Oilfield crude oil leakage detection method, system and related equipment

Abstract

The present invention is applicable to the technical field of crude oil leakage detection, and in particular, relates to a method, system and related equipment for detecting crude oil leakage in an oil field. The method for detecting crude oil leakage in an oil field comprises the following steps: obtaining multi-spectral image data and visible light image data of crude oil in the oil field; extracting feature data in the multi-spectral image data, and splicing and aligning the feature data with the visible light image data; inputting multi-channel data into a target detection network for crude oil leakage target detection; using multiple crude oil leakage data as input of a crude oil spectral data distance comparison model; judging whether the crude oil pixel point data meets a preset condition, and if so, calculating the oil leakage area data according to the crude oil pixel point data as a detection result output. Compared with the prior art, the present invention effectively improves the detection efficiency and accuracy of crude oil leakage in an oil field.

Description

Technical Field

[0001] The present invention is applicable to the technical field of crude oil leakage detection, and in particular relates to an oilfield crude oil leakage detection method, system and related equipment. Background Art

[0002] With the continuous development of oil field development and pipeline transportation, the problem of crude oil leakage has become increasingly prominent. Leakage not only causes waste of resources, but also may have a serious impact on the surrounding environment and ecosystem. In the process of crude oil extraction and transportation in the oil field area, crude oil leakage is inevitable. These leaks penetrate into the surface and have a great impact on the local environment.

[0003] At present, the detection of oil leakage location is mainly carried out by manual means, or by obtaining images of various areas of the oil field through aerial photography, and then manually checking whether there is oil leakage in each photo area. However, due to the wide range of oil fields, the location and size of oil leakage are not fixed, so manual detection is inefficient, and the detection experience methods vary from person to person, the standards are different, and it is difficult to replicate.

[0004] Therefore, there is an urgent need for a new oilfield crude oil leakage detection method, system and related equipment to solve the above technical problems. Summary of the invention

[0005] The present invention provides an oilfield crude oil leakage detection method, system and related equipment, aiming to improve the accuracy and detection efficiency of oilfield crude oil leakage detection.

[0006] In a first aspect, the present invention provides a method for detecting a crude oil leakage in an oilfield, the method comprising the following steps:

[0007] S1, obtaining multispectral image data and visible light image data of crude oil in the oil field;

[0008] S2, extracting feature data from the multispectral image data, and performing splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data;

[0009] S3, inputting the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtaining a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data;

[0010] S4, using the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screening crude oil pixel points in the crude oil leakage data by using the crude oil spectral data distance comparison model to obtain crude oil pixel point data;

[0011] S5. Determine whether the crude oil pixel data meets a preset condition. If so, calculate the oil leakage area data based on the crude oil pixel data as a detection result output.

[0012] Preferably, the crude oil spectral data distance comparison model includes a data extraction module, a data filtering module and a data screening module, and step S4 includes the following sub-steps:

[0013] S41, extracting color space data and multi-channel image data from each crude oil leakage data using the data extraction module;

[0014] S42, using the data filtering module to filter the color space data and the multi-channel image data, filtering out pixels in the color space data that do not meet the preset crude oil color characteristics, and filtering out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data;

[0015] S43, using the data screening module to screen the filtered pixel point data and the crude oil standard spectrum data corresponding to the leakage scene category data according to preset rules to obtain crude oil pixel point data.

[0016] Preferably, the preset rule is: calculating the spectral angle similarity and Mahalanobis distance between the filtered pixel point data and the crude oil standard spectral data corresponding to the leakage scene category data, judging whether the spectral angle similarity and the Mahalanobis distance are within a preset range, and if so, outputting the filtered pixel point data as the crude oil pixel point data.

[0017] Preferably, the preset conditions include a preset leakage scenario, preset crude oil standard spectrum data of the leakage scenario, a preset size and a preset number of crude oil pixels.

[0018] In a second aspect, the present invention further provides an oilfield crude oil leakage detection system, comprising:

[0019] A data acquisition module, used to acquire multi-spectral image data and visible light image data of crude oil in the oil field;

[0020] A processing module, used for extracting feature data from the multispectral image data, and performing splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data;

[0021] A detection module, used for inputting the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtaining a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data;

[0022] a screening module, used to use the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screen crude oil pixels in the crude oil leakage data through the crude oil spectral data distance comparison model to obtain crude oil pixel data;

[0023] The output module is used to determine whether the crude oil pixel point data meets the preset conditions. If so, the oil leakage area data is calculated based on the crude oil pixel point data as the detection result output.

[0024] Preferably, the crude oil spectral data distance comparison model includes a data extraction module, a data filtering module and a data screening module, and the screening module includes the following subunits:

[0025] An extraction unit, used for extracting color space data and multi-channel image data from each of the crude oil leakage data using the data extraction module;

[0026] A filtering unit, configured to filter the color space data and the multi-channel image data using the data filtering module, filter out pixels in the color space data that do not conform to preset crude oil color characteristics, and filter out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data;

[0027] The calculation unit is used to filter the filtered pixel point data and the crude oil standard spectrum data of the corresponding leakage scene category data by using the preset rules provided by the data screening module to obtain crude oil pixel point data.

[0028] Preferably, the preset rule is: calculating the spectral angle similarity and Mahalanobis distance between the filtered pixel point data and the crude oil standard spectral data corresponding to the leakage scene category data, judging whether the spectral angle similarity and the Mahalanobis distance are within a preset range, and if so, outputting the filtered pixel point data as the crude oil pixel point data.

[0029] Preferably, the preset conditions include a preset leakage scenario, preset crude oil standard spectrum data of the leakage scenario, a preset size and a preset number of crude oil pixels.

[0030] In a third aspect, the present invention further provides a computer device, comprising: a memory, a processor, and an oilfield crude oil leakage detection program stored in the memory and executable on the processor, wherein when the processor executes the oilfield crude oil leakage detection program, the processor implements the steps in the oilfield crude oil leakage detection method as described in any one of the above embodiments.

[0031] In a fourth aspect, the present invention further provides a computer-readable storage medium, on which is stored an oilfield crude oil leakage detection program, which, when executed by a processor, implements the steps in the oilfield crude oil leakage detection method as described in any one of the above embodiments.

[0032] Compared with the prior art, the present invention obtains multi-spectral image data and visible light image data of crude oil in the oil field; extracts feature data in the multi-spectral image data, performs splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data; inputs the multi-channel data into the target detection network to detect crude oil leakage targets, and obtains multiple crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data; uses multiple crude oil leakage data as input of the crude oil spectral data distance comparison model, and the crude oil spectral data distance comparison model is used to screen the crude oil pixel points in the crude oil leakage data to obtain crude oil pixel point data; judges whether the crude oil pixel point data meets the preset conditions, and if so, calculates the oil leakage area data as the detection result output according to the crude oil pixel point data. The present invention collects multi-spectral image data and visible light image data in the oil field area, uses the space and color information of the visible light data, and combines the multi-spectral feature information of the corresponding spatial range to judge whether there is crude oil leakage in the collected image range, and calculates the size of the leaked crude oil area. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be described in detail below in conjunction with the accompanying drawings. The above and other aspects of the present invention will become clearer and easier to understand through the detailed description made in conjunction with the following drawings. In the accompanying drawings:

[0034] Figure 1 It is a flowchart of a method for detecting oil field crude oil leakage provided by an embodiment of the present invention;

[0035] Figure 2 It is a structural schematic diagram of an oilfield crude oil leakage detection system provided by an embodiment of the present invention;

[0036] Figure 3 It is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0038] Embodiment 1

[0039] Please refer to Figure 1The present invention provides a method for detecting oilfield crude oil leakage, the method comprising the following steps:

[0040] S1. Acquire multispectral image data and visible light image data of crude oil in an oil field.

[0041] In an embodiment of the present invention, multispectral image data and visible light image data can be collected and acquired through a visual sensor. Multispectral image data refers to grayscale image data of multiple spectral channels, and visible light image data refers to RGB image data composed of three color spectra of red light (Red), green light (Green), and blue light (Blue).

[0042] S2. Extract feature data from the multispectral image data, and perform splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data.

[0043] In the embodiment of the present invention, after the stitching process, multi-channel data of image width*image height*(3+n) is formed (the multi-channel data is the image data), and then the alignment process is performed according to the visible light image data and the feature points of the n grayscale images extracted from the multi-spectral image data. Since the two types of data, the visible light data and the feature data, come from different acquisition lenses, they need to be aligned; and the data between each channel of the multi-spectral usually also come from different acquisition lenses, and also need to be aligned.

[0044] S3. Inputting the multi-channel data into a target detection network to perform crude oil leakage target detection to obtain a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data.

[0045] In an embodiment of the present invention, the target detection network is a multi-category detection neural network model of spatial channel dimension, the input data is 3+n multi-channel data of image width and height, and the crude oil leakage coordinate data of the crude oil leakage in the image is output, and the crude oil leakage coordinate data is embodied in the form of rectangular frame coordinates, that is, the target suspected to be the leakage location in the multi-channel image data will be framed. The leakage scene category data is due to different conditions such as seasons, weather, and leakage location scenes, and the form of leaked crude oil also presents diversified characteristics. For example, depending on the leakage location, there are floating on the water surface, mixed with grass oil, sandy ground leakage, etc.; for example, depending on the season and weather, the form of leaked oil includes pure liquid, semi-dry viscous oil and solid oil blocks; according to the leakage amount, the shape of the leaked oil presents a large area of ​​irregular shape and a star-pointed mottled shape. Therefore, the target detection model needs to distinguish various forms of leaked crude oil, output the category to which it belongs, and provide a reference for the subsequent further judgment algorithm, so that the results of the subsequent step detection are more accurate.

[0046] S4. Using the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screening crude oil pixel points in the crude oil leakage data by using the crude oil spectral data distance comparison model to obtain crude oil pixel point data.

[0047] In the embodiment of the present invention, the crude oil spectral data distance comparison model includes a data extraction module, a data filtering module and a data screening module. Step S4 includes the following sub-steps:

[0048] S41, extracting color space data and multi-channel image data from each crude oil leakage data using the data extraction module.

[0049] Specifically, the color space data is a pixel-level BGR image, where BGR represents three color channels: blue, green, and red.

[0050] S42, using the data filtering module to filter the color space data and the multi-channel image data, filtering out pixels in the color space data that do not meet the preset crude oil color characteristics, and filtering out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data.

[0051] Specifically, for the BGR image data in the crude oil leak coordinate data, it is converted to the HSV color space, and the pixels that do not conform to the crude oil color characteristics are filtered out according to the preset crude oil color characteristics; for the multispectral image data in the crude oil leak coordinate data, according to the overall light intensity information of the multispectral image data, the pixels in each spectral channel that exceed the light intensity threshold are filtered out according to the preset light intensity.

[0052] S43, using the data screening module to screen the filtered pixel point data and the crude oil standard spectrum data corresponding to the leakage scene category data according to preset rules to obtain crude oil pixel point data.

[0053] In an embodiment of the present invention, the preset rule is: calculate the spectral angle similarity (Spectral Angle Mapper, SAM) and Mahalanobis distance (Mahalanobis Distance) between the filtered pixel point data and the corresponding crude oil standard spectrum data of the leakage scene category data, and determine whether the spectral angle similarity and the Mahalanobis distance are within a preset range (the preset range can be set according to actual conditions). If so, the filtered pixel point data is output as the crude oil pixel point data. Among them, the spectral angle similarity is used to measure the spectral similarity between the original hyperspectral data and the reconstructed hyperspectral data, and the Mahalanobis distance is a measurement method for calculating the distance between data points of different dimensions.

[0054] Specifically, each pixel of the multispectral image data is a one-dimensional vector with a length equal to the number of multispectral spectra. Take the seven spectra of 450nm, 550nm, 650nm, 720nm, 750nm, 800nm, and 850nm as examples, which is a one-dimensional vector with a length of 7. By collecting a large number of multispectral image data of leaked crude oil and classifying the data according to different scenes, for each category, a clustering algorithm (clustering algorithm is a data mining algorithm used to divide the data set into different classes or clusters, so that the similarity of data objects in the same cluster is as large as possible, and the difference of data objects in different clusters is as large as possible, such as K-Means, K-Medoids, etc.) is used to extract one or more crude oil standard spectral lines of each leakage scene category data, and the mean and Mahalanobis distance covariance matrix of each leakage scene category are calculated accordingly.

[0055] For suspected crude oil scenes that are difficult to distinguish, such as black water and black liquid oil, the two are very similar in space, and the pixel-level distance algorithm results are also very close. The present invention uses a binary classification network to screen the multispectral data of these two types of samples. For example, a nonlinear SVM or a deep learning neural network. When the target detection network outputs the crude oil standard spectrum data of the leakage scene category data as black liquid oil, this network model is handed over for further judgment, so that the screening results are more accurate.

[0056] S5. Determine whether the crude oil pixel data meets a preset condition. If so, calculate the oil leakage area data based on the crude oil pixel data as a detection result output.

[0057] In an embodiment of the present invention, the oil leakage area data includes the area and position of the oil leakage area, and the preset conditions include a preset leakage scene, preset leakage scene crude oil standard spectrum data, a preset size, and a preset number of crude oil pixels, and the preset size is the size of the rectangular box corresponding to the crude oil pixel data. Further screening the crude oil pixel data by the preset conditions can effectively improve the detection accuracy of the oil leakage area.

[0058] Compared with the prior art, the present invention obtains multi-spectral image data and visible light image data of crude oil in the oil field; extracts feature data in the multi-spectral image data, performs splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data; inputs the multi-channel data into the target detection network to detect crude oil leakage targets, and obtains multiple crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data; uses multiple crude oil leakage data as input of the crude oil spectral data distance comparison model, and the crude oil spectral data distance comparison model is used to screen the crude oil pixel points in the crude oil leakage data to obtain crude oil pixel point data; judges whether the crude oil pixel point data meets the preset conditions, and if so, calculates the oil leakage area data as the detection result output according to the crude oil pixel point data. The present invention collects multi-spectral image data and visible light image data in the oil field area, uses the space and color information of the visible light data, and combines the multi-spectral feature information of the corresponding spatial range to judge whether there is crude oil leakage in the collected image range, and calculates the size of the leaked crude oil area.

[0059] Embodiment 2

[0060] The present invention also provides a crude oil leakage detection system for oil fields, please refer to Figure 2 , Figure 2 2 is a schematic diagram of a structure of an oilfield crude oil leakage detection method system 200 provided in an embodiment of the present invention, which includes:

[0061] 201. A data acquisition module, used for acquiring multi-spectral image data and visible light image data of crude oil in an oil field;

[0062] In an embodiment of the present invention, multispectral image data and visible light image data can be collected and acquired through a visual sensor. Multispectral image data refers to grayscale image data of multiple spectral channels, and visible light image data refers to RGB image data composed of three color spectra of red light (Red), green light (Green), and blue light (Blue).

[0063] 202. A processing module, configured to extract feature data from the multispectral image data, and perform splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data.

[0064] In the embodiment of the present invention, the image width is formed after the splicing process is performed. Image height The multi-channel data (image data) is then aligned according to the visible light image data and the feature points of n grayscale images extracted from the multispectral image data. Since the visible light data and the feature data come from different acquisition lenses, they need to be aligned; and the data between each channel of the multispectral image usually also comes from different acquisition lenses and also needs to be aligned.

[0065] 203. A detection module, configured to input the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtain a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data;

[0066] In an embodiment of the present invention, the target detection network is a multi-category detection neural network model of spatial channel dimension, the input data is 3+n multi-channel data of image width and height, and the crude oil leakage coordinate data of the crude oil leakage in the image is output, and the crude oil leakage coordinate data is embodied in the form of rectangular frame coordinates, that is, the target suspected to be the leakage location in the multi-channel image data will be framed. The leakage scene category data is due to different conditions such as seasons, weather, and leakage location scenes, and the form of leaked crude oil also presents diversified characteristics. For example, depending on the leakage location, there are floating on the water surface, mixed with grass oil, sandy ground leakage, etc.; for example, depending on the season and weather, the form of leaked oil includes pure liquid, semi-dry viscous oil and solid oil blocks; according to the leakage amount, the shape of the leaked oil presents a large area of ​​irregular shape and a star-pointed mottled shape. Therefore, the target detection model needs to distinguish various forms of leaked crude oil, output the category to which it belongs, and provide a reference for the subsequent further judgment algorithm, so that the results of the subsequent step detection are more accurate.

[0067] 204. A screening module, configured to use the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screen crude oil pixels in the crude oil leakage data by using the crude oil spectral data distance comparison model to obtain crude oil pixel data;

[0068] In the embodiment of the present invention, the crude oil spectrum data distance comparison model includes a data extraction module, a data filtering module and a data screening module. The screening module 204 includes the following subunits:

[0069] 2021. An extraction unit, configured to extract color space data and multi-channel image data from each crude oil leakage data using the data extraction module.

[0070] Specifically, the color space data is a pixel-level BGR image, where BGR represents three color channels: blue, green, and red.

[0071] 2022. A screening unit, configured to filter the color space data and the multi-channel image data using the data filtering module, filter out pixels in the color space data that do not conform to preset crude oil color characteristics, and filter out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data.

[0072] Specifically, for the BGR image data in the crude oil leak coordinate data, it is converted to the HSV color space, and the pixels that do not conform to the crude oil color characteristics are filtered out according to the preset crude oil color characteristics; for the multispectral image data in the crude oil leak coordinate data, according to the overall light intensity information of the multispectral image data, the pixels in each spectral channel that exceed the light intensity threshold are filtered out according to the preset light intensity.

[0073] 2023. A computing unit, configured to utilize the data screening module to screen the filtered pixel point data and the crude oil standard spectrum data corresponding to the leakage scene category data according to preset rules to obtain crude oil pixel point data.

[0074] In an embodiment of the present invention, the preset rule is: calculate the spectral angle similarity (Spectral Angle Mapper, SAM) and Mahalanobis distance between the filtered pixel point data and the corresponding crude oil standard spectrum data of the leakage scene category data, determine whether the spectral angle similarity and the Mahalanobis distance are within a preset range, and if so, output the filtered pixel point data as the crude oil pixel point data. Among them, the spectral angle similarity is used to measure the spectral similarity between the original hyperspectral data and the reconstructed hyperspectral data, and the Mahalanobis distance is a measurement method for calculating the distance between data points of different dimensions.

[0075] Specifically, each pixel of the multispectral image data is a one-dimensional vector with a length equal to the number of multispectral spectra. Take the seven spectra of 450nm, 550nm, 650nm, 720nm, 750nm, 800nm, and 850nm as examples, which is a one-dimensional vector with a length of 7. By collecting a large number of multispectral image data of leaked crude oil and classifying the data according to different scenes, for each category, a clustering algorithm (clustering algorithm is a data mining algorithm used to divide the data set into different classes or clusters, so that the similarity of data objects in the same cluster is as large as possible, and the difference of data objects in different clusters is as large as possible, such as K-Means, K-Medoids, etc.) is used to extract one or more crude oil standard spectral lines of each leakage scene category data, and the mean and Mahalanobis distance covariance matrix of each leakage scene category are calculated accordingly.

[0076] For suspected crude oil scenes that are difficult to distinguish, such as black water and black liquid oil, the two are very similar in space, and the pixel-level distance algorithm results are also very close. The present invention uses a binary classification network to screen the multispectral data of these two types of samples. For example, a nonlinear SVM or a deep learning neural network. When the target detection network outputs the crude oil standard spectrum data of the leakage scene category data as black liquid oil, this network model is handed over for further judgment, so that the screening results are more accurate.

[0077] 205. An output module is used to determine whether the crude oil pixel data meets a preset condition. If so, the oil leakage area data is calculated based on the crude oil pixel data and output as a detection result.

[0078] In an embodiment of the present invention, the preset conditions include a preset leakage scene, preset leakage scene crude oil standard spectral data, a preset size, and a preset number of crude oil pixels. Further screening the crude oil pixel data through the preset conditions can effectively improve the detection accuracy of the oil leakage area.

[0079] The oilfield crude oil leakage detection system 200 can implement the steps in the oilfield crude oil leakage detection method in the above embodiment, and can achieve the same technical effects. Please refer to the description in the above embodiment, which will not be repeated here.

[0080] Embodiment 3

[0081] The embodiment of the present invention also provides a computer device, please refer to Figure 3 , Figure 3 300 is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. The computer device 300 includes: a memory 302 , a processor 301 , and an oilfield crude oil leakage detection program stored in the memory 302 and executable on the processor 301 .

[0082] The processor 301 calls the oilfield crude oil leakage detection program stored in the memory 302 to execute the steps of the oilfield crude oil leakage detection method provided by the embodiment of the present invention. Figure 1 , specifically including the following steps:

[0083] S1, obtaining multispectral image data and visible light image data of crude oil in the oil field;

[0084] S2, extracting feature data from the multispectral image data, and performing splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data;

[0085] S3, inputting the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtaining a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data;

[0086] S4, using the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screening crude oil pixel points in the crude oil leakage data by using the crude oil spectral data distance comparison model to obtain crude oil pixel point data;

[0087] S5. Determine whether the crude oil pixel data meets a preset condition. If so, calculate the oil leakage area data based on the crude oil pixel data as a detection result output.

[0088] The computer device 300 provided in the embodiment of the present invention can implement the steps in the oilfield crude oil leakage detection method in the above embodiment, and can achieve the same technical effects. Please refer to the description in the above embodiment and will not be repeated here.

[0089] Embodiment 4

[0090] An embodiment of the present invention further provides a computer-readable storage medium, on which an oilfield crude oil leakage detection program is stored. When the oilfield crude oil leakage detection program is executed by a processor, the various processes and steps in the oilfield crude oil leakage detection method provided in the embodiment of the present invention are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0091] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0092] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0093] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0094] The embodiments of the present invention are described above in conjunction with the accompanying drawings. What is disclosed is only the preferred embodiment of the present invention. However, the present invention is not limited to the above-mentioned specific implementation manner. The above-mentioned specific implementation manner is only illustrative rather than restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms and equivalent changes without departing from the scope of protection of the purpose of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A method for detecting oil field crude oil leakage, characterized in that: The oilfield crude oil leakage detection method comprises the following steps: S1, obtaining multispectral image data and visible light image data of crude oil in the oil field; S2, extracting feature data from the multispectral image data, and performing splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data; S3, inputting the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtaining a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data; S4, using the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screening crude oil pixel points in the crude oil leakage data by using the crude oil spectral data distance comparison model to obtain crude oil pixel point data; S5, judging whether the crude oil pixel point data meets the preset conditions, and if so, calculating the oil leakage area data according to the crude oil pixel point data as the detection result output, wherein the oil leakage area data includes the area and position of the oil leakage area; The crude oil spectrum data distance comparison model includes a data extraction module, a data filtering module and a data screening module. Step S4 includes the following sub-steps: S41, extracting color space data and multi-channel image data from each crude oil leakage data using the data extraction module; S42, using the data filtering module to filter the color space data and the multi-channel image data, filtering out pixels in the color space data that do not meet the preset crude oil color characteristics, and filtering out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data; S43, using the data screening module to screen the filtered pixel point data and the crude oil standard spectrum data corresponding to the leakage scene category data according to preset rules to obtain crude oil pixel point data; The preset rule is: calculate the spectral angle similarity and Mahalanobis distance between the filtered pixel point data and the corresponding crude oil standard spectral data of the leakage scene category data, determine whether the spectral angle similarity and the Mahalanobis distance are within a preset range, and if so, output the filtered pixel point data as the crude oil pixel point data.

2. The oilfield crude oil leakage detection method according to claim 1, characterized in that: The preset conditions include a preset leakage scene, preset crude oil standard spectrum data of the leakage scene, a preset size, and a preset number of crude oil pixels.

3. An oilfield crude oil leakage detection system, characterized in that: include: A data acquisition module, used to acquire multi-spectral image data and visible light image data of crude oil in the oil field; A processing module, used for extracting feature data from the multispectral image data, and performing splicing and alignment processing on the feature data and the visible light image data to obtain multi-channel data; A detection module, used for inputting the multi-channel data into a target detection network to perform crude oil leakage target detection, and obtaining a plurality of crude oil leakage data; wherein the crude oil leakage data includes crude oil leakage coordinate data and leakage scene category data; a screening module, used to use the plurality of crude oil leakage data as input of a crude oil spectral data distance comparison model, and screen crude oil pixels in the crude oil leakage data through the crude oil spectral data distance comparison model to obtain crude oil pixel data; An output module is used to determine whether the crude oil pixel data meets the preset conditions. If so, the oil leakage area data is calculated based on the crude oil pixel data as a detection result output; the oil leakage area data includes the area and location of the oil leakage area; The crude oil spectrum data distance comparison model includes a data extraction module, a data filtering module and a data screening module, and the screening module includes the following subunits: An extraction unit, used for extracting color space data and multi-channel image data from each of the crude oil leakage data using the data extraction module; A filtering unit, configured to filter the color space data and the multi-channel image data using the data filtering module, filter out pixels in the color space data that do not conform to preset crude oil color characteristics, and filter out pixels in the multi-channel image data that exceed a preset light intensity, to obtain filtered pixel data; A calculation unit, configured to use the data screening module to screen the filtered pixel point data and the crude oil standard spectrum data corresponding to the leakage scene category data according to preset rules to obtain crude oil pixel point data; The preset rule is: calculate the spectral angle similarity and Mahalanobis distance between the filtered pixel point data and the corresponding crude oil standard spectral data of the leakage scene category data, determine whether the spectral angle similarity and the Mahalanobis distance are within a preset range, and if so, output the filtered pixel point data as the crude oil pixel point data.

4. The oilfield crude oil leakage detection system according to claim 3, characterized in that: The preset conditions include a preset leakage scene, preset crude oil standard spectrum data of the leakage scene, a preset size, and a preset number of crude oil pixels.

5. A computer device, characterized in that: include: A memory, a processor, and an oilfield crude oil leakage detection program stored in the memory and capable of running on the processor, wherein the processor implements the steps in the oilfield crude oil leakage detection method as described in any one of claims 1-2 when executing the oilfield crude oil leakage detection program.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an oilfield crude oil leakage detection program, and when the oilfield crude oil leakage detection program is executed by the processor, the steps in the oilfield crude oil leakage detection method according to any one of claims 1-2 are implemented.

Images