on a minimum-cut maximum flow principle and an overlapping area backed up by each patrol sub-area;
splicing the images to be analyzed of the patrol sub-area according to the optimal splicing mode to obtain an initial spliced image of the patrol sub-area;
and fusing the spliced images through a multi-resolution technology to obtain a panoramic image of the area to be inspected so as to realize task statistics of unmanned aerial vehicle inspection.
By effectively filtering the overlapping area acquired based on the step 105, after the overlapping area is backed up, one clear image is reserved as a subsequent image to be analyzed, and the redundant information contained in the overlapping area is avoided through filtering the overlapping area, so that the problem of high calculation complexity is solved.
In addition, by analyzing the backup of the overlapping area based on the principle of minimum cut maximum flow, the optimal splicing mode of the image pieces in each inspection area, namely the optimal splicing seam in each inspection area, can be determined. And splicing the images to be analyzed through the acquired optimal splicing seam, so that an initial spliced image of the inspection subarea can be obtained. And the spliced images can be fused based on a multi-resolution technology, so that the resolutions of the spliced images are consistent, a panoramic image of unmanned aerial vehicle inspection in an inspection area is obtained, and overall task statistics of the unmanned aerial vehicle in the inspection area can be realized based on the panoramic image. The problem of unavoidable registration errors caused by internal or external reasons such as illumination in actual conditions is solved, and meanwhile, the accuracy of the inspection task is improved by reducing the registration errors.
In one embodiment of the present disclosure, if two photos with overlapping areas are respectively an image a and an image B, based on the principle of minimum cut maximum flow and the overlapping area backed up by each patrol sub-area, the optimal stitching manner between the images in the patrol sub-area is determined as follows: pixels from image a are denoted as connection source points and pixels from image B are denoted as connection sink points. The image is divided into two sub-images respectively containing a source point and a sink point by adopting a network maximum flow method, and the weighted sum of the braiding of the two sub-images is minimum, so that the maximum flow at the moment corresponds to the minimum cut, and the minimum cut is the optimal splicing line of the image A and the image B. The method of the network maximum flow is a prior art means, and will not be described herein.
Step 107: inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
In one or more embodiments of the present disclosure, before the inputting the image to be analyzed meeting the requirements of each inspection subarea into the preset oil well hidden danger identification network model, the method further includes:
Performing gray level correction and self-adaptive histogram equalization processing on the images to be analyzed of each inspection subarea to obtain corrected images of the images to be analyzed;
processing the corrected image based on a preset filtering mode, removing noise data of the corrected image, and obtaining a denoising image;
and taking the denoising image as an image to be analyzed meeting the requirements of each inspection subarea.
In one or more embodiments of the present disclosure, the processing the corrected image based on a preset filtering manner, removing noise data of the corrected image, and obtaining a denoised image specifically includes:
determining parameters of wavelet transformation according to the estimated noise intensity of the corrected image; wherein the parameters include: wavelet basis function, decomposition layer number and threshold function;
decomposing the correction image according to the parameters of the wavelet transformation to obtain a high-frequency component and a low-frequency component of the correction image;
projecting the high-frequency component into a low-frequency space by using a Gaussian random matrix to obtain a measured value after filtering a noise vector;
and carrying out wavelet inverse transformation on the low-frequency component and the measured value so as to reconstruct the denoising image.
In one or more embodiments of the present disclosure, before the image to be analyzed that meets the requirements of each inspection subarea is input into the preset oil well hidden danger identification network, the method further includes:
acquiring a historical inspection image of an area to be inspected, and generating a sample library of the area to be inspected; wherein, the history inspection image includes: coordinates of the inspection oil well, fault points of the inspection oil well, potential safety hazard position labels and non-staff areas;
dividing the historical inspection image according to the range of the inspection subarea to be used as an image sample;
extracting and decomposing a feature vector of the image sample, and performing dimension reduction decomposition on the feature vector to serve as a training sample;
training the training sample as input, and the potential safety hazard and position coordinates of the oil well as output pairs to obtain a recognition result;
and selecting the model with small recognition result error as the potential oil well hazard recognition network model.
After the images to be analyzed of each inspection subarea are obtained, the images of the images to be analyzed are corrected by carrying out gray level correction and self-adaptive histogram equalization on the images to be analyzed of each inspection subarea. Based on the estimated noise intensity of the corrected image, parameters of the wavelet transform may be determined to noise filter the corrected image.
The parameters of the wavelet transformation include: wavelet basis, number of decomposition layers, and threshold function. Typically, the images acquired by the unmanned aerial vehicle are corrected to include smooth regions and also abrupt regions, and different regions should be selected from different wavelet bases. In the smoothing region, a wavelet function having a higher order vanishing matrix is generally selected because it can detect finer singularities contained in the image signal, thereby recovering image detail problems as much as possible. While a tightly supported wavelet basis is typically selected in the abrupt region to make the wavelet transform process of the image simple and efficient. And the number of decomposition layers is three to five based on actual conditions, so that the error of the reconstructed image is minimum, and the accuracy of the image is ensured.
And decomposing the corrected image based on the determined wavelet transformation parameters to obtain a high-frequency component and a low-frequency component of the corrected image. And projecting the transformed high frequency component into a low frequency space through a Gaussian observation matrix to obtain a measured value only containing an N-dimensional noise vector, wherein the smaller the value of N is, the more noise is removed. And reconstructing the low-frequency component after wavelet transformation decomposition and the measured value after Gaussian observation matrix processing based on wavelet inverse transformation to obtain a denoising image after noise filtering. Through the filtration of curved noise image, the problem that image analysis reliability is low that noise interference in unmanned aerial vehicle collection image leads to has been solved.
And taking the filtered denoising image as an image to be analyzed, wherein the image to be analyzed meets the requirements of the region to be inspected, and inputting a preset potential safety hazard model of the oil well for analysis, so that the potential safety hazard type and the position coordinates of the oil well in the region to be inspected are obtained.
The following description is needed: before the images to be analyzed meeting the requirements of all inspection subareas are input into a preset oil well hidden danger identification model, building and training are needed to be carried out on the oil well hidden danger identification model, and the training model meeting the requirements is used as the oil well hidden danger identification model. Specifically, in one or more embodiments of the present disclosure, a historical inspection image of an area to be inspected is obtained based on internet-based crawler technology or based on data stored in a database, where information included in the historical inspection image includes: coordinates of the oil well in the history inspection, fault points of the oil well in the history inspection, positions of potential safety hazards detected in the history inspection, related potential hazard data, the activity range of non-staff obtained in the history inspection, and the like. The obtained historical inspection image of the area to be inspected is used as a sample library of the area to be inspected, and the historical image in the sample library is segmented according to the range of the inspection subarea to be used as an image sample to be trained, so that the training model can ensure the targeted training analysis of the image in the area to be inspected. And performing dimension reduction processing on the feature vector by extracting the feature vector of the decomposed image sample, inputting the feature vector subjected to dimension reduction decomposition as training data into a training model, and outputting the type of potential safety hazards and the positions of the potential safety hazards of the oil well to train the neural network model. And evaluating the training result, and taking the model as an oil well hidden danger identification network model when the identified error is smaller than a set threshold value.
As shown in fig. 2, one or more embodiments of the present disclosure provide a multi-unmanned aerial vehicle-based oil well inspection apparatus, the apparatus comprising:
at least one processor 201; the method comprises the steps of,
a memory 202 communicatively coupled to the at least one processor 201; wherein,
the memory 202 stores instructions executable by the at least one processor 201 to enable the at least one processor 201 to:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
Filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
As shown in fig. 3, in one or more embodiments of the present specification, there is provided a nonvolatile storage medium storing executable instructions 301 of a computer, the executable instructions 301 including:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
Filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present description, is intended to be included within the scope of the claims of the present description.
Claims (9)
1. An oil well inspection method based on multiple unmanned aerial vehicles, which is characterized by comprising the following steps:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well;
before the images to be analyzed meeting the requirements of the inspection subareas are input into a preset oil well hidden danger identification network model, the method further comprises the following steps:
Acquiring a historical inspection image of an area to be inspected based on a crawler technology of the Internet or based on data stored in a database; wherein, the history inspection image includes: coordinates of the inspection oil well, fault points of the inspection oil well, potential safety hazard position labels and non-staff areas;
taking the historical inspection image of the area to be inspected as a sample library of the area to be inspected, and dividing the historical inspection image according to the range of the inspection subarea to be used as an image sample;
extracting and decomposing a feature vector of the image sample, and performing dimension reduction decomposition on the feature vector to serve as a training sample;
training the neural network model by taking the training sample as input and the potential safety hazard type and the position coordinates of the oil well as output to obtain a recognition result;
and evaluating the identification result, and taking the model as an oil well hidden danger identification network model if the identification result is smaller than a preset threshold value.
2. The multi-unmanned aerial vehicle-based oil well inspection method according to claim 1, wherein the dividing the area to be inspected of the oil well inspection task into a plurality of inspection sub-areas specifically comprises:
Acquiring historical inspection actions of all oil well equipment in the area to be inspected;
performing cluster analysis on the historical inspection actions based on a cluster algorithm to obtain a cluster center of the historical inspection actions within a preset distance threshold; wherein, the history inspection action includes: inspection of suspicious personnel, oil leakage inspection of oil well equipment, inspection of dangerous objects and inspection of combustible objects;
and dividing the region to be inspected according to the clustering range of the clustering center to obtain a plurality of inspection sub-regions.
3. The multi-unmanned aerial vehicle-based oil well inspection method according to claim 1, wherein the determining the optimal flight path of each unmanned aerial vehicle in the plurality of inspection areas specifically comprises:
determining initial oil well coordinates and target oil well coordinates of the unmanned aerial vehicle according to the coordinates of the oil wells in the inspection area and the importance level of the oil wells;
determining the flying height of the unmanned aerial vehicle based on the inspection task of the unmanned aerial vehicle;
intercepting a two-dimensional plan of the patrol sub-area according to the flying height;
determining a flight area of the unmanned aerial vehicle based on the two-dimensional plan;
determining an unmanned aerial vehicle passing path in the patrol sub-area according to the flight area of the unmanned aerial vehicle, the starting point oil well coordinates and the target oil well coordinates of the unmanned aerial vehicle;
And analyzing the unmanned aerial vehicle passing path based on a variable neighborhood search algorithm to obtain the optimal flight path of the patrol sub-area.
4. The multi-drone based oil well inspection method of claim 1, wherein after determining the optimal flight path for each drone in the plurality of inspection sub-areas, the method further comprises:
selecting a plurality of shooting positions in the optimal flight path of each patrol sub-area;
and setting shooting angles and shooting parameters at each shooting position of the unmanned aerial vehicle according to the shooting positions and the range of the patrol sub-area, so that the shooting images cover preset fault patrol points of the oil well in the patrol sub-area.
5. The multi-unmanned aerial vehicle-based oil well inspection method according to claim 1, wherein before inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, the method further comprises:
performing gray level correction and self-adaptive histogram equalization processing on the images to be analyzed of each inspection subarea to obtain corrected images of the images to be analyzed;
processing the corrected image based on a preset filtering mode, removing noise data of the corrected image, and obtaining a denoising image;
And taking the denoising image as an image to be analyzed meeting the requirements of each inspection subarea.
6. The multi-unmanned aerial vehicle-based oil well inspection method according to claim 5, wherein the processing the corrected image based on a preset filtering mode, removing noise data of the corrected image, and obtaining a denoising image specifically comprises:
determining parameters of wavelet transformation according to the estimated noise intensity of the corrected image; wherein the parameters include: wavelet basis function, decomposition layer number and threshold function;
decomposing the correction image according to the parameters of the wavelet transformation to obtain a high-frequency component and a low-frequency component of the correction image;
projecting the high-frequency component into a low-frequency space by using a Gaussian random matrix to obtain a measured value after filtering a noise vector;
and carrying out wavelet inverse transformation on the low-frequency component and the measured value so as to reconstruct the denoising image.
7. The multi-unmanned aerial vehicle-based oil well inspection method according to claim 1, wherein after the clear image of each inspection subarea is obtained as the image to be analyzed, the method further comprises:
determining an optimal splicing mode among images in the patrol sub-areas based on a minimum-cut maximum flow principle and an overlapping area backed up by each patrol sub-area;
Splicing the images to be analyzed of the patrol sub-area according to the optimal splicing mode to obtain an initial spliced image of the patrol sub-area;
and fusing the spliced images through a multi-resolution technology to obtain a panoramic image of the area to be inspected so as to realize task statistics of unmanned aerial vehicle inspection.
8. An oil well inspection device based on multiple unmanned aerial vehicles, the device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
Determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well;
before inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, the method further comprises the following steps:
acquiring a historical inspection image of an area to be inspected based on a crawler technology of the Internet or based on data stored in a database; wherein, the history inspection image includes: coordinates of the inspection oil well, fault points of the inspection oil well, potential safety hazard position labels and non-staff areas;
taking the historical inspection image of the area to be inspected as a sample library of the area to be inspected, and dividing the historical inspection image according to the range of the inspection subarea to be used as an image sample;
extracting and decomposing a feature vector of the image sample, and performing dimension reduction decomposition on the feature vector to serve as a training sample;
Training the neural network model by taking the training sample as input and the potential safety hazard type and the position coordinates of the oil well as output to obtain a recognition result;
and evaluating the identification result, and taking the model as an oil well hidden danger identification network model if the identification result is smaller than a preset threshold value.
9. A non-volatile storage medium storing executable instructions for a computer, the executable instructions comprising:
acquiring an oil well inspection task; the oil well inspection task comprises geographic coordinate positions of all oil wells in an area to be inspected;
dividing an area to be inspected of the oil well inspection task into a plurality of inspection subareas, and distributing a plurality of unmanned aerial vehicles with corresponding functions for each inspection subarea;
determining the optimal flight path of each unmanned aerial vehicle in the plurality of patrol sub-areas;
controlling the unmanned aerial vehicle to acquire images in a patrol sub-area according to the optimal flight path, and acquiring an initial shooting image of the oil well;
determining overlapping areas of a plurality of unmanned aerial vehicle initial shooting images in each inspection subarea according to the position coordinates of the initial shooting images;
filtering the overlapped area of each inspection subarea to obtain a clear image of each inspection subarea as an image to be analyzed;
Inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, and outputting the type and position coordinates of the potential safety hazards of the oil well;
before inputting the image to be analyzed meeting the requirements of each inspection subarea into a preset oil well hidden danger identification network model, the method further comprises the following steps:
acquiring a historical inspection image of an area to be inspected based on a crawler technology of the Internet or based on data stored in a database; wherein, the history inspection image includes: coordinates of the inspection oil well, fault points of the inspection oil well, potential safety hazard position labels and non-staff areas;
taking the historical inspection image of the area to be inspected as a sample library of the area to be inspected, and dividing the historical inspection image according to the range of the inspection subarea to be used as an image sample;
extracting and decomposing a feature vector of the image sample, and performing dimension reduction decomposition on the feature vector to serve as a training sample;
training the neural network model by taking the training sample as input and the potential safety hazard type and the position coordinates of the oil well as output to obtain a recognition result;
and evaluating the identification result, and taking the model as an oil well hidden danger identification network model if the identification result is smaller than a preset threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111067758.3A CN113741527B (en) | 2021-09-13 | 2021-09-13 | Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111067758.3A CN113741527B (en) | 2021-09-13 | 2021-09-13 | Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113741527A CN113741527A (en) | 2021-12-03 |
| CN113741527B true CN113741527B (en) | 2024-01-19 |
Family
ID=78738263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111067758.3A Active CN113741527B (en) | 2021-09-13 | 2021-09-13 | Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113741527B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113989682B (en) * | 2021-12-29 | 2022-05-17 | 天津天元海科技开发有限公司 | Navigation mark inspection system and inspection method based on unmanned aerial vehicle remote sensing |
| CN114578848B (en) * | 2022-01-14 | 2024-08-16 | 华东师范大学 | Unmanned aerial vehicle routing inspection path planning method based on discrete point density and global planning |
| CN115063699B (en) * | 2022-05-30 | 2023-11-28 | 广州中科云图智能科技有限公司 | Unmanned aerial vehicle-based substation inspection method, unmanned aerial vehicle-based substation inspection device, unmanned aerial vehicle-based substation inspection equipment and storage medium |
| CN114814877B (en) * | 2022-06-21 | 2022-09-06 | 山东金宇信息科技集团有限公司 | Tunnel data acquisition method, equipment and medium based on inspection robot |
| CN115048702A (en) * | 2022-06-24 | 2022-09-13 | 珠海市交通勘察设计院有限公司 | Road survey design method, device, equipment and storage medium based on 5G network |
| CN115355952B (en) * | 2022-10-20 | 2023-01-20 | 山东联合能源管道输送有限公司 | Intelligent inspection system for crude oil storage tank |
| CN115755954B (en) * | 2022-10-28 | 2023-07-25 | 佳源科技股份有限公司 | Routing inspection path planning method, system, computer equipment and storage medium |
| CN116415910A (en) * | 2023-03-27 | 2023-07-11 | 国网山东省电力公司建设公司 | Unmanned aerial vehicle-based power transmission line environment-friendly intelligent checking method |
| CN116055690B (en) * | 2023-04-03 | 2023-06-09 | 山东金宇信息科技集团有限公司 | Method and equipment for processing machine room monitoring video |
| CN116543322B (en) * | 2023-05-17 | 2024-08-30 | 深圳市保臻社区服务科技有限公司 | Intelligent property routing inspection method based on community potential safety hazards |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110231833A (en) * | 2019-06-14 | 2019-09-13 | 渤海大学 | A kind of oil field inspection fixed-point data acquisition system and method based on multiple no-manned plane |
| CN110398982A (en) * | 2019-06-21 | 2019-11-01 | 万翼科技有限公司 | A kind of method for inspecting and system of unmanned plane |
| CN110414359A (en) * | 2019-07-01 | 2019-11-05 | 中国石化销售有限公司华南分公司 | The analysis of long distance pipeline unmanned plane inspection data and management method and system |
| CN110989678A (en) * | 2019-12-23 | 2020-04-10 | 航天科技控股集团股份有限公司 | Oil field inspection fixed point data acquisition system and method based on multiple unmanned aerial vehicles |
| CN112580923A (en) * | 2020-11-13 | 2021-03-30 | 北京大学（天津滨海）新一代信息技术研究院 | Scheduling method, system and computer medium for oil field inspection unmanned aerial vehicle |
| CN112749589A (en) * | 2019-10-30 | 2021-05-04 | 中移(苏州)软件技术有限公司 | Method and device for determining routing inspection path and storage medium |
| WO2021102796A1 (en) * | 2019-11-28 | 2021-06-03 | 深圳市大疆创新科技有限公司 | Position determination method for unmanned aerial vehicle base station, terminal, and computer readable storage medium |
| CN113323618A (en) * | 2021-06-02 | 2021-08-31 | 德仕能源科技集团股份有限公司 | Ultra-low permeability reservoir artificial fracture water channeling treatment method and device |
-
2021
- 2021-09-13 CN CN202111067758.3A patent/CN113741527B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110231833A (en) * | 2019-06-14 | 2019-09-13 | 渤海大学 | A kind of oil field inspection fixed-point data acquisition system and method based on multiple no-manned plane |
| CN110398982A (en) * | 2019-06-21 | 2019-11-01 | 万翼科技有限公司 | A kind of method for inspecting and system of unmanned plane |
| CN110414359A (en) * | 2019-07-01 | 2019-11-05 | 中国石化销售有限公司华南分公司 | The analysis of long distance pipeline unmanned plane inspection data and management method and system |
| CN112749589A (en) * | 2019-10-30 | 2021-05-04 | 中移(苏州)软件技术有限公司 | Method and device for determining routing inspection path and storage medium |
| WO2021102796A1 (en) * | 2019-11-28 | 2021-06-03 | 深圳市大疆创新科技有限公司 | Position determination method for unmanned aerial vehicle base station, terminal, and computer readable storage medium |
| CN110989678A (en) * | 2019-12-23 | 2020-04-10 | 航天科技控股集团股份有限公司 | Oil field inspection fixed point data acquisition system and method based on multiple unmanned aerial vehicles |
| CN112580923A (en) * | 2020-11-13 | 2021-03-30 | 北京大学（天津滨海）新一代信息技术研究院 | Scheduling method, system and computer medium for oil field inspection unmanned aerial vehicle |
| CN113323618A (en) * | 2021-06-02 | 2021-08-31 | 德仕能源科技集团股份有限公司 | Ultra-low permeability reservoir artificial fracture water channeling treatment method and device |
Non-Patent Citations (2)
| Title |
|---|
| 基于多旋翼无人机的油田管线巡检;邢致恺;贾鹤鸣;邢国军;朱柏卓;张森;朱传旭;;科技创新与生产力(第07期);全文 * |
| 无人机巡检系统在油田生产领域的应用研究;袁鹏;孙强;薛瑾;马赟;李建平;;油气田地面工程(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113741527A (en) | 2021-12-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113741527B (en) | Oil well inspection method, equipment and medium based on multiple unmanned aerial vehicles | |
| AU2015404580B2 (en) | Condition detection using image processing | |
| CA2994508C (en) | Vegetation management for power line corridor monitoring using computer vision | |
| Sollazzo et al. | Hybrid procedure for automated detection of cracking with 3D pavement data | |
| WO2023287276A1 (en) | Geographic data processing methods and systems for detecting encroachment by objects into a geographic corridor | |
| Ye et al. | Diagnosis of sewer pipe defects on image recognition of multi-features and support vector machine in a southern Chinese city | |
| WO2019190405A1 (en) | Method and apparatus for detecting condition of a bolt on a bolted structure | |
| Chen et al. | Registering georeferenced photos to a building information model to extract structures of interest | |
| Guo et al. | Surface defect detection of civil structures using images: Review from data perspective | |
| Li et al. | RoadID: a dedicated deep convolutional neural network for multipavement distress detection | |
| Shuang et al. | Ple: Power line extraction algorithm for uav-based power inspection | |
| CN111798435A (en) | Image processing method, and method and system for monitoring invasion of engineering vehicle into power transmission line | |
| Park et al. | Deep learning-based UAV image segmentation and inpainting for generating vehicle-free orthomosaic | |
| Xiu et al. | DS-Net: A dedicated approach for collapsed building detection from post-event airborne point clouds | |
| Aendikov et al. | Integration of GIS and machine learning analytics into Streamlit application | |
| Parape et al. | Optimization of structure elements for morphological hit-or-miss transform for building extraction from VHR airborne imagery in natural hazard areas | |
| Colace et al. | FAUNO: A Machine Learning-Based Methodology for Monitoring and Predictive Maintenance of Structures in Archaeological Parks Through Image Analysis | |
| Oring | Fast and robust structural damage analysis of civil infrastructure using uav imagery | |
| Soares et al. | Segmentation and Removal of Markings in Metal Inspection Images | |
| Qin et al. | Application Research of Improved Canny Operator in Mine Foreign Body Edge Detection | |
| Pravalika et al. | Bridge Detection using Satellite Images | |
| Kumar | Leveraging Big Data and Deep Learning for Economical Condition Assessment of Wastewater Pipelines | |
| Chen et al. | Shifting research from defect detection to defect modeling in computer vision-based structural health monitoring | |
| Park et al. | Observations and Geoinformation | |
| Riegler et al. | Filament and Flare Detection in H {\alpha} image sequences |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230831 Address after: 257000 Shengli Industrial Zone, Dongying District, Shandong, Dongying Applicant after: Deshi Energy Technology Group Co.,Ltd. Address before: 257000 building 18, college student Pioneer Park, Zoucheng Road, Dongying District, Dongying City, Shandong Province Applicant before: Deshi Energy Technology Group Co.,Ltd. Applicant before: SHANDONG DESHI PETROLEUM EQUIPMENT Co.,Ltd. Applicant before: SHANDONG DESHI CHEMICAL Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |