Underground pipe intelligent desilting method, device and equipment based on deep learning and medium

Abstract

The invention relates to the technical field of farmland drainage underground pipe cleaning, in particular to an intelligent underground pipe dredging method, device and equipment based on deep learning and a medium, and the method comprises the steps: collecting a plurality of in-pipe images in an underground pipe; using a preset deep learning model to analyze the plurality of in-pipe images so as to determine a target blockage position in the concealed pipe and a blockage degree corresponding to the target blockage position; and according to the target blockage position and the blockage degree, a target relation curve of the target blockage position and the cleaning water pressure is established, and the cleaning water pressure corresponding to the target relation curve is used for conducting intelligent dredging on the target blockage position in the concealed conduit. Therefore, the problems that time and labor are wasted through manual cleaning, imported machines are high in cost and large in investment, rigid water pressure setting is often adopted, water consumption is large, utilization efficiency is low, operation complexity is increased, and applicability is low in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the technical field of cleaning concealed drainage pipes in farmland, and in particular to a method, device, equipment and medium for intelligent dredging of concealed pipes based on deep learning. Background Art

[0002] Concealed pipe drainage is a method of farmland drainage. It is widely used because of its advantages such as small footprint, easy mechanized construction and later maintenance management. Farmland concealed pipes are buried underground. As the years of operation of concealed pipes increase, fine particles in the soil and fine sand in the filter material often enter the pipe system with water flow. The accumulation of these impurities leads to gradual blockage of the pipes, which seriously affects the drainage efficiency of the system and the drainage speed of the farmland. Therefore, it is necessary to regularly evaluate the blockage of concealed pipe projects and flush the blocked concealed pipes. At present, there is a lack of effective methods for locating concealed pipe suction pipes, especially when these pipes are directly connected to underground water collection pipes or their outlets are covered by obstacles such as weeds. In this case, relying solely on manual search is not only inefficient, but also often cannot accurately locate.

[0003] Since the United States began to study pipeline desilting technology in the 1960s, Italy, Germany, Canada and other countries have also conducted in-depth exploration and development. Representative technologies include France's hydraulic balancing valve and Germany and Switzerland's hydraulic self-purification systems, which mainly use pressure differences and water flow power to flush sediments in the pipeline. Germany has also developed a self-flushing device based on the siphon principle to improve the degree of automation and safety. The above cleaning methods are all automatic flushing modes when the pipeline is operating normally, but after a long time, a large amount of silt will still accumulate in the pipeline. Although traditional desilting methods such as high-pressure water flushing and vacuum suction are effective, they are costly and have a greater impact on the environment.

[0004] Although the pipeline dredging technology abroad is relatively mature, the technology development in this field in China is still in its early stages. The domestic method of pipeline dredging mainly relies on manual cleaning and imported foreign machines. However, manual cleaning in related technologies is time-consuming and labor-intensive, and imported machines are costly and require large investments. They often use rigid water pressure settings, consume a lot of water, and have low utilization efficiency. In addition, the huge device also increases the complexity of operation and has low applicability, which needs to be solved urgently. Summary of the invention

[0005] The present application provides a method, device, equipment and medium for intelligent dredging of concealed pipes based on deep learning, so as to solve the problems in related technologies that manual dredging is time-consuming and labor-intensive, imported machines are costly and require large investments, and rigid water pressure settings are often used, resulting in high water consumption and low utilization efficiency. In addition, the huge equipment also increases the complexity of operation.

[0006] The first aspect of the present application provides a method for intelligent dredging of concealed pipes based on deep learning, comprising the following steps: collecting multiple images inside the concealed pipe; analyzing the multiple images inside the pipe using a preset deep learning model to determine a target blockage position in the concealed pipe and a blockage degree corresponding to the target blockage position; establishing a target relationship curve between the target blockage position and the cleaning water pressure according to the target blockage position and the blockage degree, so as to perform intelligent dredging on the target blockage position in the concealed pipe using the cleaning water pressure corresponding to the target relationship curve.

[0007] Optionally, in one embodiment of the present application, the multiple in-pipe images are analyzed based on a preset deep learning model to determine the target blockage position in the dark pipe and the degree of blockage corresponding to the target blockage position, including: based on the preset deep learning model, segmenting the dark pipe pipeline background and the blocked soil strips in the multiple in-pipe images to obtain segmented images; calculating the pixel length of the continuous soil strips in the segmented images, and using image resolution to convert the pixel length into the actual length of the blocked soil strip; determining the target blockage position in the dark pipe and the degree of blockage corresponding to the target blockage position according to the actual length.

[0008] Optionally, in one embodiment of the present application, the method of converting the pixel length into the actual length of the blocked soil strip using the image resolution includes: converting the segmented image into a binary image, and analyzing the soil existence status and length of each row in the binary image; calculating the average pixel length of the blocked soil strip using the soil existence status and length; and determining the actual length of the blocked soil strip based on the average pixel length.

[0009] Optionally, in one embodiment of the present application, establishing a target relationship curve between the target blockage position and the cleaning water pressure based on the target blockage position and the blockage degree includes: geometrically converting the actual length of the blockage soil strip to obtain the volume of the blockage soil strip; determining the mass of the blockage soil strip using the volume of the blockage soil strip, and determining a target critical flow rate based on the mass of the blockage soil strip and the target fluid mass; determining a target jet velocity based on the target critical flow rate, and generating the cleaning water pressure of the blockage soil strip based on the target jet velocity to establish a target relationship curve between the target blockage position and the cleaning water pressure.

[0010] Optionally, in one embodiment of the present application, after the target blockage position in the concealed pipe is intelligently dredged using the cleaning water pressure corresponding to the target relationship curve, it also includes: detecting whether the concealed pipe meets a preset exit dredging condition; when it is detected that the concealed pipe meets the preset exit dredging condition, recycling the water used for dredging in the concealed pipe.

[0011] The second aspect of the present application provides a deep learning-based intelligent dredging device for concealed pipes, comprising: an acquisition module for acquiring multiple images inside the concealed pipe; a determination module for analyzing the multiple images inside the concealed pipe using a preset deep learning model to determine a target blockage position in the concealed pipe and a blockage degree corresponding to the target blockage position; and a control module for establishing a target relationship curve between the target blockage position and the cleaning water pressure according to the target blockage position and the blockage degree, so as to perform intelligent dredging on the target blockage position in the concealed pipe using the cleaning water pressure corresponding to the target relationship curve.

[0012] Optionally, in one embodiment of the present application, the determination module includes: a segmentation unit, used to segment the dark pipe background and the blocked soil strips in the multiple images inside the pipe based on the preset deep learning model to obtain a segmented image; a calculation unit, used to calculate the pixel length of the continuous soil strips in the segmented image, and convert the pixel length into the actual length of the blocked soil strip using the image resolution; a first determination unit, used to determine the target blockage position in the dark pipe and the degree of blockage corresponding to the target blockage position based on the actual length.

[0013] Optionally, in one embodiment of the present application, the calculation unit includes: a conversion subunit, used to convert the segmented image into a binary image, and analyze the soil existence state and length of each row in the binary image; a calculation subunit, used to calculate the pixel length average of the blocked soil strip using the soil existence state and length; and a determination subunit, used to determine the actual length of the blocked soil strip based on the pixel length average.

[0014] Optionally, in one embodiment of the present application, the control module includes: an acquisition unit, used to perform a geometric conversion on the actual length of the blocking soil strip to obtain the volume of the blocking soil strip; a second determination unit, used to determine the mass of the blocking soil strip using the volume of the blocking soil strip, and determine the target critical flow rate based on the mass of the blocking soil strip and the target fluid mass; an establishment unit, used to determine the target jet velocity based on the target critical flow rate, and generate the cleaning water pressure of the blocking soil strip based on the target jet velocity, so as to establish a target relationship curve between the target blockage position and the cleaning water pressure.

[0015] Optionally, in one embodiment of the present application, the device of the embodiment of the present application also includes: a detection module, which is used to detect whether the concealed pipe meets a preset exit dredging condition after the target blockage position in the concealed pipe is intelligently dredged using the cleaning water pressure corresponding to the target relationship curve; and a processing module, which is used to recycle the water used for dredging in the concealed pipe if it is detected that the concealed pipe meets the preset exit dredging condition after the target blockage position in the concealed pipe is intelligently dredged using the cleaning water pressure corresponding to the target relationship curve.

[0016] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for intelligent dredging of concealed pipes based on deep learning as described in the above embodiment.

[0017] The fourth aspect embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and when the program is executed by a processor, it implements the above-mentioned method for intelligent dredging of concealed pipes based on deep learning.

[0018] The fifth aspect of the present application provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for intelligent dredging of concealed pipes based on deep learning.

[0019] The embodiment of the present application can use a deep learning model to analyze multiple images of the inside of the dark pipe, so as to determine the target blockage position in the dark pipe and the blockage degree corresponding to the target blockage position, and then establish a target relationship curve between the target blockage position and the cleaning water pressure, so as to use the cleaning water pressure corresponding to the target relationship curve to perform intelligent dredging on the target blockage position in the dark pipe, effectively reducing labor costs and reducing the complexity of pipeline dredging operations. Thus, the problems in the related art that manual dredging is time-consuming and labor-intensive, imported machines are costly and require large investments, and rigid water pressure settings are often used, resulting in high water consumption and low utilization efficiency, and the huge device also increases the complexity of the operation and has low applicability are solved.

[0020] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0022] Figure 1 A schematic diagram of a concealed pipe intelligent dredging system based on deep learning according to an embodiment of the present application;

[0023] Figure 2 A flowchart of a method for intelligent dredging of concealed pipes based on deep learning according to an embodiment of the present application;

[0024] Figure 3 This is a schematic diagram of an in-tube image before and after segmentation according to a specific embodiment of the present application;

[0025] Figure 4 A schematic diagram of a relationship curve between a concealed pipe blockage degree identification result and a corresponding output water pressure result in a specific embodiment of the present application;

[0026] Figure 5 A schematic diagram of the structure of a concealed pipe intelligent dredging device based on deep learning according to an embodiment of the present application;

[0027] Figure 6 It is a schematic diagram of the structure of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0029] The following describes the method, device, equipment and medium of intelligent dredging of concealed pipes based on deep learning in the embodiment of the present application with reference to the accompanying drawings. In view of the above-mentioned problems in the related art mentioned in the background technology center, such as time-consuming and labor-intensive manual cleaning, high cost and large investment of imported machines, and often using rigid water pressure settings, large water consumption, low utilization efficiency, and huge devices also increase the complexity of operation, and low applicability, the present application provides a method of intelligent dredging of concealed pipes based on deep learning, in which a deep learning model can be used to analyze multiple images in the concealed pipe, so as to determine the target blocking position in the concealed pipe and the blocking degree corresponding to the target blocking position, and then establish a target relationship curve between the target blocking position and the cleaning water pressure, so as to use the cleaning water pressure corresponding to the target relationship curve to perform intelligent dredging on the target blocking position in the concealed pipe, effectively reducing labor costs and reducing the complexity of pipeline dredging operations. Thus, the problem that time-consuming and labor-intensive manual cleaning, high cost and large investment of imported machines, and often using rigid water pressure settings, large water consumption, low utilization efficiency, and huge devices also increase the complexity of operation, and low applicability in the related art is solved.

[0030] This application establishes a deep learning-based intelligent dredging system for underground pipes. Figure 1As shown, the system includes: a rotating wheel 1, a computer 2, a telescopic rod 3, a flushing nozzle 4, a micro camera 5, a variable frequency water pump 7, a water tank 8 and a filter return port 9. Among them, the rotating wheel 1 is connected to the variable frequency water pump 7 and the telescopic rod 3, driving the entire device to move forward or backward in the dark pipe; the telescopic rod 3 is used to support the flushing nozzle 4 and the micro camera 5, and control the forward and backward movement; the micro camera 5 is fixed on the telescopic rod and is responsible for capturing the siltation at the bottom of the dark pipe; the variable frequency water pump 7 controls the water flow pressure; the water tank 8 is used to store cleaning water; the filter return port 9 allows the used water to flow back to the water tank for reuse; the computer 2 is responsible for processing the image sent back by the camera and adjusting the water pump pressure based on the image analysis results. The system can accurately identify the volume of sediment in the dark pipe, intelligently adjust different water pressures according to different blockage conditions in the pipeline, and recycle clean water to save water and energy.

[0031] Specifically, Figure 2 A schematic flow chart of a method for intelligent dredging of concealed pipes based on deep learning provided in an embodiment of the present application.

[0032] like Figure 2 As shown, the deep learning-based intelligent dredging method for concealed pipes includes the following steps:

[0033] In step S201, a plurality of in-tube images in a dark tube are collected.

[0034] It can be understood that the embodiments of the present application can collect multiple images inside the dark pipe. For example, the embodiments of the present application can drive the water spraying body into the dark pipe by the rotating wheel in the above system, and take the images inside the pipe at regular intervals through the micro camera. For example, it can be set to automatically take a picture every 1 second, and the micro camera is pointed vertically downward to ensure complete coverage of the blocked area, so as to transmit the captured images inside the pipe to the software control end in real time until the end of the dark pipe is reached, thereby effectively improving the accuracy of obtaining the blocked area in the pipe.

[0035] In step S202, a preset deep learning model is used to analyze multiple in-tube images to determine the target blockage position in the dark tube and the blockage degree corresponding to the target blockage position.

[0036] It can be understood that the embodiments of the present application can utilize the deep learning model in the following steps, such as using the DeepLabV3+ image semantic segmentation model to analyze multiple images inside the pipe to determine the target blockage position in the blind pipe and the degree of blockage corresponding to the target blockage position, thereby effectively improving the accuracy of blind pipe dredging.

[0037] Among them, in one embodiment of the present application, based on a preset deep learning model, multiple in-pipe images are analyzed to determine the target blockage position in the dark pipe and the degree of blockage corresponding to the target blockage position, including: based on the preset deep learning model, segmenting the dark pipe pipeline background and the blocked soil strips in the multiple in-pipe images to obtain segmented images; calculating the pixel length of the continuous soil strips in the segmented images, and using the image resolution to convert the pixel length into the actual length of the blocked soil strip; determining the target blockage position in the dark pipe and the degree of blockage corresponding to the target blockage position according to the actual length.

[0038] In the actual implementation process, Figure 3 As shown, the embodiment of the present application can firstly separate the concealed pipe background and the blocked soil strips from the acquired in-pipe image based on image segmentation technology, and then use the DeepLabV3+ image semantic segmentation model to build and train a deep learning network with high recognition accuracy, and use the deep learning network to further separate and identify the initially separated concealed pipe background and the blocked soil strips, so as to obtain the segmented image; then, the image is processed based on the deep learning results and prior knowledge, and the run-length theory is applied to analyze the semantically segmented image, and the pixel length of the continuous soil strips in the image is calculated, and the pixel length is converted into the actual length of the blocked soil strip in combination with the image resolution, and then the target blockage position in the concealed pipe and the degree of blockage corresponding to the target blockage position are determined according to the actual length, thereby effectively improving the accuracy of concealed pipe dredging.

[0039] It should be noted that "blocked soil strips" are defined as blockages in the pipeline formed by a mixture of mud, sand, and other impurities; the specific application of the run theory is to calculate the pixel length of continuous soil strips, and then convert it into actual length through known dimensions; the training and verification of the DeepLabV3+ model are carried out on a self-built dataset containing at least 600 images to ensure the generalization ability of the model and the adaptability of the operating environment. The input image size of the DeepLabV3+ model is 1920x1080 pixels, and the output image is a segmentation map of the soil strips and the background. The training uses a labeled dataset, which contains a total of 3 different types of blockage situations.

[0040] Optionally, in one embodiment of the present application, the pixel length is converted into the actual length of the blocked soil strip using the image resolution, including: converting the segmented image into a binary image, and analyzing the soil existence status and length of each row in the binary image; calculating the average pixel length of the blocked soil strip using the soil existence status and length; and determining the actual length of the blocked soil strip based on the average pixel length.

[0041] In some embodiments, the embodiments of the present application can convert the segmented image, that is, the separated image after deep learning, into a binary image, remove foreground noise and enhance the main image features, then mark and analyze the connected foreground objects, calculate the regional attributes and the vertical distance between adjacent objects, and obtain the average distance; calculate the soil strip spacing in the image by the ratio of the average distance to the image boundary using the known bellows spacing; after the binarization processing, use morphological operations such as dilation and erosion to accurately identify and isolate the edges of the soil strips, where 0 represents no soil and 1 represents the presence of soil, analyze the presence and length of soil in each row in the binary image, further calculate the average pixel length of the soil strips, calculate the actual soil strip length based on the ratio of the pixel length of the soil strips to the total image size, and then obtain the siltation thickness through geometric transformation, calculate the volume of the silt in the corresponding part, and effectively improve the accuracy of blind pipe dredging.

[0042] In step S203, a target relationship curve between the target blockage position and the cleaning water pressure is established according to the target blockage position and the blockage degree, so as to perform intelligent dredging on the target blockage position in the concealed pipe using the cleaning water pressure corresponding to the target relationship curve.

[0043] It is understandable that the embodiment of the present application can determine the blockage conditions of different areas in the dark pipe according to the target blockage position and the blockage degree, so as to establish a target relationship curve between the target blockage position and the cleaning water pressure, that is, Figure 4 The relationship curve between the concealed pipe distance and the water pressure shown is used to intelligently desilt the target blockage position in the concealed pipe by utilizing the cleaning water pressure corresponding to the target relationship curve. For example, the computer in the above system can send a cleaning instruction to the hardware end to prepare for cleaning. Then, the wheel in the above system is restarted to drive the water spray body back to the concealed pipe entrance along the original route. At the same time, according to the water pressure curve calculated previously, the variable frequency water pump is started to perform graded cleaning according to the set water pressure, which effectively reduces the complexity of the pipeline desilting operation and improves the automation level of concealed pipe desilting.

[0044] Optionally, in one embodiment of the present application, a target relationship curve between the target blockage position and the cleaning water pressure is established according to the target blockage position and the blockage degree, including: geometrically converting the actual length of the blockage soil bar to obtain the volume of the blockage soil bar; determining the mass of the blockage soil bar using the volume of the blockage soil bar, and determining the target critical flow rate according to the mass of the blockage soil bar and the target fluid mass; determining the target jet velocity according to the target critical flow velocity, and generating the cleaning water pressure of the blockage soil bar according to the target jet velocity, so as to establish a target relationship curve between the target blockage position and the cleaning water pressure.

[0045] As a possible implementation method, the embodiment of the present application can convert the actual length of the plugging soil strip into a geometric relationship to obtain the volume of the plugging soil strip. Based on the volume of the plugging soil strip, different volumes are selected in combination with the soil conditions in different regions to calculate the mass of the sediment in the plugging soil strip. Then, the critical flow rate is calculated by using the BC Knorrhodz formula in combination with the fluid mass, that is:

[0046]

[0047] Among them, v L is the critical flow velocity of the transported fluid, m / s; D L is the critical pipe diameter, and the diameter of the raised circle of the concealed pipe is taken as the critical pipe diameter, m; β is the sediment density correction coefficient, ρ w is the density of water.

[0048] Among them, when the sediment density ρ0≤2.7t / m 3 When β=1; when sediment density ρ0>2.7t / m 3 When , the β value is determined by the following formula: cp ≤1.5mm, d cp >1.5mm, δ t is the density of sediment ρ0 and the density of water ρ w ratio;

[0049] Density of water:

[0050]

[0051] Among them, M s is the mass of sediment, kg; M w is the mass of the fluid, kg;

[0052] Sediment quality:

[0053] M s =ρ0V s

[0054] Where ρ0 is the sediment density, V s is the volume of sediment, m 3 .

[0055] Furthermore, the embodiment of the present application can apply the flow continuity equation to the nozzle inlet and outlet interfaces to calculate the jet velocity that can flush away the silt as completely as possible, and combine it with the water pump parameters to obtain the cleaning working pressure corresponding to different siltation degrees and the final cleaning pump power. In addition, a target relationship curve between the target blockage position and the cleaning water pressure can be established, so that the water pressure can be dynamically adjusted according to the specific blockage degree of the pipeline, reducing water and energy consumption, and improving the efficiency of dredging operations and the overall utilization of water resources.

[0056] Optionally, in one embodiment of the present application, after the target blockage position in the blind pipe is intelligently dredged using the cleaning water pressure corresponding to the target relationship curve, it also includes: detecting whether the blind pipe meets the preset exit dredging conditions; when it is detected that the blind pipe meets the preset exit dredging conditions, the water used for dredging in the blind pipe is recycled.

[0057] In some embodiments, the embodiments of the present application can detect whether the concealed pipe meets the conditions for exiting dredging, such as whether all the silt in the concealed pipe has been cleared. When it is detected that the concealed pipe meets the conditions for exiting dredging, the water in the concealed pipe that has been dredged will be recycled, for example, the used water will be filtered back into the water tank through the filter port for recycling, so as to save resources and reduce environmental impact, and improve the applicability of concealed pipe dredging.

[0058] For example, now combined with Figure 1 The siltation status of the underground pipes in the saline-alkali land of a certain area was simulated in the underground pipe model with a length of 1m and a diameter of 15cm. The specific steps are as follows:

[0059] S1. Equipment startup and concealed pipe entry.

[0060] When started, the speed of the wheel 1 can be adjusted between 0.5 and 2.0 m / s to ensure the device moves forward steadily and reduce damage to the pipe wall; the micro camera 5 has a resolution of 1920x1080 pixels and is installed on the telescopic rod 3. The viewing angle is set to vertically downward and the shooting frequency is set to one frame per second to capture continuous images of the inside of the dark pipe. The camera has automatic focus and light compensation functions to ensure image quality; the images taken by the camera are instantly uploaded to the computer 2 software control terminal through the connection and processed and analyzed by matlab.

[0061] S2. Image processing and sedimentation analysis.

[0062] Computer 2 applies image processing techniques such as filtering and contrast enhancement to the received image to improve the clarity and recognizability of the image, and then uses semantic segmentation to segment the background of the concealed pipe and the blocked soil strips, establishes a soil-pipe separation data set, and builds a DeepLabV3+ image semantic segmentation model based on the self-established data set to perform network training. The DeepLabV3+ image semantic segmentation model can distinguish the background of the concealed pipe from the soil strips. This model is trained on a self-established data set containing at least 600 images to ensure good recognition of various blockage situations, and finally outputs a segmentation map of soil and non-soil; based on the run-length theory, the pixel length of the continuous soil strips in the segmented image is calculated. Through the known image resolution and pipeline size, the pixel length is converted to the actual length, and the volume of the blockage is calculated in combination with the size of the concealed pipe; based on the length and volume of the soil strips, the BC Knorrhodz formula is used in combination with fluid dynamics to calculate the required critical non-silting flow rate and determine the required water pressure, thereby providing precise guidance for the pressure output of the variable frequency water pump.

[0063] S3. Setting and executing cleaning parameters.

[0064] According to the analysis of the above steps, the variable frequency water pump 7 adjusts the output water pressure, and the range is divided into 3 flushing pressures according to 0.1 to 5.5 MPa to meet the cleaning needs of different degrees of siltation; after completing the identification of silt, the equipment operates in the reverse direction along the original route and performs high-pressure water cleaning. During the cleaning process, the flushing nozzle 4 automatically adjusts the water pressure according to the variable frequency water pump to produce different spraying intensities and coverage areas to ensure that the silt is effectively removed.

[0065] S4. Recycling and reuse of water resources.

[0066] After cleaning, the used water flows back to the water tank 8 through the filter return port 9. The filtration system can remove solid impurities in the water to ensure that the water quality meets the standards for reuse. The cleaned water is pumped into the system again for the next round of cleaning, effectively reducing the consumption of water resources and the impact on the environment.

[0067] Therefore, the embodiments of the present application can intuitively and accurately quantify the volume of silt in the concealed pipe, enhance the applicability of dredging technology in modern agriculture, demonstrate higher efficiency and accuracy in dealing with complex and changeable blockage materials, and intelligently adjust different water pressures according to different blockage conditions in the concealed pipe, and recycle clean water, saving water and energy, thereby improving the efficiency of dredging operations and the overall utilization of water resources.

[0068] According to the method for intelligent desilting of concealed pipes based on deep learning proposed in the embodiment of the present application, a deep learning model can be used to analyze multiple images in the concealed pipe, so as to determine the target blockage position in the concealed pipe and the blockage degree corresponding to the target blockage position, and then establish a target relationship curve between the target blockage position and the cleaning water pressure, so as to use the cleaning water pressure corresponding to the target relationship curve to perform intelligent desilting on the target blockage position in the concealed pipe, effectively reducing labor costs and reducing the complexity of pipeline desilting operations. Thus, the problems in the related art that manual desilting is time-consuming and labor-intensive, imported machines are costly and require large investments, and rigid water pressure settings are often used, resulting in large water consumption and low utilization efficiency, and the huge device also increases the complexity of the operation and has low applicability are solved.

[0069] Secondly, the intelligent dredging device for concealed pipes based on deep learning proposed in accordance with the embodiments of the present application is described with reference to the accompanying drawings.

[0070] Figure 5 It is a block diagram of a concealed pipe intelligent dredging device based on deep learning in an embodiment of the present application.

[0071] like Figure 5 As shown, the deep learning-based intelligent dredging device 10 for concealed pipes includes: a collection module 100, a determination module 200 and a control module 300.

[0072] Specifically, the acquisition module 100 is used to acquire multiple in-tube images in a dark tube.

[0073] The determination module 200 is used to analyze multiple in-tube images using a preset deep learning model to determine the target blockage position in the dark tube and the blockage degree corresponding to the target blockage position.

[0074] The control module 300 is used to establish a target relationship curve between the target blockage position and the cleaning water pressure according to the target blockage position and the blockage degree, so as to use the cleaning water pressure corresponding to the target relationship curve to intelligently desilt the target blockage position in the concealed pipe.

[0075] Optionally, in one embodiment of the present application, the determination module 200 includes: a segmentation unit, a calculation unit and a first determination unit.

[0076] Among them, the segmentation unit is used to segment the dark pipe background and the blocked soil strips in multiple images of the pipe based on a preset deep learning model to obtain the segmented image.

[0077] The calculation unit is used to calculate the pixel length of the continuous soil strip in the segmented image and convert the pixel length into the actual length of the blocked soil strip by using the image resolution.

[0078] The first determination unit is used to determine a target blocking position in the concealed pipe and a blocking degree corresponding to the target blocking position according to the actual length.

[0079] Optionally, in one embodiment of the present application, the calculation unit includes: a conversion subunit, a calculation subunit and a determination subunit.

[0080] The conversion subunit is used to convert the segmented image into a binary image and analyze the soil existence status and length of each row in the binary image.

[0081] The calculation subunit is used to calculate the average pixel length of the blocked soil strip by using the soil existence state and length.

[0082] A subunit is determined for determining the actual length of the plugged soil strip based on the average pixel length.

[0083] Optionally, in one embodiment of the present application, the control module 300 includes: an acquisition unit, a second determination unit and an establishment unit.

[0084] The acquisition unit is used to convert the actual length of the plugging soil strip into a geometric relationship to obtain the volume of the plugging soil strip.

[0085] The second determining unit is used to determine the mass of the plugging soil strip by using the volume of the plugging soil strip, and determine the target critical flow rate according to the mass of the plugging soil strip and the mass of the target fluid.

[0086] A unit is established for determining a target jet velocity according to a target critical flow velocity, and generating a cleaning water pressure of the plugged soil strip according to the target jet velocity, so as to establish a target relationship curve between a target plugging position and a cleaning water pressure.

[0087] Optionally, in one embodiment of the present application, the device 10 of the embodiment of the present application further includes: a detection module and a processing module.

[0088] Among them, the detection module is used to detect whether the hidden pipe meets the preset exit dredging conditions after the target blockage position in the hidden pipe is intelligently dredged using the cleaning water pressure corresponding to the target relationship curve.

[0089] The processing module is used to intelligently desilt the target blockage position in the concealed pipe using the cleaning water pressure corresponding to the target relationship curve, and then recycle the water used for desilting in the concealed pipe when it is detected that the concealed pipe meets the preset desilting exit conditions.

[0090] It should be noted that the aforementioned explanation of the embodiment of the method for intelligent dredging of concealed pipes based on deep learning is also applicable to the intelligent dredging device for concealed pipes based on deep learning in this embodiment, and will not be repeated here.

[0091] According to the deep learning-based intelligent desilting device for concealed pipes proposed in the embodiment of the present application, a deep learning model can be used to analyze multiple images in the concealed pipe, thereby determining the target blockage position in the concealed pipe and the blockage degree corresponding to the target blockage position, and then establishing a target relationship curve between the target blockage position and the cleaning water pressure, so as to use the cleaning water pressure corresponding to the target relationship curve to perform intelligent desilting on the target blockage position in the concealed pipe, effectively reducing labor costs and reducing the complexity of pipeline desilting operations. Thus, the problems in the related art that manual desilting is time-consuming and labor-intensive, imported machines are costly and require large investments, and rigid water pressure settings are often used, resulting in large water consumption and low utilization efficiency, and the huge device also increases the complexity of the operation and has low applicability are solved.

[0092] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device may include:

[0093] A memory 601 , a processor 602 , and a computer program stored in the memory 601 and executable on the processor 602 .

[0094] When the processor 602 executes the program, the dark pipe intelligent dredging method based on deep learning provided in the above embodiment is implemented.

[0095] Furthermore, the electronic device further comprises:

[0096] The communication interface 603 is used for communication between the memory 601 and the processor 602 .

[0097] The memory 601 is used to store computer programs that can be executed on the processor 602 .

[0098] The memory 601 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.

[0099] If the memory 601, the processor 602 and the communication interface 603 are implemented independently, the communication interface 603, the memory 601 and the processor 602 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0100] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can communicate with each other through an internal interface.

[0101] The processor 602 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0102] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned method for intelligent dredging of concealed pipes based on deep learning is implemented.

[0103] This embodiment also provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for intelligent dredging of concealed pipes based on deep learning.

[0104] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0105] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0106] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0107] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0108] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0109] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0110] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0111] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for intelligent dredging of concealed pipes based on deep learning, characterized in that: The following steps are involved: Collect multiple images of the dark tube; Analyzing the plurality of in-tube images using a preset deep learning model to determine a target blockage position in the dark tube and a blockage degree corresponding to the target blockage position; A target relationship curve between the target blockage position and the cleaning water pressure is established according to the target blockage position and the blockage degree, so as to intelligently desilt the target blockage position in the concealed pipe using the cleaning water pressure corresponding to the target relationship curve.

2. The method according to claim 1, characterized in that The method of analyzing the plurality of in-tube images based on a preset deep learning model to determine a target blockage position in the dark tube and a blockage degree corresponding to the target blockage position includes: Based on the preset deep learning model, segment the dark pipe background and the blocked soil strips in the multiple in-pipe images to obtain segmented images; Calculating the pixel length of the continuous soil strip in the segmented image, and converting the pixel length into the actual length of the blocked soil strip using the image resolution; A target blocking position in the blind pipe and a blocking degree corresponding to the target blocking position are determined according to the actual length.

3. The method according to claim 2, characterized in that The method of converting the pixel length into the actual length of the plugging soil strip by using the image resolution comprises: Converting the segmented image into a binary image, and analyzing the soil existence state and length of each row in the binary image; Calculating the average pixel length of the blocking soil strip using the soil existence state and length; The actual length of the plugging soil strip is determined according to the average pixel length.

4. The method according to claim 2, characterized in that: The step of establishing a target relationship curve between the target blockage position and the cleaning water pressure according to the target blockage position and the blockage degree includes: Performing geometric transformation on the actual length of the plugging soil strip to obtain the volume of the plugging soil strip; Determining the mass of the plugging soil strip by using the volume of the plugging soil strip, and determining the target critical flow rate according to the mass of the plugging soil strip and the mass of the target fluid; The target jet velocity is determined according to the target critical flow velocity, and the cleaning water pressure of the plugging soil strip is generated according to the target jet velocity to establish a target relationship curve between the target plugging position and the cleaning water pressure.

5. The method according to claim 1, characterized in that After the target blocking position in the blind pipe is intelligently desilted using the cleaning water pressure corresponding to the target relationship curve, the method further includes: Detecting whether the concealed pipe meets the preset exit and dredging conditions; When it is detected that the blind pipe meets the preset conditions for exiting dredging, the water in the blind pipe that has been dredged is recycled.

6. A blind pipe intelligent dredging device based on deep learning, characterized in that: include: An acquisition module, used for acquiring multiple images inside the dark tube; A determination module, configured to analyze the plurality of in-tube images using a preset deep learning model to determine a target blockage position in the dark tube and a blockage degree corresponding to the target blockage position; The control module is used to establish a target relationship curve between the target blockage position and the cleaning water pressure according to the target blockage position and the blockage degree, so as to use the cleaning water pressure corresponding to the target relationship curve to intelligently desilt the target blockage position in the concealed pipe.

7. The device according to claim 6, characterized in that The determination module comprises: A segmentation unit, configured to segment the dark pipe background and the blocked soil strips in the plurality of in-pipe images based on the preset deep learning model to obtain segmented images; A calculation unit, used for calculating the pixel length of the continuous soil strip in the segmented image, and converting the pixel length into the actual length of the blocked soil strip by using the image resolution; A determination unit is used to determine a target blocking position in the blind pipe and a blocking degree corresponding to the target blocking position according to the actual length.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for intelligent dredging of concealed pipes based on deep learning as described in any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the deep learning-based intelligent dredging method for concealed pipes as described in any one of claims 1 to 5.

10. A computer program product, comprising a computer program, characterized in that The computer program is executed by a processor to implement the deep learning-based intelligent dredging method for concealed pipes as described in any one of claims 1 to 5.

Images