Manhole gap detection method and related apparatus

Abstract

The application discloses a manhole cover gap detection method and related equipment, and the method comprises the following steps: processing continuous frame images of a road in a rain scene to obtain a plurality of target road images of continuous frames; the target road image comprises a manhole cover area and a road area; image segmentation is performed according to each frame of the target road image, and the segmentation result of the target road image is detected to determine the amount of water in the manhole cover area; based on the relationship between the amount of water in the manhole cover area and a preset amount of water, the detection result of the manhole cover gap is determined; whether the flow velocity of the road area is calculated according to the detection result of the manhole cover gap is judged, so as to determine whether the manhole cover gap is normal according to the flow velocity of the road area. The application can detect the manhole cover in the rain, reduce the detection difficulty of the manhole cover, and improve the detection accuracy of the manhole cover gap.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image detection, in particular to a manhole cover aperture detection method and related equipment. BACKGROUND

[0002] At present, in the process of road detection, the sewerage of the manhole cover corresponding to different roads needs to meet certain standards to avoid road waterlogging or even internal flooding due to rain. Since it is difficult to predict whether the road manhole cover will produce waterlogging on sunny days, it is necessary to detect the manhole cover on rainy days to determine whether the aperture of the manhole cover meets the sewerage standard. In related technologies, it is difficult to detect the sewer manhole cover on rainy days at present, and the detection accuracy of the sewer manhole cover aperture is easily affected by environmental factors. SUMMARY

[0003] In a first aspect, the main purpose of the present application is to provide a manhole cover aperture detection method, comprising:

[0004] processing the continuous frame images of the road in the rain scene to obtain a plurality of continuous frame target road images; the target road image comprises a manhole cover area and a road area;

[0005] performing image segmentation on each frame of the target road image, and detecting the segmentation result of the target road image to determine the sewerage of the manhole cover area;

[0006] determining the detection result of the manhole cover aperture based on the relationship between the sewerage of the manhole cover area and the preset sewerage;

[0007] judging whether to calculate the flow velocity of the road area according to the detection result of the manhole cover aperture, so as to determine whether the manhole cover aperture is normal according to the flow velocity of the road area.

[0008] Preferably, before the processing of the continuous frame images of the road in the rain scene to obtain clear road images, it comprises:

[0009] extracting the continuous frame images of the road from the collected road video stream;

[0010] performing pixel detection on the continuous frame images to determine the raindrop density in the continuous frame images;

[0011] determining the target frame image with raindrop density greater than a preset threshold according to the raindrop density in the continuous frame images, so as to judge whether it is the rain scene according to the target frame image.

[0012] Preferably, the image segmentation on each frame of the target road image and the detection of the segmentation result of the target road image to determine the sewerage of the manhole cover area comprise:

[0013] perform image segmentation on each frame of the target road image according to the pre-trained image segmentation model to determine a manhole cover pixel region and a road pixel region;

[0014] detect the manhole cover pixel region in each frame of the target road image in chronological order to determine the amount of water in the manhole cover region.

[0015] Preferably, the detecting the manhole cover pixel region in each frame of the target road image in chronological order to determine the amount of water in the manhole cover region comprises:

[0016] performing water accumulation detection on the manhole cover pixel region in each frame of the target road image to determine the pixel area of the water accumulation region in each frame of the target road image;

[0017] calculating the pixel area difference of the water accumulation region in adjacent two frames of the target road image to obtain a plurality of pixel area corresponding pixel differences;

[0018] performing weighted average calculation on the plurality of pixel differences in chronological order to determine the amount of water in the manhole cover region corresponding to the manhole cover region.

[0019] Preferably, the determining the detection result of the manhole cover aperture based on the relationship between the amount of water in the manhole cover region and the preset amount of water comprises:

[0020] comparing the amount of water in the manhole cover region with the preset amount of water;

[0021] If the amount of water in the manhole cover region is less than or equal to the preset amount of water, it is determined that the detection result of the manhole cover aperture is an abnormal state;

[0022] If the amount of water in the manhole cover region is greater than the preset amount of water, it is determined that the detection result of the manhole cover aperture is a normal state.

[0023] Preferably, the determining whether to calculate the water flow velocity of the road region according to the detection result of the manhole cover aperture to determine whether the manhole cover aperture is normal according to the water flow velocity of the road region comprises:

[0024] When the detection result of the manhole cover aperture is an abnormal state, the water flow velocity of the road region is calculated;

[0025] If the water flow velocity of the road region is greater than a preset speed, it is determined that the manhole cover aperture is in a normal state;

[0026] If the water flow velocity of the road region is less than or equal to a preset speed, it is determined that the manhole cover aperture is in an abnormal state.

[0027] Preferably, the calculating the water flow speed of the road region comprises:

[0028] detecting a road region in each frame of the target road image, and determining pixel points of a plurality of markers in the road region;

[0029] determining moving tracks and moving times of the plurality of markers in each frame of the target road image according to the pixel points of the plurality of markers;

[0030] calculating moving speeds corresponding to the plurality of markers based on the moving tracks and the moving times of the plurality of markers in each frame of the target road image;

[0031] averaging the moving speeds of the plurality of markers to obtain the water flow speed of the road region.

[0032] In a second aspect, an embodiment of the present application provides a manhole cover gap detection device, comprising:

[0033] a processing module configured to process continuous frame images of a road in a rain scene to obtain a plurality of target road images of continuous frames; the target road image comprises a manhole cover region and a road region;

[0034] a detection module configured to perform image segmentation according to each frame of the target road image, and detect a segmentation result of the target road image to determine a water inflow of the manhole cover region;

[0035] a determination module configured to determine a detection result of a manhole cover gap based on a relationship between the water inflow of the manhole cover region and a preset water inflow;

[0036] a calculation module configured to determine whether to calculate a water flow speed of the road region according to the detection result of the manhole cover gap, so as to determine whether the manhole cover gap is normal according to the water flow speed of the road region.

[0037] In a third aspect, an embodiment 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, and the processor implements steps of a manhole cover gap detection method as described above when executing the computer program.

[0038] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executable on a processor to implement steps of a manhole cover gap detection method as described above.

[0039] The above scheme of the present application at least has the following beneficial effects:

[0040] The manhole cover gap detection method provided by the application first processes continuous frame images of a road in a rain scene to obtain a plurality of continuous frame target road images, performs image segmentation on each frame target road image, detects the segmentation result of the target road image to determine the amount of water flowing into the manhole cover area, then determines the detection result of the manhole cover gap based on the relationship between the amount of water flowing into the manhole cover area and a preset amount of water, and finally determines whether to calculate the water flow velocity of the road area according to the detection result of the manhole cover gap, so as to determine whether the manhole cover gap is normal according to the water flow velocity of the road area. Thus, the manhole cover can be detected on a rainy day, the detection difficulty of the manhole cover is reduced, and the detection accuracy of the manhole cover gap is improved. BRIEF DESCRIPTION OF DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from the structures shown in the drawings without creative labor.

[0042] Figure 1 The overall flowchart of the manhole cover gap detection method provided by the embodiment of the present application is shown in the figure.

[0043] Figure 2 The structure block diagram of the manhole cover gap detection device provided by the embodiment of the present application is shown in the figure.

[0044] Figure 3 The structure block diagram of the electronic device provided by the embodiment of the present application is shown in the figure.

[0045] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0047] The terms "first", "second", and "third" and the like in the description and in the claims of the present application and above-described drawings are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. The terms "comprises", "comprising", "includes", "including" and the like are to be construed open- ended, meaning that they allow for the possibility that there are additional steps or elements that are not listed. For example, a process, method, object, or apparatus that comprises a list of steps or elements is not necessarily limited to only those steps or elements but can include additional steps or elements not expressly listed or inherent to such process, method, object, or apparatus.

[0048] The scheme of the embodiments of the present application will be described below by way of example in conjunction with the relevant drawings.

[0049] As shown in the drawings, the specific embodiments of the present application provide a manhole cover gap detection method, comprising: Figure 1

[0050] 10, processing the continuous frame images of the road in the rain scene to obtain a plurality of continuous frame target road images; the target road image includes a manhole cover area and a road area.

[0051] In the specific embodiments of the present application, the target road image represents the road image after removing the raindrops, so that the continuous frame images in the rain scene can be processed by using a raindrop removal algorithm. Optionally, when detecting raindrops in the continuous frame images, the brightness values of each pixel point in the continuous frame images can be detected, the brightness difference between adjacent two frames is determined, and then it can be determined whether the brightness difference of the pixel point is within the brightness range corresponding to the raindrop according to the brightness difference. If the brightness difference of the pixel point is within the brightness range corresponding to the raindrop, it can be determined that the pixel point is a raindrop pixel point. Thus, the raindrop pixel points can be removed, and a clear target road image is obtained, so that the detection of the manhole cover area in the target road image is more accurate.

[0052] In an optional embodiment, before the above processing the continuous frame images of the road in the rain scene to obtain a clear road image, it includes: extracting the continuous frame images of the road from the collected road video stream; performing pixel detection on the continuous frame images to determine the raindrop density in the continuous frame images; determining the target frame image with a raindrop density greater than a preset threshold according to the raindrop density in the continuous frame images, to determine whether it is a rain scene according to the target frame image.

[0053] ​In a specific embodiment of the present invention, an inspection robot or a road camera can be used to collect road video streams in the corresponding area. The inspection robot or road camera can collect the corresponding video streams in real time or at regular intervals in each road area. Thus, corresponding continuous frame images can be extracted from the road video stream. The continuous frame images can be extracted within a predetermined time period, for example, the road video stream can be extracted once per hour. By detecting raindrops in each frame image, the raindrop density in each frame image can be determined. When the raindrop density is greater than a preset threshold, it can be determined as a target frame image and the number can be accumulated. It can be understood that when the number of target frame images exceeds a preset number, it can be determined that the road video stream in that time period corresponds to a rainy scene. Optionally, the preset threshold and preset number can be preset.

[0054] In an optional embodiment, raindrop density represents the proportion of pixels in an image. When detecting the raindrop density in each frame of an image, the pixels corresponding to the raindrops in each frame can be determined. Then, based on the pixels corresponding to the raindrops, the proportion of pixels corresponding to the raindrops in each frame can be determined, thereby identifying the target frame image. When the target frame image indicates a rainy scene, the extracted consecutive frame images can be de-rained to obtain multiple consecutive target road images, making the detection of the target road image more accurate. Optionally, when the raindrop density in each frame is similar, the extracted consecutive frame images can be de-rained to avoid insufficient detection accuracy due to the influence of rainfall when detecting water volume in the manhole cover area.

[0055] 20. Perform image segmentation based on each frame of the target road image, and detect the segmentation results of the target road image to determine the amount of water flowing into the manhole cover area.

[0056] In this embodiment, the target road image includes a road area and a manhole cover area. The water volume in the manhole cover area represents the water flow change in the manhole cover in the aforementioned road video stream. By segmenting the road area and the manhole cover area, detection can be performed separately for the manhole cover area and the road area to determine whether the manhole cover opening is normal. Optionally, the segmentation of the manhole cover area and the road area can be performed by detecting the edge of the manhole cover area and completing the image segmentation, or the manhole cover area can be segmented according to a pre-set segmentation range, for example, the manhole cover area and the road area can be segmented within a 2-meter radius around the manhole cover. After segmenting the target road image, the water volume in the manhole cover area can be further determined.

[0057] Specifically, the above-mentioned image segmentation based on each frame of the target road image and detection of the segmentation results of the target road image to determine the amount of water in the manhole cover area includes: performing image segmentation on each frame of the target road image according to a pre-trained image segmentation model to determine the manhole cover pixel area and the road pixel area; and detecting the manhole cover pixel area in each frame of the target road image in chronological order to determine the amount of water in the manhole cover area.

[0058] The image segmentation model can be SegNet, etc. By identifying the pixel region of the manhole cover, detection can be performed on the pixel region of the manhole cover to determine whether the amount of water flow in the manhole cover region has changed. The time sequence represents the time sequence of the road video stream. Therefore, the target road image contains the corresponding time sequence. By detecting each frame of the target road image in the time sequence, the amount of water flowing into the manhole cover region in each frame of the target road image can be determined. It can be understood that the amount of water flowing into the manhole cover region can be determined by judging whether the water accumulation in each frame of the target road image increases or decreases in the time sequence, and thus determine whether the manhole cover opening is normal.

[0059] Furthermore, the above-mentioned detection of manhole cover pixel regions in each frame of the target road image in chronological order to determine the amount of water in the manhole cover region includes: performing water accumulation detection on the manhole cover pixel regions in each frame of the target road image to determine the pixel area of ​​the water accumulation region in each frame of the target road image; calculating the difference between the pixel areas of the water accumulation regions in two adjacent frames of the target road image to obtain the pixel difference corresponding to multiple pixel areas; and calculating the weighted average of multiple pixel differences in chronological order to determine the amount of water corresponding to the manhole cover region.

[0060] In embodiments of the present invention, a target detection model can be used to detect each frame of the target road image to determine the pixel area of ​​the water accumulation area in each frame. When detecting the manhole cover pixel area, the color of the pixel corresponding to the water accumulation can be determined, and then the area of ​​the pixel corresponding to the water accumulation area can be calculated to determine the pixel area of ​​the water accumulation area. By determining the pixel area of ​​the water accumulation area in each frame of the target road image, the difference in pixel area between two adjacent frames can be calculated. The larger the pixel difference between two adjacent frames, the smaller the water volume in the water accumulation area; the smaller the pixel difference between two adjacent frames, the more balanced the water volume in the water accumulation area. Therefore, through... By performing a weighted average calculation on the calculated pixel differences and then comparing the average value with a preset water flow coefficient, the drainage volume corresponding to the manhole cover area can be determined. Optionally, the weights of multiple pixel differences can be gradually increased in chronological order. For example, if 10 pixel differences are calculated, the corresponding weights can be set to 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07, 0.08, 0.09, and 0.1 in chronological order. It should be understood that the above pixel difference weights are only examples and are not intended to limit the calculation. By determining the drainage volume in the manhole cover area, it can be determined whether the manhole cover openings are normal.

[0061] 30. Based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume, determine the detection results of the manhole cover pores.

[0062] In this embodiment, the preset drainage volume can be a pre-set water flow change. When the detection result of the manhole cover opening is determined to be normal, it means that the manhole cover opening corresponding to the manhole cover area meets the drainage volume standard of the manhole cover area. When the detection result of the manhole cover opening is determined to be abnormal, it means that the manhole cover opening corresponding to the manhole cover area does not meet the drainage volume standard of the manhole cover area. Therefore, the manhole cover in the manhole cover area can be replaced or checked for blockages.

[0063] Specifically, the above-mentioned determination of the manhole cover gap detection result based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume includes: comparing the drainage volume in the manhole cover area with the preset drainage volume; if the drainage volume in the manhole cover area is less than or equal to the preset drainage volume, the manhole cover gap detection result is determined to be abnormal; if the drainage volume in the manhole cover area is greater than the preset drainage volume, the manhole cover gap detection result is determined to be normal.

[0064] When the detection results of the manhole cover gaps are determined, the results can be output to a visual interface for display. This allows staff to inspect or replace the manhole covers in that area, reducing water accumulation on different roads during rainy days and improving the accuracy and efficiency of road inspections.

[0065] 40. Based on the test results of the manhole cover openings, determine whether the water flow velocity in the road area can be calculated, and then determine whether the manhole cover openings are normal based on the water flow velocity in the road area.

[0066] In a specific embodiment of the present invention, the water flow velocity in the road area represents the flow velocity of water from the road to the manhole cover. The detection results of the manhole cover openings include normal and abnormal states. A normal state indicates that the drainage volume corresponding to the manhole cover is normal, while an abnormal state indicates that the drainage volume corresponding to the manhole cover is abnormal, such as the manhole cover openings being too small or blocked. It can be understood that when the water flow velocity in the road area is too fast, it indicates that the rainfall is heavy. If the drainage volume of the manhole cover is less than or equal to the preset drainage volume, it can also be determined that the manhole cover area is in a normal state. When the water flow velocity in the road area is slow, it indicates that the rainfall is light. If the drainage volume of the manhole cover is less than or equal to the preset drainage volume, it can also be determined that the manhole cover area is in an abnormal state. Therefore, the drainage volume of the manhole cover can be further judged by combining the water flow velocity in the road area to determine whether the manhole cover area is normal, thereby improving the accuracy of manhole cover detection.

[0067] Specifically, the above-mentioned determination of whether to calculate the water flow velocity in the road area based on the detection results of the manhole cover openings, and the determination of whether the manhole cover openings are normal based on the water flow velocity in the road area, includes: calculating the water flow velocity in the road area when the detection results of the manhole cover openings are abnormal; if the water flow velocity in the road area is greater than the preset velocity, determining that the manhole cover openings are normal; if the water flow velocity in the road area is less than or equal to the preset velocity, determining that the manhole cover openings are abnormal.

[0068] The water flow velocity in the road area can be detected by optical flow velocity or other methods. If the manhole cover opening is determined to be in a normal state, further detection of the road area can be omitted to improve detection efficiency. If the manhole cover opening is determined to be in an abnormal state, further detection of the water flow velocity in the road area can be performed to improve detection accuracy. Optionally, the preset velocity can be pre-set, and the preset velocity for each road area can be different. For example, the preset velocity for road areas with a higher slope can be set to a higher value, while the preset velocity for road areas with a lower slope can be set to a lower value. By comparing the calculated water flow velocity of the road area with its corresponding preset velocity, it is possible to determine whether the manhole cover opening is in an abnormal or normal state.

[0069] In an optional embodiment, the above calculation of the water flow velocity in the road area includes: detecting the road area in each frame of the target road image and identifying the pixels of multiple markers in the road area; determining the movement trajectory and movement time of the multiple markers in each frame of the target road image based on the pixels of the multiple markers; calculating the movement speed corresponding to the multiple markers based on the movement trajectory and movement time of the multiple markers in each frame of the target road image; and averaging the movement speeds of the multiple markers to obtain the water flow velocity in the road area.

[0070] In embodiments of the present invention, the marker can be any object in the road area. The aforementioned movement trajectory represents the movement trajectory of the marker in the aforementioned road video stream, and the movement time represents the time it takes for the marker to move from the starting point to the ending point in the aforementioned road video stream. By inputting the target road image into the target detection model for detection, the pixel points of multiple markers in each frame of the target road image are determined. Thus, the movement trajectories of multiple markers can be determined according to the time sequence corresponding to each frame of the target road image. The corresponding starting point and ending point can be determined through the movement trajectory of each marker. Optionally, the ending point of the marker can be the manhole cover area. Therefore, by calculating the movement speed of each marker and averaging the movement speeds of each marker, the water flow speed in the road area can be determined. Optionally, when determining the movement speed of each marker, markers with movement trajectories close to a straight line can be selected, and the water flow speed in the road area can be determined by averaging the movement speeds corresponding to the markers with movement trajectories close to a straight line. Therefore, the water flow speed in the road area is used to further judge the manhole cover openings, so as to improve the detection accuracy of manhole cover openings.

[0071] In an optional embodiment, when the detection result of the manhole cover's perforations is determined to be abnormal, manhole cover images from recent historical periods under sunny conditions can be acquired. Perforation detection is then performed on these images to determine the area of ​​each perforation. By comparing the areas of each perforation pairwise, if all perforations in the sunny conditions are the same, the manhole cover area is marked as a replacement area. If perforations with different areas exist in the sunny conditions, the manhole cover area is marked as a repair area. Optionally, the area of ​​each perforation can be calculated using the pixels of each perforation. Based on the pixels corresponding to each perforation, the corresponding pixel area can be determined, and the areas of each perforation are then compared pairwise. The detection results for the manhole cover area can then be output to a visualization interface for staff to view, allowing for subsequent replacement or repair of the manhole cover.

[0072] The manhole cover gap detection method provided by this invention first processes consecutive frame images of a road in a rainy scene to obtain multiple consecutive frame target road images. Then, image segmentation is performed on each target road image, and the segmentation results are detected to determine the amount of water flowing through the manhole cover area. Next, based on the relationship between the amount of water flowing through the manhole cover area and a preset amount of water, the detection result of the manhole cover gaps is determined. Finally, based on the detection result of the manhole cover gaps, it is determined whether the water flow velocity in the road area should be calculated to determine whether the manhole cover gaps are normal. This allows for the detection of manhole covers during rainy days, reducing the difficulty of manhole cover inspection and improving the accuracy of manhole cover gap detection.

[0073] like Figure 2 As shown, an embodiment of the present invention provides a manhole cover porosity detection device 50, comprising:

[0074] The processing module 501 is used to process the continuous frame images of the road in the rain scene to obtain multiple consecutive frame target road images; the target road images include manhole cover areas and road areas;

[0075] The detection module 502 is used to perform image segmentation based on each frame of the target road image, and to detect the segmentation results of the target road image to determine the amount of water flowing into the manhole cover area.

[0076] The determination module 503 is used to determine the detection result of the manhole cover pores based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume.

[0077] The calculation module 504 is used to determine whether the water flow velocity in the road area is calculated based on the detection results of the manhole cover openings, so as to determine whether the manhole cover openings are normal based on the water flow velocity in the road area.

[0078] The manhole cover gap detection device 50 provided by this invention first processes consecutive frame images of a road in a rainy scene to obtain multiple consecutive frame target road images. Then, it performs image segmentation on each target road image and detects the segmentation results to determine the amount of water flowing through the manhole cover area. Next, based on the relationship between the amount of water flowing through the manhole cover area and a preset amount of water, it determines the detection result of the manhole cover gaps. Finally, based on the detection result of the manhole cover gaps, it determines whether the water flow velocity in the road area should be calculated to determine whether the manhole cover gaps are normal. This allows for the detection of manhole covers during rainy days, reducing the difficulty of manhole cover inspection and improving the accuracy of manhole cover gap detection.

[0079] It should be noted that the manhole cover porosity detection device 50 provided in the specific embodiment of the present invention is a device corresponding to the above-described manhole cover porosity detection method. All embodiments of the above-described manhole cover porosity detection method are applicable to the manhole cover porosity detection device 50. Each embodiment of the above-described manhole cover porosity detection device 50 has corresponding modules corresponding to the steps in the above-described manhole cover porosity detection method, which can achieve the same or similar beneficial effects. To avoid excessive repetition, each module in the manhole cover porosity detection device 50 will not be described in detail here.

[0080] like Figure 3 As shown, a specific embodiment of the present invention also provides an electronic device 60, including a memory 602, a processor 601, and a computer program stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program, it implements the steps of the above-described manhole cover pore detection method.

[0081] Specifically, processor 601 calls the computer program stored in memory 602 and performs the following steps:

[0082] The continuous frame images of the road in the rain scene are processed to obtain multiple consecutive frame images of the target road; the target road images include the manhole cover area and the road area.

[0083] Image segmentation is performed on each frame of the target road image, and the segmentation results of the target road image are detected to determine the amount of water flowing into the manhole cover area;

[0084] Based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume, the detection results of the manhole cover pores are determined;

[0085] Based on the test results of the manhole cover openings, determine whether the water flow velocity in the road area can be calculated, and then determine whether the manhole cover openings are normal based on the water flow velocity in the road area.

[0086] Optionally, before the processor 601 processes consecutive frame images of the road in the rainy scene to obtain a clear road image, it includes:

[0087] Continuous frame images of the road are extracted from the collected road video stream;

[0088] Pixel detection is performed on consecutive frames of images to determine the raindrop density in the consecutive frames of images;

[0089] Based on the raindrop density in consecutive frame images, target frame images with raindrop density greater than a preset threshold are identified, and the target frame images are used to determine whether it is a rainy scene.

[0090] Optionally, the processor 601 performs image segmentation based on each frame of the target road image and detects the segmentation results of the target road image to determine the amount of water flowing into the manhole cover area, including:

[0091] The target road image is segmented for each frame based on a pre-trained image segmentation model to determine the pixel regions of the manhole cover and the road.

[0092] The manhole cover pixel area in each frame of the target road image is detected in chronological order to determine the amount of water discharged into the manhole cover area.

[0093] Optionally, the processor 601 performs a time-sequential detection of the manhole cover pixel region in each frame of the target road image to determine the amount of water discharged into the manhole cover region, including:

[0094] Water accumulation detection is performed on the manhole cover pixel area in each frame of the target road image to determine the pixel area of ​​the water accumulation area in each frame of the target road image;

[0095] The pixel area difference of the water accumulation area in two adjacent frames of the target road image is calculated to obtain the pixel difference corresponding to multiple pixel areas;

[0096] The weighted average of multiple pixel differences is calculated in chronological order to determine the amount of water flowing into the manhole cover area.

[0097] Optionally, the processor 601 performs a process based on the relationship between the drainage volume in the manhole cover area and a preset drainage volume to determine the detection results of the manhole cover openings, including:

[0098] Compare the drainage volume in the manhole cover area with the preset drainage volume;

[0099] If the drainage volume in the manhole cover area is less than or equal to the preset drainage volume, the detection result of the manhole cover pores is determined to be abnormal.

[0100] If the amount of water flowing through the manhole cover area is greater than the preset amount, then the test result of the manhole cover opening is considered to be in a normal state.

[0101] Optionally, the processor 601 performs the function of determining whether to calculate the water flow velocity in the road area based on the detection results of the manhole cover openings, and determining whether the manhole cover openings are normal based on the water flow velocity in the road area, including:

[0102] When the detection result of the manhole cover pores is abnormal, the water flow velocity in the road area is calculated;

[0103] If the water flow velocity in the road area is greater than the preset velocity, the manhole cover opening is determined to be in a normal state.

[0104] If the water flow velocity in the road area is less than or equal to the preset velocity, the manhole cover opening is determined to be in an abnormal state.

[0105] Optionally, the calculations performed by processor 601 to obtain the water flow velocity in the road area include:

[0106] The road region in each frame of the target road image is detected to identify the pixels of multiple markers in the road region;

[0107] Based on the pixels of multiple markers, the movement trajectory and movement time of multiple markers in each frame of the target road image are determined;

[0108] Based on the movement trajectory and movement time of multiple markers in each frame of the target road image, the movement speed of multiple markers is calculated.

[0109] The water flow velocity in the road area is obtained by averaging the moving speeds of multiple markers.

[0110] That is, in a specific embodiment of the present invention, when the processor 601 of the electronic device 60 executes the computer program, it implements the steps of the above-mentioned manhole cover hole detection method, thereby enabling the detection of manhole covers on rainy days, reducing the difficulty of manhole cover detection, and improving the detection accuracy of manhole cover holes.

[0111] It should be noted that since the processor 601 of the electronic device 60 implements the steps of the above-described manhole cover pore detection method when executing the computer program, all embodiments of the above-described manhole cover pore detection method are applicable to the electronic device 60 and can achieve the same or similar beneficial effects.

[0112] The computer-readable storage medium provided in this embodiment of the invention stores a computer program. When the computer program is executed by a processor, it implements the various processes of the manhole cover pore detection method or the application-side manhole cover pore detection method provided in this embodiment of the invention, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0113] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0114] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0115] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for detecting porosity in manhole covers, characterized in that, include: Process consecutive frames of road images in a rainy scene to obtain multiple consecutive frames of target road images; The target road image includes manhole cover areas and road areas; Image segmentation is performed on each frame of the target road image, and the segmentation results are detected to determine the amount of water in the manhole cover area. This includes: segmenting each frame of the target road image using a pre-trained image segmentation model to determine the manhole cover pixel area and the road pixel area; detecting water accumulation in the manhole cover pixel area of ​​each frame of the target road image to determine the pixel area of ​​the water accumulation area in each frame of the target road image; calculating the difference between the pixel areas of the water accumulation areas in two adjacent frames of the target road image to obtain pixel differences corresponding to multiple pixel areas; and calculating a weighted average of the multiple pixel differences in chronological order to determine the amount of water in the manhole cover area. Based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume, the detection result of the manhole cover porosity is determined; Based on the detection results of the manhole cover openings, determine whether the water flow velocity in the road area can be calculated, and determine whether the manhole cover openings are normal based on the water flow velocity in the road area.

2. The method for detecting porosity in manhole covers according to claim 1, characterized in that, Before processing consecutive frame images of roads in a rainy scene to obtain multiple consecutive frame images of the target road, the following steps are included: Continuous frame images of the road are extracted from the collected road video stream; Pixel detection is performed on the consecutive frame images to determine the raindrop density in the consecutive frame images; Based on the raindrop density in the continuous frame images, a target frame image with a raindrop density greater than a preset threshold is determined, so as to determine whether the target frame image is the rain scene.

3. The method for detecting porosity in manhole covers according to claim 1, characterized in that, The determination of the manhole cover porosity detection results based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume includes: The amount of water flowing into the manhole cover area is compared with the preset amount of water flowing into the manhole cover area. If the amount of water flowing through the manhole cover area is less than or equal to the preset amount of water flowing through, then the detection result of the manhole cover pores is determined to be abnormal. If the amount of water flowing through the manhole cover area is greater than the preset amount of water flowing through, then the detection result of the manhole cover pores is determined to be normal.

4. The method for detecting porosity in manhole covers according to claim 3, characterized in that, The step of determining whether to calculate the water flow velocity in the road area based on the detection results of the manhole cover openings, and then determining whether the manhole cover openings are normal based on the water flow velocity in the road area, includes: When the detection result of the manhole cover pores is abnormal, the water flow velocity in the road area is calculated; If the water flow velocity in the road area is greater than the preset velocity, the manhole cover opening is determined to be in a normal state. If the water flow velocity in the road area is less than or equal to a preset velocity, the manhole cover opening is determined to be in an abnormal state.

5. The method for detecting porosity in manhole covers according to claim 4, characterized in that, The calculated water flow velocity in the road area includes: The road region in each frame of the target road image is detected to determine the pixels of multiple markers in the road region; Based on the pixels of the multiple markers, the movement trajectory and movement time of the multiple markers in each frame of the target road image are determined; Based on the movement trajectory and movement time of the multiple markers in each frame of the target road image, the movement speed corresponding to the multiple markers is calculated. The water flow velocity in the road area is obtained by averaging the moving speeds of the multiple markers.

6. A manhole cover porosity detection device, characterized in that, include: The processing module is used to process consecutive frame images of roads in a rainy scene to obtain multiple consecutive frame images of the target road. The target road image includes manhole cover areas and road areas; The detection module is used to perform image segmentation based on each frame of the target road image and to detect the segmentation results of the target road image to determine the amount of water in the manhole cover area. This includes: performing image segmentation on each frame of the target road image using a pre-trained image segmentation model to determine the manhole cover pixel area and the road pixel area; performing water accumulation detection on the manhole cover pixel area in each frame of the target road image to determine the pixel area of ​​the water accumulation area in each frame of the target road image; calculating the difference between the pixel areas of the water accumulation areas in two adjacent frames of the target road image to obtain pixel differences corresponding to multiple pixel areas; and calculating a weighted average of the multiple pixel differences in chronological order to determine the amount of water corresponding to the manhole cover area. The determination module is used to determine the detection result of the manhole cover porosity based on the relationship between the drainage volume in the manhole cover area and the preset drainage volume. The calculation module is used to determine whether the water flow velocity in the road area is calculated based on the detection results of the manhole cover openings, so as to determine whether the manhole cover openings are normal based on the water flow velocity in the road area.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the manhole cover porosity detection method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the manhole cover porosity detection method as described in any one of claims 1 to 5.

Images