CN109325520B - Method, device and system for checking petroleum leakage - Google Patents
Method, device and system for checking petroleum leakage Download PDFInfo
- Publication number
- CN109325520B CN109325520B CN201810970710.5A CN201810970710A CN109325520B CN 109325520 B CN109325520 B CN 109325520B CN 201810970710 A CN201810970710 A CN 201810970710A CN 109325520 B CN109325520 B CN 109325520B
- Authority
- CN
- China
- Prior art keywords
- target image
- image
- oil leakage
- determining
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Abstract
The application provides a method, a device and a system for checking oil leakage, wherein the method comprises the following steps: inputting a first target image acquired by an unmanned aerial vehicle in real time into a pre-trained oil leakage detection model, and determining a first probability of oil leakage; if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image; inputting the second target image into the oil leakage detection model, and determining a second probability of oil leakage; determining that an oil leak has occurred after determining that the second probability exceeds the preset threshold. This application embodiment patrols and examines and train oil leakage detection model through unmanned aerial vehicle is automatic to carry out accurate aassessment to oil leakage probability, and the automatic oil leakage condition that detects can effectively reduce intensity of labour and the safety risk of patrolling and examining personnel in traditional artifical patrolling and examining the mode.
Description
Technical Field
The application relates to the technical field of automation control, in particular to a method, a device and a system for checking petroleum leakage.
Background
With the vigorous development of the petroleum industry, the total mileage of oil and gas pipelines in China is nearly 15 kilometers by 2016, the number of oil storage tanks for petroleum storage and transportation exceeds 90 kilometers, and the petroleum storage and transportation pressure vessel facilities provide important energy guarantee for national economy and daily life of residents. Because petroleum has high corrosivity, although pressure vessels used for petroleum storage and transportation at present adopt a plurality of technologies for relevant protection, petroleum leakage risks and potential safety hazards are inevitably brought about due to untimely maintenance or external force damage (oil stealing, construction damage and the like).
At present, potential safety hazards of petroleum storage and transportation pressure vessel facilities are mainly eliminated in an inspection mode. The inspection mode is generally manual inspection, and the safety condition of the petroleum storage and transportation pressure vessel is judged manually, so that the efficiency is low.
Disclosure of Invention
In view of this, an object of the present application is to provide a method, an apparatus, and a system for checking oil leakage, so as to accurately evaluate the oil leakage probability and automatically detect the oil leakage condition, thereby effectively reducing the labor intensity and the safety risk of the inspection personnel.
In a first aspect, an embodiment of the present application provides a method for checking an oil leak, including:
inputting a first target image acquired by an unmanned aerial vehicle in real time into a pre-trained oil leakage detection model, and determining a first probability of oil leakage;
if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image;
inputting the second target image into the oil leakage detection model, and determining a second probability of oil leakage;
determining that an oil leak has occurred after determining that the second probability exceeds the preset threshold.
With reference to the first aspect, an embodiment of the present application provides a first possible implementation manner of the first aspect, where, if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image, including:
dividing the first target image into a first foreground image part and a first background image part according to the pixel gray value of the first target image;
determining a segmentation threshold for segmenting the foreground and the background according to a maximum class variance between the first foreground image portion and the first background image portion;
and performing image segmentation processing on the first target image based on the determined segmentation threshold value to obtain the second target image.
With reference to the first possible implementation manner of the first aspect, the present examples provide a second possible implementation manner of the first aspect, wherein the oil leakage detection model is obtained by training according to the following steps:
adopting a deep convolutional neural network model as a basic training model;
and taking the petroleum leakage image sample and the petroleum non-leakage image sample as training sets of the basic training model, taking the known petroleum leakage result as an output result of the basic training model, and training to obtain the petroleum leakage detection model.
With reference to the first aspect, the present application provides a third possible implementation manner of the first aspect, where after determining that an oil leakage occurs, the method further includes:
and determining a real geographical area of the petroleum leakage according to the position information of the target area corresponding to the first target image.
With reference to the first aspect, the present application provides a fourth possible implementation manner of the first aspect, where after determining that an oil leakage occurs, the method further includes:
acquiring shooting parameter information of the first target image shot by the unmanned aerial vehicle; the shooting parameter information comprises the flight height of the unmanned aerial vehicle, the number of pixels of the camera, the horizontal field angle of the camera and the vertical field angle of the camera;
and calculating the petroleum leakage area according to the shooting parameter information and the pixel information of the second target image.
In a second aspect, an embodiment of the present application further provides an oil leakage inspection apparatus, including: the system comprises a first data acquisition module, a first data analysis module, a second data analysis module, a data comparison module and a first data processing module;
the data acquisition module is used for acquiring a first target image in real time;
the first data analysis module is used for inputting the first target image into a pre-trained oil leakage detection model so as to determine a first probability of oil leakage;
the data comparison module is used for comparing the first probability with a preset threshold value;
the first data processing module is configured to perform anti-interference processing on the first target image to obtain a second target image if the first probability exceeds a preset threshold;
and the second data analysis module is used for inputting the second target image into the oil leakage detection model and determining a second probability of oil leakage.
With reference to the second aspect, embodiments of the present application provide a first possible implementation manner of the second aspect, where the data processing module includes an image segmentation unit, an image determination unit, and an image generation unit;
the image segmentation unit is used for dividing the first target image into a first foreground image part and a first background image part according to the pixel gray value of the first target image;
the image determining unit is used for determining a segmentation threshold value for segmenting the foreground and the background according to the maximum class variance between the first foreground image part and the first background image part;
the image generation unit is configured to perform image segmentation processing on the first target image based on the determined segmentation threshold to obtain the second target image.
With reference to the second aspect, embodiments of the present application provide a second possible implementation manner of the second aspect, where the method further includes: an image determination module;
and the image determining module is used for determining the real geographical area of the petroleum leakage according to the position information of the target area corresponding to the first target image.
With reference to the second aspect, an embodiment of the present application provides a third possible implementation manner of the second aspect, where the method further includes: the second data acquisition module and the second data processing module;
the second data acquisition module is used for acquiring shooting parameter information of the first target image shot by the unmanned aerial vehicle;
and the second data processing module is used for calculating the oil leakage area according to the shooting parameter information and the pixel information of the second target image.
In a third aspect, an embodiment of the present application further provides an oil leakage inspection system, including: the system comprises a data processing server, an unmanned aerial vehicle body, a ground communication base station and an enterprise intelligent control center;
the data processing server is used for executing the method according to any one of claims 1 to 5 and transmitting the execution result to the enterprise intelligent control center;
the enterprise intelligent control center is used for sending an unmanned aerial vehicle body operation instruction to the ground communication base station based on the execution result transmitted by the data processing server;
and the ground communication base station is used for responding to the unmanned aerial vehicle body operation instruction sent by the enterprise intelligent control center.
According to the method, the device and the system for checking the petroleum leakage, the first target image acquired by the unmanned aerial vehicle in real time is input into a pre-trained petroleum leakage detection model, and the first probability of petroleum leakage is determined; if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image; inputting the second target image into the oil leakage detection model, and determining a second probability of oil leakage; determining that an oil leak has occurred after determining that the second probability exceeds the preset threshold. Whether oil leakage occurs is determined through a trained oil leakage detection model, the recognition accuracy is higher than that of manual judgment, and the operator does not rely on visual inspection. Simultaneously, for the method of artifical inspection or manual control unmanned aerial vehicle patrolling and examining, efficiency is higher, and the cost is littleer, and degree of automation is higher, receives manual operation influence less. In addition, a two-stage detection mode is adopted, after the first probability of petroleum leakage is determined to exceed the preset threshold value based on the collected first target image, whether the second probability of petroleum leakage exceeds the preset threshold value is determined based on the second target image subjected to anti-interference processing on the first target image, and the petroleum leakage is determined only after the second probability also exceeds the second threshold value, so that the probability of misjudgment is reduced.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
FIG. 1 is a flow chart illustrating a method for checking for oil leakage according to an embodiment of the present disclosure;
FIG. 2 is a flow chart illustrating the optimization of a method for checking for oil leakage according to an embodiment of the present disclosure;
fig. 3 is a diagram illustrating an effect of the anti-interference processing provided by the embodiment of the present application;
FIG. 4 is a flow chart illustrating the optimization of another method for checking for oil leakage according to the embodiment of the present application;
FIG. 5 is a flow chart illustrating the optimization of another method for checking for oil leakage according to the embodiment of the present application;
FIG. 6 is a diagram illustrating the effect of the maximum between-class variance algorithm provided by the embodiment of the present application;
FIG. 7 is a schematic diagram illustrating the calculation of an oil leakage area using a maximum extremum stable region algorithm provided by an embodiment of the present application;
FIG. 8 is a schematic structural diagram illustrating an apparatus for checking for oil leakage according to an embodiment of the present disclosure;
fig. 9 shows a schematic diagram of a front end of an enterprise smart production center monitor provided by an embodiment of the present application.
Fig. 10 shows an effect diagram after the inspection method using oil leakage provided by the embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
With the vigorous development of the petroleum industry, the total mileage of oil and gas pipelines in China is nearly 15 kilometers by 2016, the number of oil storage tanks for petroleum storage and transportation exceeds 90 kilometers, and the petroleum storage and transportation pressure vessel facilities provide important energy guarantee for national economy and daily life of residents. Because the petroleum has high corrosivity, although the pressure vessel for petroleum storage and transportation at present adopts a plurality of technologies to carry out related protection, the petroleum leakage risk and the potential safety hazard are inevitably brought because of untimely maintenance or external force damage (oil stealing, construction damage and the like). Once petroleum is leaked, irreversible damage can be caused to the environment, and simultaneously, the petroleum is easy to burn and explode under the conditions of open fire and the like, so that the personal and property safety of people is threatened.
In order to avoid leakage of an oil storage and transportation system, oil enterprises currently adopt a plurality of technical means to monitor the integrity of the pressure container, but due to the influence of physical characteristics such as viscosity and complex components of oil, the most effective method is to inspect the pressure container at present.
The traditional inspection work of the petroleum storage and transportation pressure vessel mainly adopts two modes of regular inspection by a security inspector and irregular spot inspection by a higher supervising department, and both adopt a manual contact mode, so that the labor intensity is high, the operation procedure is complex, the resource allocation is too fat, the work efficiency is low, the period is longer, and particularly, the measurement task cannot be completed on time with quality and quantity protection in difficult places. In recent years, unmanned aerial vehicles have begun to be applied in scenes where oil leaks are being photographed. However, the current inspection work still highly depends on manual flight control and judgment of the petroleum leakage, and the enterprise operation cost can be increased by carrying out related training on professionals. And even personnel trained by professionals are susceptible to great influence on the detection effect of the 'leakage and leakage' condition of petroleum due to the influence of the resolution of aerial pictures and the external environment (such as illumination, shooting angle and the like) during shooting, so that the method has great limitation in practical use.
Based on the above problems, embodiments of the present application provide a method, an apparatus, and a system for checking oil leakage, which are described below by way of example.
For the understanding of the present embodiment, a method for checking for oil leakage disclosed in the embodiments of the present application will be described in detail first.
As shown in fig. 1, a method for checking oil leakage includes the following steps:
s101, inputting a first target image acquired in real time into a pre-trained oil leakage detection model, and determining a first probability of oil leakage;
s102, if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image;
s103, inputting the second target image into the oil leakage detection model, and determining a second probability of oil leakage;
and S104, determining that the petroleum leakage occurs after the second probability is determined to exceed the preset threshold value.
In step S101, a first target image acquired in real time is input into a pre-trained oil leakage detection model. The first target image is an image including an oil leak scene, captured in real time by the drone in the field. The pre-trained oil leakage detection model is characterized in that an unmanned aerial vehicle of the same model is used for shooting on the spot to obtain an oil leakage image sample, the obtained oil leakage image sample is preprocessed and then is used as a training set of the model together with an oil non-leakage image sample, then a deep convolution neural network is trained on the basis of a basic training model to obtain an oil leakage detection model, and the oil leakage detection model is deployed on a data processing server. The base training model may be the Google Net inclusion V3 pre-training model. And finally, determining the probability of the oil leakage of the region in the input picture through an oil leakage model, and determining the probability as a first probability.
In step S102, when the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image. The preset threshold value is an artificially set value, and when the probability obtained after the oil leakage detection model is greater than the preset threshold value, the oil leakage is determined to occur through secondary judgment.
Step S103 is to input the second target image subjected to the interference rejection processing to the oil leakage detection model again to obtain a second probability on the basis of step S102. This is a secondary judgment of whether or not oil leakage occurs in the region displayed by the first target image input to the oil leakage detection model, so as to reduce the possibility of occurrence of erroneous judgment.
And step S104, comparing the second probability obtained in the previous step with a preset threshold value, determining whether the second probability exceeds the preset threshold value, and judging that the petroleum leakage occurs when the second probability still exceeds the preset threshold value.
Further, as shown in fig. 2, step S102 includes the following steps:
s201, dividing a first target image into a first foreground image part and a first background image part according to the pixel gray value of the first target image;
s202, determining a segmentation threshold for segmenting the foreground and the background according to the maximum class variance between the first foreground image part and the first background image part;
s203, based on the determined segmentation threshold, performing image segmentation processing on the first target image to obtain the second target image.
The steps from S201 to S203 are to divide the first target image into a first foreground image portion and a first background image portion according to the pixel gray value of the first target image, and when an optimal threshold is adopted, the difference between the two portions should be the largest, and the criterion for measuring the difference here is the most common maximum inter-class variance. If the inter-class variance between the first foreground image part and the first background image part is larger, the difference between the two parts forming the image is larger, when a part of objects are wrongly classified into the background or a part of the background is wrongly classified into the objects, the difference between the two parts is reduced, and when the threshold value is taken to divide the inter-class variance to be maximum, the probability of wrong classification is minimum. When the variance is the largest, the difference between the foreground and the background at this time can be considered as the largest, and the gray value at this time is the optimal threshold. At this time, according to the determined segmentation threshold, the first target image is segmented, and the first foreground image part and the first background image part are segmented and displayed. Meanwhile, as shown in fig. 3, the foreground image portion, i.e., the oil leakage area, is circled on the picture, and the background, which may be a disturbing object, is replaced with white to be in sharp contrast with the black color of the oil leakage area. Or the background of the interference object is shielded to highlight the foreground image part, namely the petroleum leakage area, so as to obtain the second target image subjected to anti-interference processing.
As shown in fig. 4, the step of training the oil leak detection model includes:
s401, adopting a deep convolution neural network model as a basic training model;
s402, taking the petroleum leakage image sample and the petroleum non-leakage image sample as training sets of the basic training model, taking the known petroleum leakage result as an output result of the basic training model, and training to obtain the petroleum leakage detection model.
In step S401, the Google Net inclusion V3 is used as a basic training model to participate in training. Step S402 specifically includes the following processing procedure. Firstly, acquiring real-time videos of petroleum leakage of a petroleum storage and transportation pressure container shot by an unmanned aerial vehicle from multiple angles, and intercepting the real-time videos into frames according to a certain time interval to obtain petroleum leakage image samples; secondly, preprocessing the oil leakage image sample. The pretreatment process comprises the following steps: adjusting the contrast of the original picture to simulate the petroleum leakage scene under different illumination conditions, generating a new image by rotating the original picture anticlockwise for every 45 degrees to simulate the petroleum leakage condition under different shooting angles, and amplifying and reducing the original picture according to a certain proportion to simulate the petroleum leakage condition shot under different heights; finally, the parameters of the Google Net addition V3 basic model are adjusted by using the preprocessed pictures, a new deep convolutional neural network model for detecting the oil leakage is trained, and the network is deployed on a data processing server. Specifically, a deep convolution neural network is adopted to extract features and design a detector to complete detection of an oil leakage area; and taking the petroleum leakage image sample marked out of the petroleum leakage area as a training set of the basic training model, and supervising and guiding the training of the model by using the marked petroleum leakage result so that the output result of the model is continuously converged to the marked result, thereby finishing the training of the model.
After the oil leakage is determined, the oil leakage area can be determined according to the position information of the target area corresponding to the first target image. The position information comprises longitude and latitude of the shot area, and the position of the petroleum leakage area can be accurately positioned.
As shown in fig. 5, after the oil leakage is determined according to the above steps, the method further comprises the following steps:
s501, acquiring shooting parameter information of the first target image shot by the unmanned aerial vehicle; the shooting parameter information comprises the flight height of the unmanned aerial vehicle, the number of pixels of the camera, the horizontal field angle of the camera and the vertical field angle of the camera;
and S502, calculating the oil leakage area according to the shooting parameter information and the pixel information of the second target image.
An unmanned aerial vehicle inspection line is preset according to GPS coordinates of the petroleum storage and transportation pressure vessel, and the unmanned aerial vehicle flies according to specific flight parameters (speed, height and the like). Planning an unmanned aerial vehicle inspection line according to GPS coordinates of the petroleum storage and transportation pressure container, and writing the line into an unmanned aerial vehicle flight control system.
In step S501, shooting parameter information of the first target image shot by the unmanned aerial vehicle is acquired. The flying height of the unmanned aerial vehicle can be obtained from flying parameters transmitted back by the unmanned aerial vehicle, the number of the camera pixels is determined according to the type of the camera carried by the unmanned aerial vehicle, and the horizontal field angle and the vertical field angle of the camera are also related to the type of the camera. The pixel information of the second target image in step S502 refers to a second target image obtained by performing anti-interference processing on the first target image, and the data processing server automatically extracts a pixel value P included in a connected portion detected as an oil-containing region in the second target image by using a maximum extremum stable region algorithm. As shown in fig. 6, it is an effect graph after the maximum between-class variance algorithm. Substituting the pixel value P and the shooting parameter information of the first target image obtained in the step S401 into a formula
Wherein P in the formula0The total number of the pixels of the camera is determined by the model of the camera; in the formula H, the flight height of the unmanned aerial vehicle can be obtained from flight parameter data returned by the unmanned aerial vehicle; α and β in the formula are the horizontal and vertical field angles of the lens, respectively, and are related only to the camera model. FIG. 7 is a schematic diagram of the calculation of oil leakage area using the maximum extremum stable region algorithm. Meanwhile, data are packaged and sent to a ground data processing server through an unmanned aerial vehicle airborne data transmission station, the data sampling frequency is consistent with the frame interception frequency, and the one-to-one correspondence between each picture and the flight parameter data is guaranteed.
In summary, in the above method, firstly, a field shooting is performed by using an unmanned aerial vehicle of the same model to obtain an oil leakage image sample, the oil leakage image sample and an oil non-leakage image sample are used as a training set of the basic training model, and a known oil leakage result is used as an output result of the basic training model; then training a deep convolutional neural network according to the artificially labeled data set on the basis of a Google Net initiation V3 basic model, and deploying the model on a data processing server; and finally, inputting an image shot by the unmanned aerial vehicle in actual use into the model for judgment, eliminating the interference object by using a maximum between-class variance algorithm for the picture exceeding a certain threshold value, if the model still judges that the oil leakage condition exists, automatically extracting the oil leakage area by using a maximum extreme value stable area algorithm, calculating the leakage area, evaluating the oil leakage degree, and displaying related information on the enterprise intelligent control center and the mobile terminal.
In addition, as shown in fig. 8, the present application also provides an oil leak inspection apparatus including: a first data acquisition module 801, a first data analysis module 802, a second data analysis module 803, a data comparison module 804 and a first data processing module 805; a first data obtaining module 801, configured to obtain a first target image in real time; a first data analysis module 802 for inputting a first target image to a pre-trained oil leak detection model to determine a first probability of an oil leak occurring; a data comparison module 804, configured to compare the first probability with a preset threshold; a first data processing module 805, configured to perform anti-interference processing on the first target image to obtain a second target image if the first probability exceeds a preset threshold; a second data analysis module 803, configured to input the second target image into the oil leakage detection model, and determine a second probability of oil leakage.
Further, the first data processing module 805 includes an image segmentation unit 8051, an image determination unit 8052, and an image generation unit 8053;
an image segmentation unit 8051, configured to divide the first target image into a first foreground image portion and a first background image portion according to a pixel grayscale value of the first target image;
an image determining unit 8052 for determining a segmentation threshold for segmenting the foreground and the background based on a maximum class variance between the first foreground image portion and the first background image portion;
an image generating unit 8053, configured to perform image segmentation processing on the first target image based on the determined segmentation threshold to obtain a second target image.
Further, the apparatus further comprises: an image determination module 806;
and an image determining module 806, configured to determine a real geographic area of the oil leakage according to the location information of the target area corresponding to the first target image.
In addition, the device still includes: a second data acquisition module 807 and a second data processing module 808;
a second data obtaining module 807, configured to obtain shooting parameter information of the first target image shot by the unmanned aerial vehicle;
and a second data processing module 808, configured to calculate an oil leakage area according to the shooting parameter information and the pixel information of the second target image.
The present application also provides an oil leak inspection system, comprising: the system comprises a data processing server, an unmanned aerial vehicle body, a ground communication base station and an enterprise intelligent control center;
the data processing server is used for executing the method and transmitting the execution result to the enterprise intelligent control center;
the enterprise intelligent control center is used for sending an unmanned aerial vehicle body operation instruction to the ground communication base station based on the execution result transmitted by the data processing server;
and the ground communication base station is used for responding to the unmanned aerial vehicle body operation instruction sent by the enterprise intelligent control center.
An aircraft control system, a communication system, a task load and a power system are arranged in the unmanned aerial vehicle body, wherein the aircraft control system comprises a combined navigation system and a flight control system, the combined navigation system is composed of a GPS module and an Inertial navigation module (IMU), and the flight control system can receive ground control instructions and send the instructions to the power system. The control system comprises an airborne microprocessor, a laser ranging and obstacle avoiding system, a data transmission radio station, a picture transmission radio station, a receiver, a data storage module, a binocular vision sensor, an infrared ranging sensor, a gyroscope, an IMU (inertial measurement unit) and GPS (global positioning system) combined navigation module, an electronic speed regulator and an auxiliary motor thereof, wherein the airborne microprocessor is connected with a communication system, a task load and a power system through a bus. The communication system comprises a receiver for receiving the ground control signal, a transmitter for communicating with the ground communication base station through a wireless network, and a transmitter for communicating with the satellite through a 4G network. The task load comprises a high-definition camera (used in daytime), an infrared camera (used in night time), a tripod head for installing and fixing the camera and an attached stepping motor. The task load system adopts a modular design, and different modules can be switched by field maintainers according to natural environments such as illumination and the like. The power system comprises an unmanned aerial vehicle power supply, a power supply voltage stabilizing system, a motor and a control system thereof. The ground communication base station comprises a base station power supply system, a wireless signal transceiving station, a base station controller and an antenna. The power supply system provides power for the base station controller and the wireless signal receiving and transmitting station, and the receiving and transmitting station comprises a data transmission radio station used for sending control instructions to the unmanned aerial vehicle, a data transmission radio station used for receiving and transmitting flight parameters of the unmanned aerial vehicle and a picture transmission radio station used for receiving and transmitting pictures shot by the unmanned aerial vehicle. The base station controller is used for the allocation, release and management of the wireless channel of the unmanned aerial vehicle, and the antenna is used for receiving and transmitting radio waves.
And a petroleum leakage detection model which is trained in advance according to the petroleum leakage image sample is deployed on the data processing server and used for processing, judging and evaluating whether petroleum leakage occurs or not and evaluating the leakage severity. As shown in fig. 9, the enterprise intelligent control center comprises a ground monitoring display, a real-time remote monitoring platform and an unmanned aerial vehicle remote control system, the real-time monitoring platform can display historical production data of an enterprise, real-time flight state data of the unmanned aerial vehicle and safety conditions of production areas on the ground monitoring display, and the unmanned aerial vehicle remote control system is used for sending flight attitude control instructions to the unmanned aerial vehicle and controlling task loads to complete specified actions. Fig. 10 is a diagram illustrating the effect of the oil leakage area processed by the method and the system.
In addition, the system also comprises a cloud server and the mobile terminal. The cloud server can be connected with the ground data processing server through a network, the result processed by the server is uploaded to the cloud end, the real-time inspection state is displayed in a specific webpage through a visualization technology, the mobile terminal can check the state through accessing the webpage, and the content of the webpage cannot be modified and the unmanned aerial vehicle cannot be operated at the mobile terminal.
Meanwhile, the ground data processing server packs and uploads the processing result and the flight parameter data transmitted back by the unmanned aerial vehicle data transmission radio station to the cloud server, and the enterprise intelligent control center and the mobile terminal can check the unmanned aerial vehicle inspection result displayed in real time in the visual interface in a webpage access mode. According to the visual interface that shows on the intelligent control center monitor of enterprise, operating personnel can long-rangely assign the instruction to unmanned aerial vehicle, and attitude, speed, the height isoparametric of flight when adjusting unmanned aerial vehicle and patrolling and examining automatically also can control unmanned aerial vehicle and circle around a certain point of interest is automatic. Simultaneously, operating personnel also can switch to the manual operation mode with patrolling the line system, and the flight of nimble control unmanned aerial vehicle. If the unmanned aerial vehicle does not detect the oil leakage condition, then continue to patrol the line according to the preset line.
According to the enterprise production statistical data, the probability of oil leakage occurring again after the oil pipeline or the oil storage tank body in a certain area has an oil leakage accident is higher than that in other areas. When the unmanned aerial vehicle is used for inspection, when an area with petroleum leakage in history is inspected, the unmanned aerial vehicle can reduce to 1/n of the original set speed (n is a positive number greater than 1 and can be freely set by enterprises according to historical production data).
The oil stealing event generally occurs at night according to the actual production situation, and the portrait image shot by the infrared camera is added when the deep convolutional neural network is trained. When the unmanned aerial vehicle is used for routing inspection at night, when the routing inspection system detects suspicious personnel in a certain range of the petroleum storage and transportation pressure vessel, the unmanned aerial vehicle can be controlled to fly around related personnel and prompt an intelligent control center of an enterprise to check the suspicious personnel.
In addition, the automatic unmanned aerial vehicle inspection system can be used for inspecting the repairing condition of the pressure vessel after maintenance except for daily line patrol in normal production of enterprises: unmanned aerial vehicle surrounds the oil tank body after the maintenance and encircles the flight, injects the oil into the jar body through the maintenance, if takes place the oil and spills over the condition, unmanned aerial vehicle system of patrolling and examining can send out the oil leak warning automatically.
The data processing server consists of a series of high-performance computers, wherein a model trained on a data set of petroleum leakage pictures (including pictures shot by a common camera and pictures shot by an infrared camera) acquired in a large number of actual production is deployed, and the picture data set is subjected to picture preprocessing to simulate environments with different illumination and angles before the model is trained; the model is obtained by retraining the Google Net initiation V3 pre-training model according to the adjustment parameters of the actual data set, the model combines the maximum between-class variance algorithm (also called Otsu method, OTSU) in addition to the deep convolutional neural network to eliminate the influence of the background interferent similar to petroleum on the deep convolutional neural network model, and the model is added with the maximum extremum Stable region algorithm (MSER) to automatically realize the evaluation of the petroleum leakage region area; the data processing server is connected with the cloud server, processes pictures shot by the unmanned aerial vehicle, packages and uploads the pictures to the cloud server together with received flight parameters of the unmanned aerial vehicle, and the flight position, flight state parameters, production history data, petroleum leakage overview and severity of the unmanned aerial vehicle can be visually displayed at the front end of an enterprise intelligent control center and the mobile terminal.
The method, the device and the system in the application can realize that the unmanned aerial vehicle automatically inspects the oil storage and transportation pressure vessel, and automatically detect whether an oil leakage accident happens by utilizing a computer vision technology, and do not depend on visual inspection of operators. Constructing a model based on a high-performance pre-training network, wherein the recognition accuracy is higher than that of manual judgment; real samples which are subjected to contrast adjustment, rotation and scaling are added into the model during training, so that a real environment can be well simulated, and the robustness to a complex environment is high; eliminating the influence of interferents by using a maximum between-class variance algorithm; automatically calculating the petroleum leakage area by using a maximum extremum stable region algorithm; and can control the unmanned aerial vehicle to operate in batches; uninterrupted inspection for 24 hours can be realized by replacing batteries and the like. Therefore, compared with the original manual inspection or manual control unmanned aerial vehicle inspection method, the method is higher in efficiency, lower in cost, higher in automation degree and less influenced by manual operation.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions when actually implemented, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer-readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
Finally, it should be noted that: the above-mentioned embodiments are only specific embodiments of the present application, and are used for illustrating the technical solutions of the present application, but not limiting the same, and the scope of the present application is not limited thereto, and although the present application is described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: any person skilled in the art can modify or easily conceive the technical solutions described in the foregoing embodiments or equivalent substitutes for some technical features within the technical scope disclosed in the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the exemplary embodiments of the present application, and are intended to be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (8)
1. A method of checking for oil leakage, comprising:
inputting a first target image acquired by an unmanned aerial vehicle in real time into a pre-trained oil leakage detection model, and determining a first probability of oil leakage;
if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image;
inputting the second target image into the oil leakage detection model, and determining a second probability of oil leakage;
determining that oil leakage occurs after determining that the second probability exceeds the preset threshold;
if the first probability exceeds a preset threshold, performing anti-interference processing on the first target image to obtain a second target image, including:
dividing the first target image into a first foreground image part and a first background image part according to the pixel gray value of the first target image;
determining a segmentation threshold for segmenting the foreground and the background according to a maximum class variance between the first foreground image portion and the first background image portion;
and performing image segmentation processing on the first target image based on the determined segmentation threshold, segmenting and displaying a first foreground image part and a first background image part, marking the first foreground image part, and replacing or shielding the background of an interfering object to obtain a second target image.
2. The method of claim 1, wherein the oil leak detection model is trained according to the following steps:
adopting a deep convolutional neural network model as a basic training model;
and taking the petroleum leakage image sample and the petroleum non-leakage image sample as training sets of the basic training model, taking the known petroleum leakage result as an output result of the basic training model, and training to obtain the petroleum leakage detection model.
3. The method of claim 1, after determining that an oil leak has occurred, further comprising:
and determining a real geographical area of the petroleum leakage according to the position information of the target area corresponding to the first target image.
4. The method of claim 1, after determining that an oil leak has occurred, further comprising:
acquiring shooting parameter information of the first target image shot by the unmanned aerial vehicle; the shooting parameter information comprises the flight height of the unmanned aerial vehicle, the number of pixels of the camera, the horizontal field angle of the camera and the vertical field angle of the camera;
and calculating the petroleum leakage area according to the shooting parameter information and the pixel information of the second target image.
5. An oil leak inspection device, comprising: the system comprises a first data acquisition module, a first data analysis module, a second data analysis module, a data comparison module and a first data processing module;
the first data acquisition module is used for acquiring a first target image in real time;
the first data analysis module is used for inputting the first target image into a pre-trained oil leakage detection model so as to determine a first probability of oil leakage;
the data comparison module is used for comparing the first probability with a preset threshold value;
the first data processing module is configured to perform anti-interference processing on the first target image to obtain a second target image if the first probability exceeds a preset threshold;
the second data analysis module is used for inputting the second target image into the oil leakage detection model and determining a second probability of oil leakage;
the first data processing module comprises an image segmentation unit, an image determination unit and an image generation unit;
the image segmentation unit is used for dividing the first target image into a first foreground image part and a first background image part according to the pixel gray value of the first target image;
the image determining unit is used for determining a segmentation threshold value for segmenting the foreground and the background according to the maximum class variance between the first foreground image part and the first background image part;
the image generation unit is configured to perform image segmentation processing on the first target image based on the determined segmentation threshold, segment and display a first foreground image portion and a first background image portion, mark the first foreground image portion, and replace or block a background of an interfering object to obtain the second target image.
6. The apparatus of claim 5, further comprising: an image determination module;
and the image determining module is used for determining the real geographical area of the petroleum leakage according to the position information of the target area corresponding to the first target image.
7. The apparatus of claim 5, further comprising: the second data acquisition module and the second data processing module;
the second data acquisition module is used for acquiring shooting parameter information of the first target image shot by the unmanned aerial vehicle;
and the second data processing module is used for calculating the oil leakage area according to the shooting parameter information and the pixel information of the second target image.
8. An oil leak inspection system, comprising: the system comprises a data processing server, an unmanned aerial vehicle body, a ground communication base station and an enterprise intelligent control center;
the data processing server is used for executing the method according to any one of claims 1 to 4 and transmitting the execution result to the enterprise intelligent control center;
the enterprise intelligent control center is used for sending an unmanned aerial vehicle body operation instruction to the ground communication base station based on the execution result transmitted by the data processing server;
and the ground communication base station is used for responding to the unmanned aerial vehicle body operation instruction sent by the enterprise intelligent control center.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810970710.5A CN109325520B (en) | 2018-08-24 | 2018-08-24 | Method, device and system for checking petroleum leakage |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810970710.5A CN109325520B (en) | 2018-08-24 | 2018-08-24 | Method, device and system for checking petroleum leakage |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109325520A CN109325520A (en) | 2019-02-12 |
| CN109325520B true CN109325520B (en) | 2021-06-29 |
Family
ID=65263795
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810970710.5A Active CN109325520B (en) | 2018-08-24 | 2018-08-24 | Method, device and system for checking petroleum leakage |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109325520B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110363109B (en) * | 2019-06-26 | 2023-03-24 | 成都新舟锐视科技有限公司 | Computer vision-based coverage type oil leakage detection system and method |
| CN110553151B (en) * | 2019-07-17 | 2021-05-07 | 石化盈科信息技术有限责任公司 | Pipeline leakage monitoring method and system |
| CN110705601A (en) * | 2019-09-09 | 2020-01-17 | 安徽继远软件有限公司 | Transformer substation equipment oil leakage image identification method based on single-stage target detection |
| CN111126196A (en) * | 2019-12-10 | 2020-05-08 | 安徽银河物联通信技术有限公司 | Equipment oil leakage detection method |
| CN111414955B (en) * | 2020-03-17 | 2023-08-25 | 昆仑数智科技有限责任公司 | Intelligent detection method and device for petroleum drilling lost circulation overflow and electronic equipment |
| CN111445462A (en) * | 2020-03-30 | 2020-07-24 | 国家计算机网络与信息安全管理中心 | Storage battery leakage detection method based on neural network and thermography |
| CN113255737B (en) * | 2021-04-30 | 2023-08-08 | 超节点创新科技(深圳)有限公司 | Method for sorting baggage in folded package on civil aviation sorting line, electronic equipment and storage medium |
| CN113542595B (en) * | 2021-06-28 | 2023-04-18 | 北京沧沐科技有限公司 | Capturing and monitoring method and system based on day and night images |
| CN113514193B (en) * | 2021-07-13 | 2022-12-02 | 江铃汽车股份有限公司 | Automobile collision fuel leakage detection method based on image recognition |
| CN113670524B (en) * | 2021-07-13 | 2023-12-19 | 江铃汽车股份有限公司 | Detection method and detection system for automobile collision fuel leakage |
| CN114992813B (en) * | 2022-06-17 | 2023-05-12 | 珠海格力电器股份有限公司 | Method and device for detecting substance leakage, air conditioning equipment and storage medium |
| CN115457297B (en) * | 2022-08-23 | 2023-09-26 | 中国航空油料集团有限公司 | Oil leakage detection method and device for aviation oil depot and aviation oil safety operation and maintenance system |
| CN116052061B (en) * | 2023-02-21 | 2024-02-27 | 嘉洋智慧安全科技(北京)股份有限公司 | Event monitoring method, event monitoring device, electronic equipment and storage medium |
| CN116259012B (en) * | 2023-05-16 | 2023-07-28 | 新疆克拉玛依市荣昌有限责任公司 | Monitoring system and method for embedded supercharged diesel tank |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101170641A (en) * | 2007-12-05 | 2008-04-30 | 北京航空航天大学 | A method for image edge detection based on threshold sectioning |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101571915B (en) * | 2009-06-16 | 2011-06-01 | 大连海事大学 | Method for identifying oil spill of SAR image based on characteristic value |
| US9727800B2 (en) * | 2015-09-25 | 2017-08-08 | Qualcomm Incorporated | Optimized object detection |
| CN108073933B (en) * | 2016-11-08 | 2021-05-25 | 杭州海康威视数字技术股份有限公司 | Target detection method and device |
| US10726244B2 (en) * | 2016-12-07 | 2020-07-28 | Samsung Electronics Co., Ltd. | Method and apparatus detecting a target |
| US10380741B2 (en) * | 2016-12-07 | 2019-08-13 | Samsung Electronics Co., Ltd | System and method for a deep learning machine for object detection |
| CN107194368A (en) * | 2017-06-08 | 2017-09-22 | 广东容祺智能科技有限公司 | A kind of unmanned plane Crude Oil at Sea leakage is searched and emergency processing method |
| CN108010156A (en) * | 2017-11-01 | 2018-05-08 | 北京航天福道高技术股份有限公司 | A kind of round-the-clock autonomous oil field cruising inspection system |
| CN107871119B (en) * | 2017-11-01 | 2021-07-06 | 西安电子科技大学 | Target detection method based on target space knowledge and two-stage prediction learning |
| CN108154159B (en) * | 2017-12-25 | 2018-12-18 | 北京航空航天大学 | A kind of method for tracking target with automatic recovery ability based on Multistage Detector |
| CN108363998A (en) * | 2018-03-21 | 2018-08-03 | 北京迈格威科技有限公司 | A kind of detection method of object, device, system and electronic equipment |
-
2018
- 2018-08-24 CN CN201810970710.5A patent/CN109325520B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101170641A (en) * | 2007-12-05 | 2008-04-30 | 北京航空航天大学 | A method for image edge detection based on threshold sectioning |
Non-Patent Citations (4)
| Title |
|---|
| Detection of stored-grain insects using deep learning;Yufeng Shen等;《Computers and Electronics in Agriculture》;20180228;第145卷;第319-325页 * |
| Image segmentation based on gray stretch and threshold algorithm;Leiming Liu等;《Optik》;20150331;第126卷(第6期);第626-629页 * |
| SAR图像去噪、分割及目标检测方法研究;郝亚冰;《中国优秀硕士学位论文全文数据库 信息科技辑》;20130415(第04期);第I136-795页 * |
| 区域建议网络的细粒度车型识别;杨娟等;《中国图象图形学报》;20180630;第23卷(第6期);第837-845页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109325520A (en) | 2019-02-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109325520B (en) | Method, device and system for checking petroleum leakage | |
| CN107014827B (en) | Transmission line defect analysis method, device and system based on image processing | |
| CN111537515A (en) | Iron tower bolt defect display method and system based on three-dimensional live-action model | |
| AU2017339771B2 (en) | Method and system for remote processing and analysis of industrial asset inspection data | |
| CN106099763A (en) | A kind of power transmission line unmanned machine inspection device | |
| JP7065477B2 (en) | Disaster situation judgment system and disaster judgment flight system | |
| Alharam et al. | Real time AI-based pipeline inspection using drone for oil and gas industries in Bahrain | |
| CN111178424A (en) | Petrochemical production site safety compliance real-time detection system and method | |
| CN111815576B (en) | Method, device, equipment and storage medium for detecting corrosion condition of metal part | |
| Jeon et al. | A real-time drone mapping platform for marine surveillance | |
| CN115082813A (en) | Detection method, unmanned aerial vehicle, detection system and medium | |
| CN105447431A (en) | Docking airplane tracking and positioning method and system based on machine vision | |
| CN112184903A (en) | Method, device, equipment and medium for detecting high-voltage line tree obstacle risk points | |
| CN210511074U (en) | Full-intelligent oil-gas pipeline inspection system based on unmanned aerial vehicle remote sensing and deep learning | |
| KR102514703B1 (en) | An unmanned aerial vehicle for photographing a damaged area of a structure using a plurality of cameras and an apparatus for inspecting damage to a structure using the same | |
| CN116363397A (en) | Equipment fault checking method, device and inspection system | |
| US20240103537A1 (en) | Methods, systems, and devices for inspecting structures and objects | |
| CN111126771B (en) | Safety inspector identification picture quality supervision and guarantee system and method based on regional attention prediction | |
| CN113516120A (en) | Raise dust detection method, image processing method, device, equipment and system | |
| Obaid et al. | Deep Learning Approach for Oil Pipeline Leakage Detection Using Image-Based Edge Detection Techniques. | |
| Astapov et al. | Improvement of energy efficiency of agricultural enterprises through the survey of high-voltage transmission lines using unmanned aerial vehicles | |
| CN106779327B (en) | Power line patrol aircraft, aircraft system and method | |
| US20240061100A1 (en) | Information processing device, information processing method, and non-transitory computer-readable storage medium | |
| KR102569406B1 (en) | System and server for monitoring green and red tide using multi-spectral camera, and method thereof | |
| CN116013018B (en) | Forest fire prevention early warning analysis method and system based on unmanned aerial vehicle detection |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |