A pipeline inspection and cleaning robot and control system

Abstract

The present invention relates to the field of robot control technology, and in particular to a pipeline detection cleaning robot and control system. First, a grayscale image of the inside of a pipeline is obtained and regional division is performed. Image segmentation technology is used to accurately identify dirty areas in the images of each region, and the standard spraying speed of the detergent is adjusted based on the area and number of the dirty areas to achieve accurate matching with the dirty areas. At the same time, by analyzing the grayscale values ​​of the pixels in the grayscale image, the degree of dirtiness is quantified, and then the unit area residence time of the cleaning robot in each dirty area is determined. Finally, the pipeline detection cleaning robot is used to clean the pipeline in combination with the detergent spraying speed and residence time. This method achieves precise dosage control of the detergent and differentiated cleaning strategies by accurately adjusting the spraying speed of the detergent and the residence time of the robot, effectively improving the cleaning effect.

Description

Technical Field

[0001] The present invention relates to the technical field of robot control, and in particular to a pipeline detection and cleaning robot and a control system. Background Art

[0002] In many industrial and infrastructure fields, such as municipal drainage, petrochemicals, electricity, mining, etc., pipeline cleaning is a crucial link. Traditional pipeline cleaning mainly relies on manual entry into the pipeline for operation, but due to the small and complex environment inside the pipeline, manual operation will have a high safety risk. With the continuous advancement of robot technology, pipeline inspection and cleaning robots have emerged. Such robots can be used specifically to perform inspection and cleaning tasks inside pipelines. They can work in small, complex or dangerous environments, effectively improving the safety and efficiency of operations.

[0003] At present, when cleaning dirt inside the pipeline, the pipeline inspection and cleaning robot uses the camera on board to identify the pipeline environment, and then controls the robot in the cloud to spray cleaning agent on the dirty areas in the pipeline, so as to achieve all-round cleaning without dead ends. However, due to the different degrees of dirt in different areas of the pipeline, if the same amount of cleaning agent is sprayed on all locations, the targeting is poor, which may result in incomplete cleaning of some dirt inside the pipeline. Summary of the invention

[0004] In order to solve the technical problem that due to the different degrees of dirt in different areas inside the pipeline, if the same amount of cleaning agent is sprayed on all positions, the pertinence is poor, and some dirt inside the pipeline may not be cleaned thoroughly, the purpose of the present invention is to provide a pipeline inspection and cleaning robot and a control system, and the technical solutions adopted are as follows:

[0005] An image data acquisition module, used to acquire a grayscale image of the interior of the pipeline to be cleaned;

[0006] A detergent spray speed analysis module is used to divide the grayscale image into regions to obtain a plurality of region images, perform image segmentation in each region image to obtain a dirty region in each region image in the grayscale image, and adjust the detergent standard spray speed based on the area and number of the dirty regions in each region image to determine the detergent spray speed for each dirty region;

[0007] The residence time analysis module is used to analyze the grayscale values ​​of the pixels in the grayscale image, so as to obtain the dirtiness value of each dirty area in each area image, and in each area image, based on the area of ​​the dirty area and the dirtiness value, determine the residence time of the cleaning robot per unit area of ​​each dirty area;

[0008] The pipeline cleaning module is used to clean the inside of the pipeline to be cleaned by using a cleaning robot based on the spraying speed of the cleaning agent and the residence time per unit area in each dirty area.

[0009] Furthermore, the method for obtaining the dirty area includes:

[0010] In each regional image, threshold segmentation is performed on the regional image based on the OTSU threshold segmentation method to obtain a binary image;

[0011] Taking the area with a gray value of 0 in the binary image as the foreground area, and taking the area corresponding to the foreground area in the gray image as the target area;

[0012] In the target area, connected domain analysis and morphological closing operation are performed to obtain all connected domains, and each connected domain is regarded as a dirty area.

[0013] Furthermore, the method for determining the cleaning agent spraying speed includes:

[0014] In each region image, a cleaning importance factor of each region image is calculated based on the area and number of dirty regions;

[0015] Based on the differences between the important cleaning factors of all the regional images and in combination with the preset cleaning agent dosage, the total amount of cleaning agent for each regional image is determined;

[0016] The value after normalizing the ratio of the total amount of detergent in each regional image to the total area of ​​all dirty regions in each regional image is used as the detergent spray speed adjustment factor corresponding to the dirty region in each regional image;

[0017] In each area image, the detergent spraying speed of the dirty area in each area image is determined according to the corresponding detergent spraying speed adjustment factor and the preset spraying speed, and the detergent spraying speed adjustment factor and the preset spraying speed are both positively correlated with the detergent spraying speed.

[0018] Furthermore, the method for obtaining the cleaning important factor includes:

[0019] In each region image, a value obtained by normalizing the product of the area mean of all dirty regions and the number of dirty regions is used as the cleaning importance factor of each region image.

[0020] Furthermore, the method for obtaining the total amount of detergent in each area image includes:

[0021] The difference between the cleaning importance factor of each regional image and the minimum cleaning importance factor of all regional images is used as the first adjustment factor of the cleaning agent dosage corresponding to each regional image;

[0022] The difference between the preset maximum detergent dosage and the preset minimum detergent dosage is normalized to a value as a second detergent dosage adjustment factor;

[0023] The product of the first adjustment factor of the detergent dosage corresponding to each regional image and the second adjustment factor of the detergent dosage is used as the adjustment parameter corresponding to each regional image;

[0024] The sum of the minimum preset detergent dosage and the adjustment parameter corresponding to each regional image is used as the total detergent dosage of each regional image.

[0025] Furthermore, the method for obtaining the dirtiness value includes:

[0026] In each area image, the mean grayscale value of all pixels in each dirty area is calculated, and the mean grayscale value is negatively correlated and normalized to obtain a value that serves as the dirtiness value of each dirty area.

[0027] Furthermore, the method for obtaining the residence time includes:

[0028] In each region image, all dirty regions are sorted in ascending order according to their area to obtain a sorted sequence;

[0029] Determining a time factor based on the position of each dirty region in the corresponding sorted sequence;

[0030] The sum of the time factor, the dirtiness level and the preset stay time corresponding to each dirty area is used as the stay time of the cleaning robot per unit area in each dirty area.

[0031] Furthermore, the method for obtaining the time factor includes:

[0032] In the sorting sequence, the ratio of the sequence number of each dirty area to the median value of the sequence number in the sorting sequence is used as the time factor corresponding to each dirty area, wherein the sequence number value of the sorting sequence starts from 1.

[0033] Furthermore, in each dirty area, based on the spraying speed of the cleaning agent and the residence time per unit area, the cleaning robot is used to clean the inside of the pipeline to be cleaned, including:

[0034] When the cleaning robot moves to a dirty area, the residence time of the cleaning robot per unit area is set to the residence time per unit area of ​​the dirty area, and the detergent spraying speed of the cleaning robot is set to the detergent spraying speed of the dirty area, so that the cleaning robot is used to clean the dirty area in the pipeline.

[0035] A pipeline inspection and cleaning robot comprises a processor and a memory, wherein the memory stores at least one instruction, a program, a code set or an instruction set, and when the at least one instruction, a program, a code set or an instruction set is loaded and executed by the processor, steps of a pipeline inspection and cleaning robot control system are implemented.

[0036] The present invention has the following beneficial effects:

[0037] First, a grayscale image of the inside of the pipe to be cleaned is obtained to provide basic data for the subsequent identification of dirty areas inside the pipe. In order to more accurately identify the dirty areas inside the pipe, the grayscale image can be divided into regions to obtain multiple regional images, and then in each regional image, the dirty areas can be accurately identified based on image segmentation technology. Since the number and size of dirty areas can be used as a standard for evaluating the amount of detergent required, the standard detergent spraying speed can be adjusted based on the area and number of dirty areas to determine the detergent spraying speed of each dirty area. At this time, the detergent spraying speed can better match the dirty area. Furthermore, after determining the detergent spraying speed of each dirty area, it is necessary to determine the residence time at each position, so that the amount of detergent can be more accurately matched with the dirtiness of each position. Since the grayscale value of the dirty area is usually quite different from the grayscale value of the pipe wall, the grayscale value of the pixel in the grayscale image can be analyzed to obtain the degree of dirtiness of each dirty area in each area image. The degree of dirtiness is analyzed from the level of grayscale value to quantify the degree of dirtiness. Then, the residence time of the cleaning robot per unit area of ​​each dirty area can be determined according to the area of ​​the dirty area and the degree of dirtiness. By optimizing the residence time, it can be ensured that each position in the dirty area is fully cleaned. Finally, in each dirty area, the pipeline detection cleaning robot is used to clean each dirty position of the pipeline in combination with the corresponding detergent spraying speed and residence time. The present invention can accurately control the amount of detergent used by accurately adjusting the spraying speed of the detergent in each dirty area and the residence time of the robot, and implement differentiated cleaning strategies for different dirty areas, thereby effectively improving the cleaning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1A system block diagram of a pipeline inspection and cleaning robot control system provided by one embodiment of the present invention;

[0040] Figure 2 A method flow chart of a method for obtaining a cleaning agent spraying speed provided by an embodiment of the present invention;

[0041] Figure 3 A flow chart of a method for obtaining residence time provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a pipeline inspection and cleaning robot and control system proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The specific scheme of a pipeline inspection and cleaning robot and a control system provided by the present invention is described in detail below with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a system block diagram of a pipeline inspection and cleaning robot control system provided by an embodiment of the present invention, the system includes: an image data acquisition module 101, a cleaning agent spray speed analysis module 102, a residence time analysis module 103, and a pipeline cleaning module 104.

[0046] The image data acquisition module 101 is used to acquire a grayscale image of the interior of the pipeline to be cleaned.

[0047] Pipeline inspection and cleaning robots are robots specially designed to perform inspection and cleaning tasks inside pipelines. These robots can work in small, complex or dangerous environments, improving the safety and efficiency of operations. When cleaning dirt inside the pipeline, the pipeline cleaning robot does not need to enter the pipeline manually. The built-in controller analyzes and processes the collected images inside the pipeline to guide the robot to complete the cleaning work.

[0048] Install a high-resolution camera on the pipeline inspection and cleaning robot to ensure that the camera can clearly capture the image inside the pipeline. At the same time, since the internal environment of the pipeline is usually dim, it is necessary to start the built-in lighting system of the pipeline inspection and cleaning robot to provide sufficient light. The brightness and color temperature of the lighting system should be adjusted according to the actual situation to ensure that the image of the inside of the pipeline to be cleaned has a clear contrast and brightness distribution. In the process of shooting the image of the inside of the pipeline to be cleaned, the position, angle and focal length of the camera need to be debugged to ensure that the quality of the image obtained meets the requirements of subsequent processing.

[0049] Because the collected images are usually color images, in the internal environment of the pipeline, factors such as lighting and shadow may affect the identification of dirt in the color image, and the grayscale image is more conducive to the subsequent identification and feature extraction of dirty areas. Therefore, in order to facilitate subsequent processing and analysis, the acquired color image can be converted into a grayscale image, thereby obtaining a grayscale image of the inside of the pipeline to be cleaned.

[0050] It should be noted that the method for acquiring the grayscale image can adopt methods such as average grayscale, maximum value method or weighted grayscale, which are all technical means well known to technical personnel in this field and are not limited or elaborated here; the image acquisition method can be adjusted according to the specific configuration of the pipeline inspection and cleaning robot and is not limited here.

[0051] The detergent spray speed analysis module 102 is used to divide the grayscale image into regions to obtain multiple region images, perform image segmentation and connected domain analysis in each region image to obtain the dirty region in each region image in the grayscale image, and in each region image, adjust the standard detergent spray speed based on the area and number of dirty regions to determine the detergent spray speed for each dirty region.

[0052] Since the information in the entire grayscale image is relatively complex, in order to more carefully determine the dirty area in the grayscale image, the grayscale image can be first divided into regions to obtain multiple regional images; since the grayscale value of the dirty area and the grayscale value of the inner wall of the pipe will show a large difference, image segmentation can be performed in each regional image to identify the dirty area in each regional image. In the regional image, due to the different number and area of ​​dirty areas, it means that the dirtiness of the regional image is different. The more serious the dirtiness of the regional image, the greater the amount of detergent required, and the spraying speed of the detergent needs to be increased. Therefore, in each regional image, the standard spraying speed of the detergent can be adjusted based on the area and number of dirty areas, so as to determine the detergent spraying speed of each dirty area. At this time, the detergent spraying speed of the dirty area in each regional image can be preliminarily matched with the dirtiness in each regional image, thereby improving the cleaning effect and pertinence.

[0053] The grayscale image is divided into regions to obtain a plurality of regional images. Specifically, the grayscale image can be divided into four equal parts to obtain four small regional images. The specific number of divisions and the division method can be adjusted according to the implementation scenario and are not limited here.

[0054] Image segmentation is one of the effective means to simplify machine vision algorithms. After image segmentation, each region has some similar characteristics, while the characteristics between different regions are significantly different. In the pipe to be cleaned, there is a significant grayscale difference between the dirty area and the background area of ​​the pipe without dirt. Therefore, image segmentation can be performed in each regional image to obtain the dirty area in each regional image in the grayscale image.

[0055] Preferably, in one embodiment of the present invention, the method for obtaining the dirty area includes:

[0056] The OTSU threshold segmentation method can automatically calculate an optimal global threshold based on the grayscale histogram of the image. This threshold can be used to segment the image into foreground and background. Therefore, in each regional image, the regional image is threshold segmented based on the OTSU threshold segmentation method to obtain a binary image.

[0057] Usually, in a binary image, an area with a grayscale value of 255 (or 1) is regarded as a foreground area, and an area with a grayscale value of 0 is regarded as a background area. However, inside the pipe to be cleaned, the material of the pipe wall and other surface workmanship will lead to its strong reflective ability for light, so the grayscale value will be higher than the grayscale value of dirt such as oil. Therefore, in an embodiment of the present invention, an area with a grayscale value of 0 in the binary image is taken as a foreground area, and an area corresponding to the foreground area in the grayscale image is taken as a target area. At this time, the target area contains all the information of the dirty area and is in the form of a grayscale image.

[0058] Finally, in the target area, connected domain analysis and morphological closing operation are performed to obtain all independent connected domains. These connected domains usually represent different parts of the dirt, so each connected domain is regarded as a dirty area.

[0059] It should be noted that the OTSU threshold segmentation method, the connected domain analysis method and the morphological closing operation are all well-known technologies, and the specific processes are not described here.

[0060] For regional images with a large number of dirty areas or a large area, the detergent spraying speed should be increased to facilitate faster dirt removal and improve cleaning efficiency; while for regional images with a small area or a small number of dirty areas, the detergent spraying speed should be reduced to avoid unnecessary waste. Therefore, in each regional image, the standard detergent spraying speed can be adjusted based on the area and number of dirty areas, so as to determine the adaptive detergent spraying speed for each dirty area.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the cleaning agent spraying speed includes:

[0062] See also Figure 2 , which shows a method flow chart of a method for obtaining a cleaning agent spraying speed in one embodiment of the present invention, the method comprising the following steps:

[0063] Step S201: In each region image, based on the area and number of dirty regions, calculate the cleaning importance factor of each region image.

[0064] In each area image, the area and number of dirty areas can reflect the degree to which each area image needs to be cleaned, so in each area image, the product of the area mean of all dirty areas and the number of dirty areas is normalized to the value obtained as the cleaning importance factor of each area image. The larger the area mean of the dirty areas and the more dirty areas there are, the higher the degree of dirtiness, and the larger the cleaning importance factor, indicating that the cleaning importance corresponding to the area image is higher, and more detergent is needed. Normalization is a technical means well known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0065] Step S202: Based on the differences between the important cleaning factors of all the regional images and in combination with the preset cleaning agent dosage, the total amount of cleaning agent for each regional image is determined.

[0066] The total amount of detergent needs to comprehensively consider the cleaning importance factor of the regional image and the preset detergent dosage, so as to ensure that both the cleaning needs are met and waste is avoided. The difference between the cleaning importance factor of each regional image and the minimum cleaning importance factor of all regional images is used as the first adjustment factor for the detergent dosage corresponding to each regional image. The minimum cleaning importance factor can be used as a benchmark value. At this time, the larger the first adjustment factor of the detergent dosage, the higher the degree of cleaning required in a certain regional image, and the greater the amount of detergent required.

[0067] Then the difference between the preset maximum detergent dosage and the preset minimum detergent dosage is used as the second detergent dosage adjustment factor. The role of the second detergent dosage adjustment factor is to control the detergent dosage within a reasonable range to prevent the detergent dosage from exceeding the maximum range.

[0068] Then, the product of the first adjustment factor of the detergent usage corresponding to each regional image and the second adjustment factor of the detergent usage is used as the adjustment parameter corresponding to each regional image. The adjustment parameter represents the amount of detergent that needs to be increased for each regional image. The larger the adjustment parameter, the higher the required cleaning level of the regional image and the greater the required detergent content.

[0069] Finally, the sum of the minimum preset detergent dosage and the adjustment parameter corresponding to each regional image is taken as the total detergent amount for each regional image. At this time, the larger the total detergent amount, the higher the degree of cleaning required for the dirty area in the regional image.

[0070] It should be noted that the preset maximum detergent usage is set to 20 kg, and the preset minimum detergent usage is set to 5 kg. The specific values ​​can be adjusted according to the implementation scenario and are not limited here.

[0071] Step S203: In each area image, based on the corresponding total amount of detergent, the total area of ​​the dirty area and the preset spraying speed, the detergent spraying speed of the dirty area is obtained.

[0072] Since a reasonable spraying speed of the detergent is also one of the key factors affecting the cleaning effect, the detergent spraying speed is calculated according to the dirtiness of each area image, which can effectively improve the cleaning efficiency.

[0073] Based on the above steps, the total amount of detergent required for the dirty area in each area image can be calculated, and then the ratio of the total amount of detergent in each area image to the total area of ​​all dirty areas in each area image can be normalized as the detergent spray speed adjustment factor corresponding to the dirty area in each area image. The detergent spray speed adjustment factor reflects the distribution density of the detergent in the dirty area in each area image, and the larger the value, the higher the required detergent distribution density. Normalization is a technical means well known to those skilled in the art, and the normalization function can be linear normalization or standard normalization, etc. The specific normalization method is not limited here.

[0074] Finally, in each regional image, the detergent spraying speed of the dirty area in each regional image is determined according to the corresponding detergent spraying speed adjustment factor and the preset spraying speed. Since the larger the detergent spraying speed adjustment factor, the higher the detergent distribution density required for the dirty area in the regional image, the detergent spraying speed should be increased. Therefore, the detergent spraying speed adjustment factor and the preset spraying speed are positively correlated with the detergent spraying speed. In this embodiment of the present invention, the sum of the detergent spraying speed adjustment factor corresponding to each regional image and the preset spraying speed can be directly used as the detergent spraying speed of the dirty area in each regional image.

[0075] It should be noted that the preset spray rate is set to 20ml / s; the specific value can be adjusted according to the actual scenario and is not limited here.

[0076] At this point, the detergent spraying speed corresponding to each dirty area in each area image can be obtained.

[0077] The residence time analysis module 103 is used to analyze the grayscale values ​​of the pixels in the grayscale image to obtain the dirtiness value of each dirty area in each area image. In each area image, based on the area of ​​the dirty area and the dirtiness value, the residence time of the cleaning robot per unit area of ​​each dirty area is determined.

[0078] After determining the detergent spraying speed corresponding to each dirty area, it is also necessary to determine the residence time of the cleaning robot at each position. For dirty areas with higher levels of dirtiness, the cleaning robot should stay longer. For the quantification of the degree of dirtiness, since the material and surface treatment of the pipeline will affect its reflection of light, increase the reflectivity of light, and thus appear as a higher grayscale value in the grayscale image, and the dirty area inside the pipeline often has a lower reflectivity or a higher absorption rate, so under the same lighting conditions, the light intensity reflected back by the dirty area is weaker, resulting in a lower grayscale value in the grayscale image. Therefore, the degree of dirtiness of each dirty area in each regional image can be quantified based on the grayscale value of the pixel point in the grayscale image. Then, in each regional image, the residence time of the cleaning robot per unit area of ​​each dirty area can be determined based on the area of ​​the dirty area and the degree of dirtiness.

[0079] Preferably, in one embodiment of the present invention, the method for obtaining the dirtiness value includes:

[0080] Based on the above analysis, we know that the more dirty an area is, the lower its grayscale value will be. Therefore, in each area image, the mean grayscale value of all pixels in each dirty area is calculated. The smaller the mean grayscale value is, the higher the degree of dirtiness of the dirty area is. Therefore, the mean grayscale value is negatively correlated and normalized to achieve logical relationship correction, and the value after negative correlation mapping and normalization is used as the degree of dirtiness of each dirty area. At this time, the larger the degree of dirtiness, the higher the degree of dirtiness of the dirty area, and the longer the cleaning robot should stay per unit area in the dirty area to ensure the cleaning effect. It should be noted that the negative correlation mapping and normalization method here can be used. Function, where It represents an exponential function with the natural constant e as the base, and x represents the independent variable.

[0081] The detergent spraying speed of each dirty area is determined in the detergent spraying speed analysis module 102. In this module, by accurately identifying the area and degree of dirtiness of the dirty area, the cleaning robot can allocate cleaning time and resources more specifically. For areas with a larger area or a higher degree of dirtiness, the robot can appropriately increase the dwelling time to ensure that the detergent content can meet the cleaning requirements, thereby thoroughly cleaning the dirty area; while for areas with a smaller area or a lower degree of dirtiness, the dwelling time can be appropriately reduced, thereby improving the overall cleaning efficiency.

[0082] Preferably, in one embodiment of the present invention, the method for obtaining the residence time includes:

[0083] See also Figure 3 , which shows a method flow chart of a method for obtaining residence time in one embodiment of the present invention, the method comprising the following steps:

[0084] Step S301: In each region image, the dirty regions are sorted based on their areas to obtain a sorting sequence.

[0085] In order to distinguish the size of the dirty area and provide a basis for the subsequent determination of the time factor, in each regional image, all the dirty areas are sorted in ascending order according to the area to obtain a sorting sequence.

[0086] Step S302: Determine a time factor based on the position of each dirty area in the corresponding sorting sequence.

[0087] In the sorting sequence, the ratio of the sequence number of each dirty area to the median of the sequence number in the sorting sequence is used as the time factor corresponding to each dirty area. At this time, the larger the time factor of a dirty area, the larger the sequence number of the dirty area in the sorting sequence, that is, the larger the area, so the cleaning robot should increase the stay time when cleaning the dirty area; conversely, the smaller the time factor of a dirty area, the smaller the sequence number of the dirty area in the sorting sequence, that is, the smaller the area, so the cleaning robot should reduce the stay time when cleaning the dirty area.

[0088] It should be noted that the sequence number value should start from 1.

[0089] Step S303: Determine the residence time of the cleaning robot per unit area of ​​each dirty area according to the time factor, dirtiness value and preset residence time corresponding to each dirty area.

[0090] Based on the above analysis, it can be known that the larger the dirtiness value of the dirty area, the higher the dirtiness of the dirty area, and the longer the cleaning robot should stay per unit area in the dirty area; the larger the time factor corresponding to the dirty area, the larger the sequence value of the dirty area in the sorting sequence, that is, the larger the area, and the longer the cleaning robot stays when cleaning the dirty area. Therefore, the multiplication of the time factor corresponding to each dirty area, the dirtiness value, and the preset stay time is used as the stay time of the cleaning robot per unit area in each dirty area. At this time, it can be ensured that the cleaning robot stays longer when cleaning dirty areas with a large dirtiness and large area, thereby ensuring the cleaning effect.

[0091] It should be noted that the preset stay time is set to 2s, and the specific value can be adjusted according to the implementation scenario and is not limited here.

[0092] At this point, the residence time of the cleaning robot per unit area of ​​each dirty area in each area image can be obtained.

[0093] The pipeline cleaning module 104 is used to clean the inside of the pipeline to be cleaned by using a cleaning robot based on the spraying speed of the cleaning agent and the residence time per unit area in each dirty area.

[0094] In module 102, the detergent spraying speed of each dirty area can be obtained, which is one of the key factors affecting the cleaning effect; in module 103, the residence time of the cleaning robot per unit area of ​​each dirty area can be determined, which represents the length of time the cleaning robot performs the cleaning task per unit area of ​​the dirty area. Therefore, combining the two can achieve targeted cleaning of different dirty areas inside the cleaning pipe to be cleaned, which not only ensures the cleaning effect but also improves the cleaning efficiency.

[0095] Preferably, in one embodiment of the present invention, in each dirty area, based on the spraying speed of the cleaning agent and the residence time per unit area, a cleaning robot is used to clean the inside of the pipeline to be cleaned, including:

[0096] When the cleaning robot moves to different dirty areas, the residence time of the cleaning robot per unit area is set to the residence time per unit area of ​​the corresponding dirty area, and the detergent spraying speed of the cleaning robot is set to the detergent spraying speed of the corresponding dirty area. In this way, dirty areas with different cleaning needs can obtain detergent content that matches them, so that when the cleaning robot is used to clean the dirty areas in the pipe, targeted cleaning can be achieved and the cleaning effect can be improved.

[0097] In summary, first obtain a grayscale image of the inside of the pipe to be cleaned to provide basic data for the subsequent identification of dirty areas inside the pipe. In order to more accurately identify the dirty areas inside the pipe, the grayscale image can be divided into regions to obtain multiple regional images, and then in each regional image, the dirty areas can be accurately identified based on image segmentation technology. Since the number and size of dirty areas can be used as a standard for evaluating the amount of detergent required, the standard detergent spraying speed can be adjusted based on the area and number of dirty areas to determine the detergent spraying speed of each dirty area. At this time, the detergent spraying speed can better match the dirty area. Furthermore, after determining the detergent spraying speed of each dirty area, it is also necessary to determine the residence time at each position, so that the amount of detergent can be more accurately matched with the dirtiness of each position. Since the grayscale value of the dirty area is usually quite different from the grayscale value of the pipe wall, the grayscale value of the pixel in the grayscale image can be analyzed to obtain the degree of dirtiness of each dirty area in each area image. The degree of dirtiness is analyzed from the level of grayscale value to quantify the degree of dirtiness. Then, the residence time of the cleaning robot on the unit area of ​​each dirty area can be determined according to the area of ​​the dirty area and the degree of dirtiness. By optimizing the residence time, it can be ensured that each position in the dirty area is fully cleaned. Finally, in each dirty area, the pipeline detection cleaning robot is used to clean each dirty position of the pipeline in combination with the corresponding detergent spraying speed and residence time. The embodiment of the present invention can accurately control the amount of detergent by accurately adjusting the spraying speed of the detergent in each dirty area and the residence time of the robot, and implement differentiated cleaning strategies for different dirty areas, thereby effectively improving the cleaning effect.

[0098] An embodiment of the present invention also provides a pipeline inspection and cleaning robot, which includes a processor and a memory, wherein the memory stores at least one instruction, a program, a code set or an instruction set, and when the at least one instruction, a program, a code set or an instruction set is loaded and executed by the processor, the steps in a pipeline inspection and cleaning robot control system are implemented.

[0099] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A pipeline inspection and cleaning robot control system, characterized in that: The system comprises: An image data acquisition module, used to acquire a grayscale image of the interior of the pipeline to be cleaned; A detergent spray speed analysis module is used to divide the grayscale image into regions to obtain a plurality of region images, perform image segmentation in each region image to obtain a dirty region in each region image in the grayscale image, and adjust the detergent standard spray speed based on the area and number of the dirty regions in each region image to determine the detergent spray speed for each dirty region; The residence time analysis module is used to analyze the grayscale values ​​of the pixels in the grayscale image, so as to obtain the dirtiness value of each dirty area in each area image, and in each area image, based on the area of ​​the dirty area and the dirtiness value, determine the residence time of the cleaning robot per unit area of ​​each dirty area; A pipeline cleaning module is used to clean the inside of the pipeline to be cleaned by using a cleaning robot based on the spraying speed of the cleaning agent and the residence time per unit area in each dirty area; The method for determining the cleaning agent spraying speed includes: In each region image, a cleaning importance factor of each region image is calculated based on the area and number of dirty regions; Based on the differences between the important cleaning factors of all the regional images and in combination with the preset cleaning agent dosage, the total amount of cleaning agent for each regional image is determined; The value after normalizing the ratio of the total amount of detergent in each regional image to the total area of ​​all dirty regions in each regional image is used as the detergent spray speed adjustment factor corresponding to the dirty region in each regional image; In each area image, the detergent spraying speed of the dirty area in each area image is determined according to the corresponding detergent spraying speed adjustment factor and the preset spraying speed, and the detergent spraying speed adjustment factor and the preset spraying speed are both positively correlated with the detergent spraying speed.

2. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: The method for obtaining the dirty area comprises: In each regional image, threshold segmentation is performed on the regional image based on the OTSU threshold segmentation method to obtain a binary image; Taking the area with a gray value of 0 in the binary image as the foreground area, and taking the area corresponding to the foreground area in the gray image as the target area; In the target area, connected domain analysis and morphological closing operation are performed to obtain all connected domains, and each connected domain is regarded as a dirty area.

3. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: The method for obtaining the cleaning important factor includes: In each region image, a value obtained by normalizing the product of the area mean of all dirty regions and the number of dirty regions is used as the cleaning importance factor of each region image.

4. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: The method for obtaining the total amount of detergent in each area image includes: The difference between the cleaning importance factor of each regional image and the minimum cleaning importance factor of all regional images is used as the first adjustment factor of the cleaning agent dosage corresponding to each regional image; The difference between the preset maximum detergent dosage and the preset minimum detergent dosage is used as a second detergent dosage adjustment factor; The product of the first adjustment factor of the detergent dosage corresponding to each regional image and the second adjustment factor of the detergent dosage is used as the adjustment parameter corresponding to each regional image; The sum of the minimum preset detergent dosage and the adjustment parameter corresponding to each regional image is used as the total detergent dosage of each regional image.

5. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: The method for obtaining the dirtiness value comprises: In each area image, the mean grayscale value of all pixels in each dirty area is calculated, and the mean grayscale value is negatively correlated and normalized to obtain a value that serves as the dirtiness value of each dirty area.

6. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: The method for obtaining the residence time includes: In each region image, all dirty regions are sorted in ascending order according to their area to obtain a sorted sequence; Determining a time factor based on the position of each dirty region in the corresponding sorted sequence; The multiplication of the time factor corresponding to each dirty area, the dirtiness value and the preset stay time is used as the stay time per unit area of ​​the cleaning robot in each dirty area.

7. A pipeline inspection and cleaning robot control system according to claim 6, characterized in that: The method for obtaining the time factor includes: In the sorting sequence, the ratio of the sequence number of each dirty area to the median value of the sequence number in the sorting sequence is used as the time factor corresponding to each dirty area, wherein the sequence number value of the sorting sequence starts from 1.

8. A pipeline inspection and cleaning robot control system according to claim 1, characterized in that: In each dirty area, based on the spraying speed of the cleaning agent and the residence time per unit area, the cleaning robot is used to clean the inside of the pipeline to be cleaned, including: When the cleaning robot moves to a dirty area, the residence time of the cleaning robot per unit area is set to the residence time per unit area of ​​the dirty area, and the detergent spraying speed of the cleaning robot is set to the detergent spraying speed of the dirty area, so that the cleaning robot is used to clean the dirty area in the pipeline.

9. A pipeline inspection and cleaning robot, characterized in that: It includes a processor and a memory, wherein at least one instruction, a program, a code set or an instruction set is stored in the memory, and when the at least one instruction, a program, a code set or an instruction set is loaded and executed by the processor, the steps of a pipeline inspection and cleaning robot control system as described in any one of claims 1 to 8 are implemented.

Images