Method, apparatus, device and readable storage medium for detecting liquid fuel leakage in a pipeline

Abstract

The application provides a method, device, equipment and readable storage medium for detecting liquid fuel leakage in a pipeline, the method comprising: acquiring a global image and a local image of a target pipeline respectively; establishing a neural network pipeline detection model based on optical flow according to the global image and the local image; acquiring a global image and a local image of a target pipeline to be detected; inputting the global image and the local image of the target pipeline to be detected into the pipeline detection model, and predicting whether liquid fuel leakage occurs in the pipeline through the pipeline detection model; and in the case that liquid fuel leakage occurs in the pipeline through the pipeline detection model, sending corresponding global images and local images to a terminal user. The method provided by the application improves detection accuracy.

Description

Technical Field

[0001] This invention relates to the field of pipeline inspection technology, and in particular to a method, apparatus, equipment, and readable storage medium for detecting liquid fuel leaks in pipelines. Background Technology

[0002] In complex test environments, numerous pipes, flanges, high-temperature components, and supporting test equipment exist. Different pipes transport different media; for example, gas pipelines are specifically designed for transporting high-temperature, high-pressure gases, while fuel pipelines are specifically designed for transporting liquid fuels. During testing, leaks of high-temperature gases or liquid fuels may occur due to various reasons, such as vibration, high-temperature deformation, loose or uneven flange connections, or damage. In such situations, the test equipment, test pieces, or test devices at the test site are unmonitored. Therefore, it is necessary to monitor the pipelines for liquid leaks in real time to ensure that no accidents occur due to liquid fuel leaks throughout the entire test.

[0003] The current method for detecting liquid leaks in pipelines involves first acquiring images of the pipeline and then analyzing these images to determine if a leak exists. However, the accuracy of current methods for detecting leaks in pipelines using pipeline images is not high enough. Summary of the Invention

[0004] This invention provides a method, apparatus, device, and readable storage medium for detecting liquid fuel leaks in pipelines, thereby overcoming the shortcomings of existing technologies in accurately detecting whether liquid fuel leaks in pipelines and achieving efficient detection of liquid fuel leaks in pipelines.

[0005] A method for detecting liquid fuel leaks in a pipeline includes:

[0006] At the same time, a global image and a local image of the target pipeline are acquired respectively; the global image has the global view of the target pipeline; the local image has images of key parts of the target pipeline.

[0007] A neural network pipeline detection model based on optical flow is established based on the global and local images;

[0008] Obtain global and local images of the target pipeline to be detected;

[0009] The global and local images of the target pipeline to be detected are input into the pipeline detection model, and the pipeline detection model is used to predict whether a liquid fuel leak has occurred in the pipeline.

[0010] When a pipeline leak is predicted to occur due to liquid fuel leakage using a pipeline detection model, the corresponding global and local images are sent to the end user.

[0011] Furthermore, in the method for detecting liquid fuel leakage in a pipeline as described above, acquiring global and local images of the target pipeline to be detected includes:

[0012] Global and local images of the target pipe to be detected are acquired at a preset frequency.

[0013] Determine whether there are differences between the currently acquired global image and local image and the corresponding previous global image and local image;

[0014] If there are differences between two consecutive global or local images, the corresponding global or local images are stored in memory respectively.

[0015] The global and local images stored in the memory are used as input to the pipeline detection model.

[0016] Furthermore, in the method for detecting liquid fuel leakage in a pipeline as described above, the step of using the global image and local image stored in memory as input to the pipeline detection model includes:

[0017] Determine whether the total number of global and local images with discrepancies exceeds a preset number. If it does, extract the global and local images exceeding the preset number from the memory and use them as input to the pipeline detection model.

[0018] Furthermore, in the method for detecting liquid fuel leakage in a pipeline as described above, if there is no difference between two consecutive global or local images, the image in which there is no difference is deleted.

[0019] Furthermore, in the method for detecting liquid fuel leakage in a pipeline as described above, if the pipeline detection model predicts that no liquid fuel leakage has occurred, the global and local images input to the pipeline detection model are deleted from memory.

[0020] Furthermore, in the method for detecting liquid fuel leakage in a pipeline as described above, when there are differences between two consecutive global or local images, storing the corresponding global or local images in their respective memories includes:

[0021] Create corresponding subdirectory folders according to the year, month, day, and hour of image acquisition, and store the global image and local image in the corresponding subdirectory folders according to the time of image acquisition.

[0022] An apparatus for detecting liquid fuel leaks in pipelines, comprising:

[0023] The acquisition unit is used to acquire a global image and a local image of the target pipeline at the same time; the global image has the global view of the target pipeline; the local image has images of key parts of the target pipeline.

[0024] The training unit is used to establish a neural network pipeline detection model based on optical flow based on the global image and local images;

[0025] The acquisition unit is also used to acquire global and local images of the target pipeline to be detected;

[0026] The prediction unit is used to input the global image and local image of the target pipeline to be detected into the pipeline detection model, and predict whether the pipeline has a liquid fuel leak through the pipeline detection model.

[0027] The sending unit is used to send the corresponding global and local images to the terminal when a liquid fuel leak is predicted to occur in the pipeline through the pipeline detection model.

[0028] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method for detecting liquid fuel leakage in a pipeline as described above.

[0029] A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the method for detecting liquid fuel leakage in a pipeline as described above.

[0030] A computer program product includes a computer program that, when executed by a processor, implements the method for detecting liquid fuel leakage in a pipeline as described above.

[0031] The method provided by this invention, by setting the focal length and field of view of different optical sensors, completes the image capture of different sensors for different fields of view and different leakage events, ensuring the capture of phenomena from tiny fuel particle dripping to liquid fuel overflow, splashing and fuel evaporation, and the acquired images are predicted by a neural network pipeline detection model based on optical flow. This can meet the online detection of different liquid fuel leakage modes in complex test sites and improve the detection accuracy. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figure 1 This is one of the flowcharts of the method for detecting liquid fuel leakage in a pipeline provided by the present invention;

[0034] Figure 2 This is a schematic diagram illustrating the principle of the present invention for detecting liquid fuel leaks in pipelines.

[0035] Figure 3 This is the second flowchart of the method for detecting liquid fuel leakage in pipelines according to the present invention;

[0036] Figure 4 A flowchart illustrating the process of creating a subdirectory for this invention;

[0037] Figure 5 A schematic diagram of the device for detecting liquid fuel leakage in pipelines provided by the present invention;

[0038] Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0040] Figure 1 This is one of the flowcharts of the method for detecting liquid fuel leakage in a pipeline provided by the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0041] Step 101: At the same time, acquire the global image and local image of the target pipeline respectively; the global image has the global view of the target pipeline; the local image has the key part image of the target pipeline.

[0042] Specifically, in actual testing scenarios, leak detection of liquid fuels mainly focuses on two scenarios:

[0043] (1) The fuel pipeline network that supplies liquid fuel to the test piece is concentrated in a dense fuel pipeline network. The possible forms of fuel leakage include: tiny fuel droplets (mainly appearing at valves, flanges and interfaces), large-scale fuel overflow (i.e., the leakage fuel flow is large, forming an oil film that drips as a whole on the pipeline, or dripping in multiple places at the same time in the form of droplets), and fuel splashing (the leakage volume is large, and under pressure, it is sprayed out through the rupture).

[0044] (2) The above-mentioned problems that occurred on the test line / test piece, as well as the phenomenon of leaked fuel evaporation caused by high wall temperature.

[0045] Of the fuel leak types mentioned above, overflow, splashing, and evaporation are relatively obvious leak events that can be directly identified using optical cameras. However, tiny fuel droplets are more difficult to detect. For example, in a complex fuel pipeline system, a tiny fuel droplet leak might be missed due to insufficient optical resolution in a wide field of view. Therefore, in real-world scenarios where fuel pipelines are dense and pipe diameters are small, the diameter of leaking fuel particles is also small. When shooting from a wide field of view, insufficient optical resolution may cause tiny fuel droplets to be overlooked. Therefore, this study uses two optical sensors to acquire optical images from different scenarios for analysis, thereby improving the accuracy of image recognition. The specific approach is as follows:

[0046] A single-channel optical sensor is used for capturing distant, global views. Images acquired from this sensor are primarily used for detecting significant leaks, including spills, splashes, and fuel evaporation.

[0047] The optical sensor in the other channel uses zoom and local field of view to capture images of key monitoring locations (such as valves, flanges, joints, etc.), mainly to monitor minute fuel droplet events, and can also monitor localized overflow, splashing and other phenomena.

[0048] Step 102: Establish a neural network pipeline detection model based on optical flow based on the global image and local images.

[0049] Step 103: Obtain the global and local images of the target pipeline to be detected;

[0050] Step 104: Input the global image and local image of the target pipeline to be detected into the pipeline detection model, and predict whether liquid fuel leakage has occurred in the pipeline through the pipeline detection model;

[0051] Step 105: If a liquid fuel leak is predicted in the pipeline using the pipeline detection model, the corresponding global and local images are sent to the end user.

[0052] The method provided by this invention, by setting the focal length and field of view of different optical sensors, completes the image capture of different sensors for different fields of view and different leakage events, ensuring the capture of phenomena from tiny fuel particle dripping to liquid fuel overflow, splashing and fuel evaporation, and the acquired images are predicted by a neural network pipeline detection model based on optical flow. This can meet the online detection of different liquid fuel leakage modes in complex test sites and improve the detection accuracy.

[0053] Furthermore, acquiring the global and local images of the target pipeline to be detected includes:

[0054] The global and local images of the target pipeline to be detected are acquired at a preset frequency; it is determined whether there is a difference between the currently acquired global and local images and the previous corresponding global and local images; if there is a difference between the two images, the corresponding global and local images are stored in memory; if there is no difference between the two images, the image with no difference is deleted.

[0055] Determine whether the total number of global and local images with discrepancies exceeds a preset number. If it does, extract the global and local images exceeding the preset number from the memory and use them as input to the pipeline detection model.

[0056] If a liquid fuel leak is predicted in a pipeline based on the pipeline detection model, the corresponding global and local images are sent to the end user; if no liquid fuel leak is predicted in a pipeline based on the pipeline detection model, the global and local images input to the pipeline detection model are deleted from memory.

[0057] Specifically, since each optical camera has a frame rate of 25 frames per second, and the processing of optical images cannot affect the acquisition of normal optical images, it can be concluded that after acquiring an image frame, if the resolution scaling and comparison with the previous image cannot be completed within 40ms, the acquired image must be stored in a cache, the image file in the cache must be read for differential comparison, and images with significant differences must be pushed to shared memory to meet the time requirements. Through practice, it can be found that: the time consumption for acquiring an optical image frame is about 10ms, the resolution scaling takes about 2ms, the time consumption for pushing images with significant differences to shared memory is about 10ms, the image comparison time is about 10ms when there are no significant differences between adjacent images, and the comparison time is about 60ms when there are significant differences. Figure 2 This is a schematic diagram illustrating the principle of the present invention for detecting liquid fuel leaks in pipelines, as shown below. Figure 2 As shown, this invention uses a high-speed cache to store real-time acquired image files, and a timer scans the cache to obtain images for comparison. This invention employs a combination of caching, image comparison, and pushing to shared memory to perform image difference checks. Only when adjacent images show significant differences (exceeding a set difference threshold) will the image be pushed to shared memory, thereby effectively reducing the computational load of the deep neural network module.

[0058] Figure 3This is a second flowchart of the method for detecting liquid fuel leakage in pipelines according to the present invention, as shown below. Figure 3 As shown, the method provided by this invention will be described in detail below:

[0059] Step 301: Activate the optical sensor and begin capturing images of the detected object at a frame rate of 25 frames per second.

[0060] Step 302: Start the program that resides in the computer memory (called a daemon process on Linux-like operating systems and a service on Windows operating systems), start a timer, and scan the cache for captured optical images in the timer event. If they exist, retrieve the captured images from the cache.

[0061] Multi-channel refers to the use of two optical sensors in this invention to capture the same scene. Since the fuel pipeline network in the real scene is relatively dense and the pipe diameter is relatively small, the diameter of the fuel leak particles is small. When the field of view is large, the small fuel leak droplets may be ignored due to insufficient optical resolution. Therefore, the two optical sensors capture different levels of the scene.

[0062] Step 303: Perform difference analysis on the captured optical images. Here, peak signal-to-noise ratio is used as the method to determine the differences between adjacent optical images. The core algorithm is as follows:

[0063] Peak Signal-to-Noise Ratio (PSNR) is commonly used in image compression. Let the image before compression be P1, and the image after compression be P2. By objectively evaluating the degradation of P1 and P2, the degree of image distortion can be assessed. The PSNR evaluation result is usually expressed in logarithmic decibels (dB). In images P1 and P2, a higher PSNR value indicates less degradation; when the degradation is significant, the PSNR value approaches 0 dB. Therefore, if P1 and P2 are considered as two images to compare their similarity, a higher PSNR value indicates that P2 is less compressed than P1, meaning P2 is more similar to P1; conversely, a lower PSNR value indicates that P2 is more compressed than P1, meaning P2 is less similar to P1. Thus, PSNR can be applied to image similarity comparisons, and a similarity value can be determined experimentally or using defined error standards to measure whether P2 has changed compared to P1.

[0064] The formula for PSNR is as follows:

[0065]

[0066] Among them, MAX I MAX represents the maximum grayscale value of the denoised image. IThe size depends on the number of grayscale bits of the image (generally 8-bit, N = 255). In equation (1), MSE (mean square error) is the mean square error, as shown in equation (2):

[0067]

[0068] In equation (2), I(i,j) and K(i,j) are the corresponding gray values ​​or color values ​​in P1 and P2, respectively, and m-1 and n-1 are m×n pixels.

[0069] Compared with other image difference comparison algorithms, the peak signal-to-noise ratio method has the advantage of high speed. It can complete the comparison of two 1280*720 resolution (24-bit color) images within milliseconds, which meets the time requirements for multi-channel optical image comparison.

[0070] Step 304: Images that are detected to be significantly different from the previous frame are pushed into a pre-constructed shared memory. When the number of images in the shared memory reaches 4, the deep neural network learning module is notified to retrieve the image data from the shared memory.

[0071] Step 305: The deep neural network learning module retrieves the image from shared memory and uses an optical flow-based model to detect whether there is a liquid fuel leak in the image. Once a leak is detected, it marks the image with a rectangular area and text, notifies the control / monitoring computer via TCP network, and sends the detection result image to the computer as a file. The computer will display the result as an alarm, and the user decides on the next step.

[0072] Furthermore, the step of storing the corresponding global image and local image in memory when there are differences between the two images includes: creating corresponding subdirectory folders according to the year, month, day, and hour of image acquisition, and storing the global image and local image in the corresponding subdirectory folders according to the time of image acquisition.

[0073] Specifically, images that are significantly different from the previous frame are saved to a designated directory. This directory structure is based on time and device configuration to avoid slowing down processing caused by storing a large number of image files (hundreds of thousands or more, mainly due to excessive storage files during long-term monitoring) in the same directory. Figure 4 A flowchart illustrating the process of creating a subdirectory for this invention is shown below. Figure 4 As shown, this invention decomposes storage time and saves the corresponding image files into subdirectories refined to the hour-device level, thereby achieving distributed storage of massive image data and avoiding the decline in read and query performance caused by too many files in a single directory.

[0074] The method provided by this invention uses high-speed caching, image comparison, and pushing to shared memory to complete the image difference check. Only when there are significant differences between adjacent images (exceeding the set difference threshold) will the image be pushed to shared memory, effectively reducing the computational load of the deep neural network module.

[0075] The following describes the device for detecting liquid fuel leaks in pipelines provided by the present invention. The device for detecting liquid fuel leaks in pipelines described below can be referred to in correspondence with the method for detecting liquid fuel leaks in pipelines described above.

[0076] Figure 5 The schematic diagram of the device for detecting liquid fuel leakage in pipelines provided by the present invention is as follows: Figure 5 As shown, the device includes:

[0077] The acquisition unit 501 is used to acquire a global image and a local image of the target pipeline at the same time; the global image has the global view of the target pipeline; the local image has images of key parts of the target pipeline.

[0078] Training unit 502 is used to establish a neural network pipeline detection model based on optical flow based on the global image and local images;

[0079] The acquisition unit 501 is also used to acquire a global image and a local image of the target pipeline to be detected;

[0080] Prediction unit 503 is used to input the global image and local image of the target pipeline to be detected into the pipeline detection model, and predict whether the pipeline has a liquid fuel leak through the pipeline detection model;

[0081] The sending unit 504 is used to send the corresponding global image and local image to the terminal when a liquid fuel leak is predicted to occur in the pipeline through the pipeline detection model.

[0082] Figure 6 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 6As shown, the electronic device may include a processor 610, a communications interface 620, a memory 630, and a communication bus 640, wherein the processor 610, communications interface 620, and memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a method for detecting liquid fuel leakage in a pipeline. This method includes: simultaneously acquiring a global image and a local image of the target pipeline; the global image has a global view of the target pipeline; the local image has images of key parts of the target pipeline.

[0083] A neural network-based pipeline detection model based on optical flow is established based on the global and local images;

[0084] Obtain global and local images of the target pipeline to be detected;

[0085] The global and local images of the target pipeline to be detected are input into the pipeline detection model, and the pipeline detection model is used to predict whether a liquid fuel leak has occurred in the pipeline.

[0086] When a pipeline leak is predicted to occur due to liquid fuel leakage using a pipeline detection model, the corresponding global and local images are sent to the end user.

[0087] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0088] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer is able to execute the methods for detecting liquid fuel leakage in pipelines provided by the above methods.

[0089] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the methods for detecting liquid fuel leakage in pipelines provided by the above methods.

[0090] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. 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 the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of detecting a leak of a liquid fuel in a pipeline, characterized by, The method comprises the following steps: acquiring a global image and a local image of a target pipeline at the same time, respectively; the global image has a global field of view of the target pipeline; the local image has an image of a key part of the target pipeline; establishing a neural network pipeline detection model based on optical flow according to the global image and the local image; acquiring a global image and a local image of a target pipeline to be detected; inputting the global image and the local image of the target pipeline to be detected into the pipeline detection model to predict whether liquid fuel leakage occurs in the pipeline through the pipeline detection model; in the case that the pipeline detection model predicts that liquid fuel leakage occurs in the pipeline, sending the corresponding global image and local image to a terminal user; the acquiring of the global image and the local image of the target pipeline to be detected comprises: acquiring the global image and the local image of the target pipeline to be detected at a preset frequency; respectively judging whether the current acquired global image and local image are different from the previous corresponding global image and local image; in the case that the global image or the local image has differences between the previous and the current images, respectively storing the corresponding global image and local image in a memory; using the global image and the local image stored in the memory as the input of the pipeline detection model; the judging whether the current acquired global image and local image are different from the previous corresponding global image and local image comprises: starting an optical sensor to capture the detection object at a frame rate of 25 frames per second, and acquiring the captured optical image from a cache; The peak signal-to-noise ratio is used as a method for determining the difference between adjacent optical images, and the captured optical images are analyzed for difference: ; wherein, MSE is the mean square error, and M is the maximum value of the gray scale of the denoised image. ； wherein, and are the corresponding gray or color values in the image before compression and after compression, respectively, and m-1 and n-1 are the m x n pixel points.

2. The method of detecting a leak of a liquid fuel in a pipeline according to claim 1, wherein, the using of the global image and the local image stored in the memory as the input of the pipeline detection model comprises: respectively judging whether the total number of the global image and the local image with differences exceeds a preset number, and if yes, extracting the global image and the local image exceeding the preset number from the memory and using them as the input of the pipeline detection model.

3. The method of detecting a leak of a liquid fuel in a pipeline according to claim 2, wherein, in the case that the global image or the local image has no differences between the previous and the current images, deleting the images with no differences.

4. The method of detecting a leak of a liquid fuel in a pipeline according to claim 3, wherein, in the case that the pipeline detection model predicts that liquid fuel leakage does not occur in the pipeline, deleting the global image and the local image input into the pipeline detection model from the memory.

5. The method of detecting a leak of a liquid fuel in a pipeline according to claim 3, wherein, in the case that the global image or the local image has differences between the previous and the current images, respectively storing the corresponding global image and local image in a corresponding memory comprises: respectively establishing a corresponding subdirectory folder according to the year, month, day and hour of the acquired image, and storing the global image and the local image in the corresponding subdirectory folder according to the time of the acquired image.

6. An apparatus for detecting a leak of a liquid fuel in a pipe, the apparatus comprising: The method comprises the following steps: an acquiring unit is configured to acquire a global image and a local image of a target pipeline at the same time, respectively; the global image has a global field of view of the target pipeline; the local image has an image of a key part of the target pipeline; a training unit is configured to establish a neural network pipeline detection model based on optical flow according to the global image and the local image; the acquiring unit is further configured to acquire a global image and a local image of a target pipeline to be detected; The prediction unit is configured to input the global image and the local image of the target pipeline to be detected into the pipeline detection model, and predict whether the pipeline leaks liquid fuel by using the pipeline detection model. The sending unit is configured to send the corresponding global image and local image to a terminal when the pipeline detection model predicts that the pipeline leaks liquid fuel. The acquisition unit is further configured to: acquire the global image and the local image of the target pipeline to be detected at a preset frequency; determine whether the current acquired global image and local image are different from the previous corresponding global image and local image, respectively; store the corresponding global image and local image in the memory when the previous and current global image or local image are different; use the global image and the local image stored in the memory as the input of the pipeline detection model; The determination of whether the current acquired global image and local image are different from the previous corresponding global image and local image, respectively, includes: starting the optical sensor to capture the detection object at a frame rate of 25 frames per second, and obtaining the captured optical image from the cache; The peak signal-to-noise ratio is used as a method for determining the difference between adjacent optical images, and the captured optical images are analyzed for difference: ; wherein, MSE is the mean square error, and M is the maximum value of the grayscale of the denoised image. ； wherein, and are the corresponding gray or color values in the image before compression and after compression, respectively, and m-1 and n-1 are the m x n pixel points.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the method for detecting liquid fuel leakage in a pipeline according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method for detecting liquid fuel leakage in a pipeline according to any one of claims 1 to 5.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the method for detecting liquid fuel leakage in a pipeline according to any one of claims 1 to 5.

Images