Method, device, equipment and storage medium for evaluating corrosion of storage tank

Abstract

The present invention discloses a method, apparatus, device, and storage medium for evaluating storage tank corrosion. The method comprises: preprocessing a current inspection photo, including at least extracting the illumination field; generating a rectangular frame label for the tank to be evaluated in the inspection photo using a rectangle recognition sub-model; generating a tank mask label for the tank to be evaluated in the tank rectangular image using a mask recognition sub-model; obtaining a preliminary rust recognition result based on the tank rectangular image using a preliminary rust recognition sub-model; and determining a rust recognition result for the tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result. The present invention eliminates the need for subjective human judgment, resulting in more accurate and stable evaluation results.

Description

Technical Field

[0001] The present invention relates to the field of chemical equipment, and in particular to a method, device, equipment and storage medium for evaluating corrosion of a storage tank. Background Art

[0002] Storage tanks are large containers used to store liquid media. They are commonly found in refineries, oil fields, and oil depots. Tanks form tank farms and are used to store various media. Storage tanks are widely used to store liquid media in oil fields, refineries, stations, and storage areas. During operation, tanks are often affected by the external environment, and corrosion is a common phenomenon. Tank maintenance and repair procedures require periodic assessments of tank corrosion levels to inform inspection and maintenance decisions.

[0003] With the intelligent transformation of refineries, drone inspections have been widely used. By using drones equipped with high-definition cameras, they can capture images of inspection objects (storage tanks) along certain inspection routes to obtain inspection photos including storage tanks.

[0004] In the prior art, after taking inspection photos captured by drones, the assessment of the tank corrosion condition generally requires manual comparison with the naked eye.

[0005] After research, the inventors found that the existing methods for evaluating the corrosion condition of tank bodies have at least the following defects:

[0006] Since manual comparison methods lack objective quantitative indicators and have no unified standards, manual evaluation of tank corrosion conditions on photos is prone to poor stability and accuracy due to personal subjective factors. Summary of the Invention

[0007] The main purpose of the present invention is to improve the stability and accuracy when evaluating the corrosion condition of storage tanks.

[0008] To achieve the above object, the technical solution adopted by the present invention is:

[0009] The present invention discloses a method for evaluating storage tank corrosion, comprising:

[0010] S11, performing preprocessing on the current inspection photo, including at least extracting the illumination field; the inspection photo includes a graphic of the storage tank to be evaluated;

[0011] S12. Generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangular recognition sub-model; the rectangular recognition sub-model is trained and generated using the original inspection photo combined with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on a drone;

[0012] S13. Generate, by a mask recognition sub-model, a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0013] S14, obtaining a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is trained and generated using the rectangular image of the tank body in combination with a rust label;

[0014] S15. Determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0015] Preferably, the present invention further comprises the steps of:

[0016] S16. Determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

[0017] Preferably, the present invention further comprises the steps of:

[0018] S17. Based on the current scan data, generate a SAR image having a coordinate system consistent with the inspection photo corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data and the inspection photo acquired by the image acquisition device are synchronized and have a one-to-one correspondence;

[0019] S18, determining, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result;

[0020] S19. Determine the corrosion depth of the reflective rust area according to the scattering characteristics of the reflective rust area.

[0021] Preferably, in the present invention, determining the corrosion depth of the reflective rust area according to the scattering characteristics of the reflective rust area includes:

[0022] Preset the corresponding relationship between the grayscale value of the SAR image and the corrosion depth;

[0023] The rust depth of the reflective rust area is determined according to the grayscale value and the corresponding relationship of the reflective rust area in the SAR image.

[0024] Preferably, in the present invention, it includes:

[0025] The rectangular frame label is obtained in the original inspection photo by manual frame selection.

[0026] Preferably, in the present invention, the step of generating a rust label comprises:

[0027] The rust seed area features were selected by clustering method;

[0028] The preliminary rust labels were obtained through the seed region growth method;

[0029] Perform clustering again and update the regional characteristics of the central area;

[0030] Updated rust tags again via seed region growth method;

[0031] The iteration ends when the cluster center no longer changes or the iteration termination signal is reached.

[0032] Preferably, in the present invention, generating a tank mask label includes:

[0033] Select the candidate area of the tank through edge detection;

[0034] Select the largest polygonal object as the coarse label;

[0035] The final tank mask label is obtained after manual fine-tuning.

[0036] In another aspect of the present invention, a tank corrosion assessment device is provided, comprising:

[0037] A preset unit is used to perform preprocessing on a current inspection photo, including at least extracting a light field; the inspection photo includes a graphic of the storage tank to be evaluated;

[0038] A rectangle recognition unit is configured to generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangle recognition sub-model; the rectangle recognition sub-model is generated by training the original inspection photo in combination with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on the drone;

[0039] a mask recognition unit configured to generate a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image using a mask recognition sub-model; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0040] A preliminary rust recognition unit is configured to obtain a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is generated by training the rectangular image of the tank body in combination with a rust label;

[0041] A photo rust recognition unit is used to determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0042] Preferably, the present invention further comprises:

[0043] An area ratio calculation unit is used to determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

[0044] Preferably, the present invention further comprises:

[0045] A coordinate conversion unit is configured to generate, based on current scan data, a SAR image having a coordinate system consistent with that of the inspection photograph corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data is synchronized with and has a one-to-one correspondence with the inspection photograph acquired by the image acquisition device;

[0046] a reflective rust determination unit, configured to determine, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result;

[0047] The rust depth determining unit determines the rust depth of the reflective rust area according to the scattering characteristics of the reflective rust area.

[0048] In another embodiment of the present invention, a tank corrosion assessment device is provided, comprising:

[0049] Memory for storing computer programs;

[0050] A processor is used to call and execute the computer program to implement each step of the tank corrosion assessment method as described in any one of the above items.

[0051] In another aspect of the embodiment of the present invention, a storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, each step of the tank corrosion assessment method as described above is implemented.

[0052] Beneficial effects

[0053] The present invention uses original inspection photos, which serve as historical data, as modeling data to construct a rectangular recognition sub-model for generating a rectangular frame label for a storage tank to be evaluated based on the inspection photos; a mask recognition sub-model for generating a tank mask label for the storage tank to be evaluated in the tank rectangular image based on the tank rectangular image; and a preliminary rust recognition sub-model for obtaining a preliminary rust recognition result based on the tank rectangular image. In this way, based on the current inspection photos obtained by the inspection equipment (such as a drone equipped with an image acquisition device), the preliminary rust recognition result of the tank rectangular image can be obtained through the above-mentioned sub-models; then, the rust on the tank top can be identified based on the tank mask label.

[0054] The present invention can accurately identify rust on the tank roof, so the rust area ratio of the tank roof can be accurately calculated based on the identification result; thus, an accurate result of the tank corrosion assessment can be obtained based on the rust area ratio; because the present invention eliminates the human subjective judgment link, its assessment result is more accurate and stable.

[0055] Furthermore, the present invention may also include SAR images collected by a synthetic aperture radar synchronized with the image acquisition device. Since the scattering characteristics of the SAR image can reflect the roughness of the surface of the photographed object, the rust area can be determined in the SAR image based on the recognition results of the inspection photos. Then, the corresponding rust depth is determined based on the different scattering characteristics in the rust area. Since the embodiment of the present invention can not only determine the rust area (or rust area ratio) of the tank roof, but also identify different rust depths in the rust area, it is possible to obtain a more accurate and comprehensive assessment result of the rust degree and rust condition.

[0056] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application and to implement it according to the contents of the specification, and to make the above and other purposes, technical features and advantages of the present application easier to understand, one or more preferred embodiments are listed below and described in detail with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 Schematic diagram of the steps of the tank corrosion assessment method of the present invention;

[0059] Figure 2 Schematic diagram of another step of the tank corrosion assessment method of the present invention;

[0060] Figure 3 Schematic diagram of the structure of the tank corrosion assessment device of the present invention;

[0061] Figure 4 Schematic diagram of another structure of the storage tank corrosion assessment device of the present invention;

[0062] Figure 5 Schematic diagram of the structure of the tank corrosion assessment equipment described in the present invention. DETAILED DESCRIPTION

[0063] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0064] Example 1

[0065] In order to improve the stability and accuracy of evaluating the corrosion condition of storage tanks, refer to Figure 1 , an embodiment of the present invention provides a method for evaluating corrosion of a storage tank, comprising:

[0066] S11, performing preprocessing on the current inspection photo, including at least extracting the illumination field; the inspection photo includes a graphic of the storage tank to be evaluated;

[0067] In an embodiment of the present invention, inspection photos may be obtained by an image acquisition device installed on a drone. For example, a drone equipped with a video or photo device may conduct an inspection above the tank area to obtain orthographic images of the tanks to be evaluated below.

[0068] In the embodiment of the present invention, the evaluation of the corrosion degree of the storage tank is achieved by a computer according to a preset prediction model. That is, based on the inspection photos, the prediction model is used to generate an evaluation result of the corrosion degree of the storage tank to be evaluated.

[0069] Since environmental factors such as light intensity may change at any time, there may be differences in brightness and other factors in different batches of inspection photos; therefore, in order to facilitate image recognition by the prediction model, in an embodiment of the present invention, the obtained inspection photos are preprocessed such as extracting the light field.

[0070] S12. Generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangular recognition sub-model; the rectangular recognition sub-model is trained and generated using the original inspection photo combined with the rectangular frame label; the original inspection photo, which serves as historical data, is framed to obtain a rectangular frame label for the tank body pattern, and a tank body mask label and a rust label are generated; the original inspection photo is acquired by an image acquisition device provided on a drone;

[0071] The prediction model in the embodiment of the present invention includes multiple sub-models (i.e., a rectangle recognition sub-model, a mask recognition sub-model, and a preliminary rust recognition sub-model) to respectively perform different sub-functions, wherein the rectangle recognition sub-model is used to generate a rectangular frame label for the tank to be evaluated in the inspection photo based on the inspection photo.

[0072] In practical applications, the rectangle recognition sub-model can be trained and generated by combining original inspection photos with rectangular frame labels; that is, the original inspection photos as historical data are used as modeling data for training, which can specifically include: firstly, obtaining the rectangular frame label of the tank body from the original inspection photos through manual frame selection; after the rectangular frame selection, rust labels and tank mask labels can be further generated through automatic and semi-automatic labeling methods.

[0073] In practical applications, the specific steps of generating rust labels through automatic label generation may include:

[0074] The rust seed area features were selected by clustering method;

[0075] The preliminary rust labels were obtained through the seed region growth method;

[0076] Perform clustering again and update the regional characteristics of the central area;

[0077] Updated rust tags again via seed region growth method;

[0078] The iteration ends when the cluster center no longer changes or the iteration termination signal is reached.

[0079] In practical applications, the specific steps for generating tank mask labels through semi-automatic label generation can be as follows:

[0080] Edge detection selects candidate areas of the tank;

[0081] Select the largest polygonal object as the coarse label;

[0082] The final tank mask label is obtained after manual fine-tuning.

[0083] It should be noted that, based on the same reason, in the embodiment of the present invention, the original inspection photos also need to be pre-processed, such as extracting the illumination field.

[0084] In practical applications, original inspection photos can also be obtained by an image acquisition device installed on a drone.

[0085] The mask recognition sub-model in the embodiment of the present invention is trained and generated by combining a tank rectangular image with a tank mask label. Specifically, after obtaining a tank rectangular frame label based on the original inspection photo, a corresponding tank rectangular image can be further generated. Then, the tank rectangular image is used as modeling data and trained in combination with the tank mask label generated by automatic and semi-automatic labeling methods, thereby constructing a mask recognition sub-model that can generate a corresponding tank mask label based on the tank rectangular image.

[0086] The preliminary rust recognition sub-model in the embodiment of the present invention is trained and generated using rectangular tank images combined with rust labels. Specifically, the training is performed using rectangular tank images as modeling data, combined with automatic and semi-automatic rust labeling methods to construct a preliminary rust recognition sub-model capable of generating corresponding rust labels based on rectangular tank images. It should be noted that the preliminary rust recognition sub-model in the embodiment of the present invention generates recognition results based on rectangular tank images, while the orthophoto image of the tank body should be circular. Therefore, the recognition results of this preliminary rust recognition sub-model are likely to include recognition of the vicinity of the tank body's exterior. This means that the recognition results of the preliminary rust recognition sub-model are not yet accurate enough.

[0087] Next, the embodiment of the present invention can also use the tank mask label to remove redundant parts in the recognition result of the rust recognition sub-model, thereby obtaining the final photo rust recognition result of the storage tank to be evaluated.

[0088] In practical applications, multiple current inspection photos will be obtained during each inspection. After preprocessing the inspection photos, the rectangle recognition sub-model can be used to generate rectangular frame labels for each tank to be evaluated based on each inspection photo.

[0089] S13. Generate, by a mask recognition sub-model, a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0090] After obtaining the rectangular frame label of the storage tank to be evaluated, it is also necessary to use the mask recognition sub-model to generate the tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image.

[0091] S14, obtaining a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is trained and generated using the rectangular image of the tank body in combination with the rust label;

[0092] In the embodiment of the present invention, each storage tank to be evaluated corresponds to a corresponding rectangular image of the tank body, and the rust corrosion points therein can be identified through the preliminary rust recognition sub-model.

[0093] S15. Determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0094] The recognition result of the preliminary rust recognition sub-model is generated based on the rectangular image of the tank body, while the orthographic image of the tank body should be circular. Therefore, the recognition result is likely to include the recognition of the ground or other equipment near the outside of the tank body; that is, the recognition result of the preliminary rust recognition sub-model is not accurate enough.

[0095] To this end, in an embodiment of the present invention, the tank mask label generated by the mask recognition sub-model is also utilized. The tank mask label and the preliminary rust recognition result are used to eliminate the redundant parts in the preliminary recognition result of the rust recognition sub-model, thereby obtaining the final photo rust recognition result of the storage tank to be evaluated.

[0096] In practical applications, the corrosion degree of the tank roof can be evaluated by the corrosion area ratio. To this end, in an embodiment of the present invention, the following steps are further included:

[0097] S16. Determine the rust area ratio of the storage tank to be evaluated based on the rust identification results in the photo.

[0098] In practical applications, the correspondence between the degree of corrosion of a storage tank and the corrosion area ratio can be set based on the experience of those skilled in the art or a limited number of experiments. Thus, after obtaining the corrosion area ratio of the storage tank to be evaluated, the degree of corrosion of the storage tank to be evaluated can be determined based on the correspondence.

[0099] In summary, the embodiment of the present invention uses the original inspection photos as historical data as modeling data to construct a rectangle recognition sub-model for generating a rectangular frame label for the storage tank to be evaluated based on the inspection photos; a mask recognition sub-model for generating a tank mask label for the storage tank to be evaluated in the tank rectangular image based on the tank rectangular image; and a preliminary rust recognition sub-model for obtaining a preliminary rust recognition result based on the tank rectangular image. In this way, based on the current inspection photos obtained by the inspection equipment (such as a drone equipped with an image acquisition device), the preliminary rust recognition result of the tank rectangular image can be obtained using the above-mentioned sub-models. Then, the rust on the tank roof can be identified based on the tank mask label.

[0100] The present invention can accurately identify rust on the tank roof, so the rust area ratio of the tank roof can be accurately calculated based on the identification result; thus, an accurate result of the tank corrosion assessment can be obtained based on the rust area ratio; because the present invention eliminates the human subjective judgment link, its assessment result is more accurate and stable.

[0101] Example 2

[0102] refer to Figure 2 Based on the first embodiment, the embodiment of the present invention may further include the following steps:

[0103] S17. Based on the current scan data, generate a SAR image having a coordinate system consistent with the inspection photo corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data and the inspection photo acquired by the image acquisition device are synchronized and have a one-to-one correspondence;

[0104] In an embodiment of the present invention, the correlation between the scattering characteristics of the SAR image of the scanning object and the roughness of the scanning object surface is further utilized to determine the corrosion depth of the storage tank. Specifically, the SAR image of the scanning object obtained by the SAR can be used to infer the roughness of the scanning object surface based on the scattering characteristics of the SAR image. For a storage tank, the higher the roughness of the rust on the tank, the greater the corrosion depth value. The corrosion depth is also an important indicator for evaluating the degree of corrosion of the storage tank.

[0105] If a drone carries a synthetic aperture radar alone to scan and obtain scanning data of the tank area, the scanning data obtained by the moving synthetic aperture radar will cause position accuracy deviation due to the spatial inertial navigation combination solution, which will make the SAR image obtained thereby unusable directly. For this reason, in an embodiment of the present invention, a drone is used to carry an image acquisition device and a synthetic aperture radar at the same time to synchronously obtain inspection photos and scanning data. In this way, a SAR image is generated with reference to the coordinates of the inspection photos corresponding to the scanning data, which can avoid the problem of position accuracy deviation of the SAR image generated by the drone carrying a synthetic aperture radar alone.

[0106] S18, determining, in the SAR image, a reflective rust area corresponding to the rust area in the rust identification result of the photo;

[0107] Since the coordinates of the SAR image and the inspection photo are consistent, in the embodiment of the present invention, the corresponding area can be determined from the SAR image according to the rust area in the photo rust recognition result as the reflective rust area.

[0108] S19. Determine the corrosion depth of the reflective rust area based on the scattering characteristics of the reflective rust area.

[0109] Synthetic aperture radar can be used to obtain the scattering characteristics of the scanned object's surface. A typical example of this is that the roughness of the scanned object's surface determines the grayscale value (i.e., echo intensity) of the SAR image. The inventors have discovered through research that the roughness of rust on the surface of a storage tank is strongly correlated with the corrosion depth. Therefore, in an embodiment of the present invention, the scattering characteristics characterized by the SAR image are used as a measurement parameter for the corrosion depth. That is, the greater the grayscale value of the SAR image, the deeper the corrosion depth, and vice versa.

[0110] Specifically, when determining the corrosion depth of the reflective rust area, the correspondence between the grayscale value of the SAR image and the corrosion depth can be preset; then the corrosion depth of the reflective rust area is determined based on the grayscale value of the reflective rust area in the SAR image and the correspondence.

[0111] In summary, the embodiment of the present invention, based on the first embodiment, may further include SAR images acquired by a synthetic aperture radar synchronized with the image acquisition device. Since the scattering characteristics of the SAR image can reflect the roughness of the surface of the photographed object, the rust area can be determined in the SAR image based on the recognition results of the inspection photos. Then, the corresponding rust depth is determined based on the different scattering characteristics in the rust area. Since the embodiment of the present invention can not only determine the rust area (or rust area ratio) of the tank roof, but also identify different rust depths in the rust area, it is possible to obtain a more accurate and comprehensive assessment result of the rust degree and rust condition.

[0112] Example 3

[0113] In another embodiment of the present invention, a tank corrosion assessment device is provided. Figure 3 The structural diagram of the tank corrosion assessment device provided by the embodiment of the present invention is shown. Figure 1 or Figure 2 The device corresponding to the tank corrosion assessment method described in the corresponding embodiment, that is, the device is implemented by means of a virtual device. Figure 1 or Figure 2 In the corresponding embodiment of the tank corrosion assessment method, each virtual module constituting the tank corrosion assessment device can be executed by an electronic device, such as a network device, a terminal device, or a server. The tank corrosion assessment device in the embodiment of the present invention can implement the tank corrosion assessment required for industrial control. Specifically, the tank corrosion assessment device in the embodiment of the present invention includes:

[0114] The preset unit 01 is configured to perform preprocessing on a current inspection photo, including at least extracting a light field; the inspection photo includes a graphic of a storage tank to be evaluated;

[0115] Rectangle recognition unit 02, configured to generate a rectangular frame label for the tank to be evaluated based on the inspection photo using a rectangle recognition sub-model; the rectangle recognition sub-model is trained and generated using original inspection photos combined with rectangular frame labels; the original inspection photos are acquired by an image acquisition device provided on a drone;

[0116] The mask recognition unit 03 is configured to generate a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image using a mask recognition sub-model; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0117] A preliminary rust recognition unit 04 is configured to obtain a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is generated by training the rectangular image of the tank body in combination with a rust label;

[0118] The photo rust recognition unit 05 is configured to determine the photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0119] Preferably, in the embodiment of the present invention, the following may also be included:

[0120] The area ratio calculation unit 06 is used to determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

[0121] For further reference, Figure 4 In an embodiment of the present invention, the method may further include:

[0122] A coordinate conversion unit 07 is configured to generate, based on current scan data, a SAR image having a coordinate system consistent with the inspection photo corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data is synchronized with and has a one-to-one correspondence with the inspection photos acquired by the image acquisition device;

[0123] a reflective rust determination unit 08, configured to determine, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result;

[0124] The corrosion depth determining unit 09 determines the corrosion depth of the reflective rust area according to the scattering characteristics of the reflective rust area.

[0125] Since the working principle and beneficial effects of the tank corrosion assessment device in the embodiment of the present invention have been Figure 1 The corresponding tank corrosion assessment method is also recorded and explained, so you can refer to it and will not repeat it here.

[0126] Example 4

[0127] Corresponding to the above-mentioned method embodiments, the present application also provides a tank corrosion assessment device, such as a terminal or server. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be, but is not limited to, a smartphone, tablet computer, laptop computer, or desktop computer.

[0128] An example of a hardware structure block diagram of a storage tank corrosion assessment device provided by an embodiment of the present invention is shown in FIG. Figure 5 As shown, this may include:

[0129] Processor 1, communication interface 2, memory 3 and communication bus 4;

[0130] The processor 1, the communication interface 2, and the memory 3 communicate with each other via the communication bus 4;

[0131] Optionally, the communication interface 2 may be an interface of a communication module, such as an interface of a GSM module; the processor 1 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement an embodiment of the present invention.

[0132] The memory 3 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0133] The processor 1 is specifically configured to execute the computer program stored in the memory 3 to perform the following steps:

[0134] S11, performing preprocessing on the current inspection photo, including at least extracting the illumination field; the inspection photo includes a graphic of the storage tank to be evaluated;

[0135] S12. Generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangular recognition sub-model; the rectangular recognition sub-model is trained and generated using the original inspection photo combined with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on a drone;

[0136] S13. Generate, by a mask recognition sub-model, a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0137] S14, obtaining a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is trained and generated using the rectangular image of the tank body in combination with a rust label;

[0138] S15. Determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0139] Preferably, in the embodiment of the present invention, the steps may also be included:

[0140] S16. Determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

[0141] Furthermore, in the embodiment of the present invention, the steps may also be included:

[0142] S17. Based on the current scan data, generate a SAR image having a coordinate system consistent with the inspection photo corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data and the inspection photo acquired by the image acquisition device are synchronized and have a one-to-one correspondence;

[0143] S18, determining, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result;

[0144] S19. Determine the corrosion depth of the reflective rust area according to the scattering characteristics of the reflective rust area.

[0145] The storage tank corrosion assessment device in the embodiment of the present invention, when the program instructions included in its computer program product are executed by a computer, can enable the computer to execute the storage tank corrosion assessment method described in the above aspects and achieve the same technical effects.

[0146] Example 5

[0147] In an embodiment of the present invention, a storage medium is further provided. The storage medium may store a program suitable for execution by a processor, wherein the program is used to:

[0148] S11, performing preprocessing on the current inspection photo, including at least extracting the illumination field; the inspection photo includes a graphic of the storage tank to be evaluated;

[0149] S12. Generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangular recognition sub-model; the rectangular recognition sub-model is trained and generated using the original inspection photo combined with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on a drone;

[0150] S13. Generate, by a mask recognition sub-model, a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label;

[0151] S14, obtaining a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; the preliminary rust recognition sub-model is trained and generated using the rectangular image of the tank body in combination with a rust label;

[0152] S15. Determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result.

[0153] Furthermore, in the embodiment of the present invention, the steps may also be included:

[0154] S16. Determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

[0155] Furthermore, in the embodiment of the present invention, the steps may also be included:

[0156] S17. Based on the current scan data, generate a SAR image having a coordinate system consistent with the inspection photo corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data and the inspection photo acquired by the image acquisition device are synchronized and have a one-to-one correspondence;

[0157] S18, determining, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result;

[0158] S19. Determine the corrosion depth of the reflective rust area according to the scattering characteristics of the reflective rust area.

[0159] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0160] The above-mentioned product can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0161] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0162] In the several embodiments provided herein, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0163] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0164] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0165] It should be understood that in the embodiments of the present application, the various embodiments and features can be combined with each other to solve the aforementioned technical problems.

[0166] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0167] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating tank corrosion, characterized in that: Including steps: S11, performing preprocessing on the current inspection photo, including at least extracting the illumination field; the inspection photo includes a graphic of the storage tank to be evaluated; S12. Generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangular recognition sub-model; the rectangular recognition sub-model is trained and generated using the original inspection photo combined with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on a drone; S13. Generate, by a mask recognition sub-model, a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label; S14, obtaining a preliminary rust recognition result based on the rectangular image of the tank body using a preliminary rust recognition sub-model; The preliminary rust recognition sub-model is generated by training the tank rectangular image in combination with the rust label; Generating the rust label includes: selecting rust seed region features by a clustering method; obtaining preliminary rust labels by a seed region growing method; performing clustering again to update the central region features; updating the rust label by the seed region growing method again; and ending when the cluster center no longer changes or an iteration termination signal is reached; S15, determining a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result; Also includes the steps: Based on the current scan data, a synthetic aperture radar (SAR) image is generated that is consistent with the inspection photo coordinate system corresponding to the scan data; the UAV is also provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data is synchronized with the inspection photos acquired by the image acquisition device and has a one-to-one correspondence; determining, in the SAR image, a reflective rust region corresponding to the rust region in the photo rust recognition result; Determining the rust depth of the reflective rust area according to the scattering characteristics of the reflective rust area includes: presetting a correspondence between the grayscale value of the SAR image and the rust depth; and determining the rust depth of the reflective rust area according to the grayscale value of the reflective rust area in the SAR image and the correspondence.

2. The tank corrosion assessment method according to claim 1, characterized in that: Also includes the steps: S16. Determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

3. The tank corrosion assessment method according to claim 1 or 2, characterized in that: include: The rectangular frame label is obtained in the original inspection photo by manual frame selection.

4. The tank corrosion assessment method according to claim 1 or 2, characterized in that: The step of generating a tank mask label includes: Select the candidate area of the tank through edge detection; Select the largest polygonal object as the coarse label; The final tank mask label is obtained after manual fine-tuning.

5. A tank corrosion assessment device, characterized in that: include: A preset unit, configured to perform preprocessing on the current inspection photo, including at least extracting the illumination field; The inspection photos include graphics of the storage tank to be evaluated; A rectangle recognition unit is configured to generate a rectangular frame label for the storage tank to be evaluated based on the inspection photo using a rectangle recognition sub-model; the rectangle recognition sub-model is generated by training the original inspection photo in combination with the rectangular frame label; the original inspection photo is acquired by an image acquisition device provided on the drone; a mask recognition unit configured to generate a tank mask label of the storage tank to be evaluated in the tank rectangular image according to the tank rectangular image using a mask recognition sub-model; the mask recognition sub-model is trained and generated by combining the tank rectangular image with the tank mask label; the tank rectangular image is obtained according to the rectangular label; a preliminary rust recognition unit, configured to obtain a preliminary rust recognition result based on the rectangular image of the tank body by using a preliminary rust recognition sub-model; The preliminary rust recognition sub-model is generated by training the tank rectangular image in combination with the rust label; Generating the rust label includes: selecting rust seed region features by a clustering method; obtaining preliminary rust labels by a seed region growing method; performing clustering again to update the central region features; updating the rust label by the seed region growing method again; and ending when the cluster center no longer changes or an iteration termination signal is reached; a photo rust recognition unit, configured to determine a photo rust recognition result of the storage tank to be evaluated based on the tank mask label generated by the mask recognition sub-model and the preliminary rust recognition result; A coordinate conversion unit is configured to generate, based on current scan data, a SAR image having a coordinate system consistent with that of the inspection photograph corresponding to the scan data; the drone is further provided with a synthetic aperture radar for acquiring scan data synchronized with the image acquisition device; the scan data is synchronized with and has a one-to-one correspondence with the inspection photograph acquired by the image acquisition device; a reflective rust determination unit, configured to determine, in the SAR image, a reflective rust area corresponding to the rust area in the photo rust recognition result; The rust depth determining unit determines the rust depth of the reflective rust area based on the scattering characteristics of the reflective rust area, including: presetting a correspondence between the grayscale value of the SAR image and the rust depth; and determining the rust depth of the reflective rust area based on the grayscale value of the reflective rust area in the SAR image and the correspondence.

6. The tank corrosion assessment device according to claim 5, characterized in that: Also includes: An area ratio calculation unit is used to determine the rust area ratio of the storage tank to be evaluated based on the rust identification result of the photo.

7. A tank corrosion assessment device comprising: memory for storing computer programs; A processor is used to call and execute the computer program to implement each step of the tank corrosion assessment method according to any one of claims 1 to 4.

8. A storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the computer program implements the steps of the tank corrosion assessment method according to any one of claims 1 to 4.

Images