Life prediction method and system based on concrete pipe inspection data

Abstract

The present invention relates to the field of image data processing technology, and in particular to a life prediction method and system based on concrete pipe detection data. The method comprises the following steps: each edge in a concrete pipe image divides a corresponding edge window into two parts, which are respectively recorded as a first area or a second area, and determines the edge boundary coefficient; the edges with the same vector direction in the group with the largest boundary coefficient in the boundary coefficient segmentation result of all edges by the Otsu threshold method are taken as the same edge group; the boundary line in the concrete pipe image is obtained by two edge groups containing the largest number of edges, the concrete pipe image is divided into a dry area and a wet area, the life index of the concrete pipe image is determined, and a concrete pipe life prediction model is constructed according to the concrete pipe image and the corresponding life index to obtain the life prediction result of the concrete pipe, thereby effectively improving the accuracy of the life prediction of the concrete pipe.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a life prediction method and system based on concrete pipe detection data. Background Art

[0002] Concrete pipes have durability, load-bearing capacity and corrosion resistance, and can be used in road construction, tunnel construction, water conservancy facilities construction and other fields. The length of its life is directly related to the safety and stability of the overall structure of the relevant construction. By predicting the life of concrete pipes, potential safety hazards of concrete pipes can be discovered in time to avoid safety accidents caused by pipeline aging, damage and other reasons.

[0003] Research in the prior art tends to predict the life of the pipeline by obtaining internal data of the pipeline. For example, a patent application document with publication number CN115561153A discloses an online corrosion detection method and a remaining life prediction method for a natural gas marine pipeline. The method obtains a simulated corrosion curve and an effective wall thickness change curve of a simulated sample segment; obtains monitoring data of at least three groups of monitoring segments in a unit time, obtains a correction parameter k, obtains the actual corrosion curve and the actual effective wall thickness change curve of the monitoring segment, and obtains the remaining life of the monitoring segment.

[0004] Although the above-mentioned prior art can improve the accuracy of the remaining life prediction of marine pipelines to a certain extent, it mainly predicts the life by obtaining the overall corrosion degree of the pipeline. The interior of the pipeline is usually divided into a dry area and a wet area. If the same detection method is used for all areas, the accuracy of the obtained pipeline life prediction value will be low.

[0005] Based on this, how to accurately predict the life of concrete pipes based on the inspection data is an urgent problem to be solved by technical personnel in this field. Summary of the invention

[0006] In order to solve the technical problem of how to accurately obtain the life prediction of a concrete pipe based on the detection data of the concrete pipe, the present invention provides a life prediction method and system based on the detection data of the concrete pipe.

[0007] In a first aspect, the present invention provides a life prediction method based on concrete pipe inspection data, which adopts the following technical solution:

[0008] The life prediction method based on concrete pipe detection data includes the following steps:

[0009] The edge window of each edge in the concrete pipe image is preset, wherein the length of the edge is equal to the length of the corresponding edge window; the edge divides the corresponding edge window into two parts, which are respectively recorded as the first area or the second area, and the absolute value of the difference between the ratio of the gray level mean of the first area and the gray level mean of the second area and 1 is used as the boundary coefficient of the edge; the group with the largest boundary coefficient in the boundary coefficient segmentation result of the Otsu threshold method for all edges is recorded as the boundary line group, and the edges with the same vector direction in the boundary line group are regarded as the same edge group; the boundary line in the concrete pipe image is obtained by the two edge groups containing the largest number of edges, and the concrete pipe image is divided into a dry area and a wet area; the maximum value of the life index of the dry area and the wet area is recorded as the life index of the concrete pipe image, and a concrete pipe life prediction model is constructed according to the concrete pipe image and the corresponding life index to obtain the life prediction result of the concrete pipe.

[0010] The present invention takes into account that the concrete pipe image is divided into an upper dry area and a lower wet area formed by the impact of flowing water. There may be errors in directly predicting the life of the concrete pipe image as a whole. Therefore, by analyzing the grayscale expression of all edges in the concrete pipe image in its edge window, the possibility that the edge is the boundary between the dry area and the wet area is obtained. Combined with the vector direction of the edge, the boundary that can distinguish the dry area from the wet area is finally accurately obtained. On this basis, the present invention can accurately obtain the life index of the concrete pipe image by calculating the life index of the dry area and the wet area respectively, thereby effectively improving the accuracy of the life prediction of the concrete pipe.

[0011] According to the life prediction method based on concrete pipe detection data provided by the present invention, the edge windows of each edge in the preset concrete pipe image also include: obtaining a concrete pipe video stream through an acquisition device, and pre-processing the concrete pipe video stream after frame extraction to obtain a concrete pipe image.

[0012] The present invention takes into account that there may be noise interference, data missing, etc. in the originally collected concrete pipe video stream, and therefore improves the quality of the concrete pipe image through preprocessing to prepare for subsequent data processing.

[0013] According to the life prediction method based on concrete pipe detection data provided by the present invention, the edge window acquisition method of the edge includes: acquiring edges parallel to the edge on each side of the edge as parallel lines of the edge, the length of each parallel line is equal to the length of the corresponding edge; acquiring the minimum value of the segmentation index in the area surrounded by the parallel lines on both sides of the edge, and taking the area surrounded by the parallel lines on both sides corresponding to the minimum value of the segmentation index as the edge window of the edge.

[0014] The present invention takes into account that the grayscale features of the damaged area in the concrete pipe image may be close to the grayscale features of the normal wet area. In order to reduce the calculation of the life index of the damaged area, the parallel lines on both sides corresponding to the minimum value of the segmentation index are obtained as the two side edges of the edge window. When determining the edge window, the damaged area is excluded, so that the life index of the concrete pipe image can be accurately obtained based on this in the subsequent process.

[0015] According to the life prediction method based on concrete pipe detection data provided by the present invention, the segmentation index in the area surrounded by the parallel lines on both sides of the edge satisfies the relationship:

[0016] ;

[0017] For the edge One of the parallel lines and the The segmentation index in the area enclosed by the parallel lines on the other side, For the The gray level in the area enclosed by the parallel lines on one side and the edge is The proportion of For the The gray level in the area enclosed by the parallel lines on the other side and the edge is The proportion of is the minimum function.

[0018] The present invention provides a precise method for calculating the segmentation index in an area surrounded by parallel lines on both sides. By respectively obtaining the degree of chaos in the area surrounded by the parallel lines on one side and the parallel lines on the other side and the edge, the segmentation index in the area surrounded by the parallel lines on both sides is obtained, so that the edges on both sides of the edge window can be accurately obtained.

[0019] According to the life prediction method based on concrete pipe detection data provided by the present invention, the edge vector acquisition method includes: subtracting the coordinates of the two end points of the edge to obtain the edge vector.

[0020] According to the life prediction method based on concrete pipe inspection data provided by the present invention, the method for obtaining the life index of the dry area and the wet area includes: ; is the life index of the dry area, is the maximum gray level in the dry area, The gray level in the dry area is ratio; determine the life index of the wet area.

[0021] According to the life prediction method based on concrete pipe detection data provided by the present invention, the life prediction result of the concrete pipe is obtained, and then the method further includes: judging the necessity of maintenance of the concrete pipe by comparing the life prediction result with a preset threshold value.

[0022] The present invention takes into account that the life prediction result determines the remaining service life of the concrete pipe. In order to reduce safety hazards, after the life prediction result is obtained, it is compared with a preset threshold, and corresponding processing is performed based on the comparison result.

[0023] In a second aspect, the present invention provides a life prediction system based on concrete pipe inspection data, which adopts the following technical solution:

[0024] The life prediction system based on concrete pipe detection data comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the life prediction method based on concrete pipe detection data is implemented.

[0025] By adopting the above technical solution, the above life prediction method based on concrete pipe detection data is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is manufactured according to the memory and the processor for easy use.

[0026] The present invention has the following technical effects:

[0027] Based on the above technical scheme, when determining the life prediction result of the concrete pipe based on the detection data of the concrete pipe, the present invention analyzes the grayscale performance of all edges in the concrete pipe image in its edge window to obtain the possibility that the edge is a dividing line, and combines the vector direction of the edge to finally accurately obtain the dividing line that can distinguish the dry area and the wet area; on this basis, the present invention calculates the life index of the dry area and the wet area respectively, which can reduce the influence of the grayscale difference between the dry area and the wet area on the calculation of the life index of the concrete pipe image, so that the life index of the concrete pipe image can be accurately obtained, and the accuracy of the life prediction of the concrete pipe is effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0029] Figure 1 A schematic flow chart of a life prediction method based on concrete pipe inspection data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0031] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0032] It should be noted that concrete pipes have durability, load-bearing capacity and corrosion resistance, and can be used in road construction, tunnel construction, water conservancy facilities construction and other fields. The length of its life is directly related to the safety and stability of the overall structure of the relevant construction. By predicting the life of concrete pipes, potential safety hazards of concrete pipes can be discovered in time to avoid safety accidents caused by pipeline aging, damage and other reasons.

[0033] Existing technologies mostly predict the life of pipelines by collecting images of them. However, there is usually water accumulation inside the pipeline, which causes the entire pipeline to be divided into dry areas and wet areas. If the same life detection method is used for all areas without distinction, the accuracy of the pipeline life prediction value obtained will be low.

[0034] Based on this, an embodiment of the present invention discloses a life prediction method based on concrete pipe detection data. The method analyzes the pixel changes at the edge of the concrete pipe to obtain a dividing line for dividing the dry area and the wet area, and then calculates the life of the dry area and the wet area respectively, so that the life prediction value of the concrete pipe can be accurately obtained based on the concrete pipe detection data.

[0035] For details, please refer to Figure 1 As shown, Figure 1 A schematic flow chart of a life prediction method based on concrete pipe inspection data provided by an embodiment of the present invention, the method specifically comprises the following steps S1 to S5.

[0036] S1: Preset an edge window of each edge in the concrete pipe image, wherein the length of the edge is equal to the length of the corresponding edge window.

[0037] The edge may be obtained by using a canny edge detection algorithm, which may be specifically set according to actual needs.

[0038] For example, in an embodiment of the present invention, an edge window of each edge in the concrete pipe image is preset, and the method also includes: obtaining a concrete pipe video stream through a collection device, and pre-processing the concrete pipe video stream after extracting frames to obtain a concrete pipe image.

[0039] The preprocessing may include image denoising, image grayscale processing, missing data interpolation, etc., which may be specifically configured according to actual needs, and the embodiment of the present invention does not impose too many restrictions thereon.

[0040] For example, the acquisition device may be a closed-circuit television robot, which may be configured according to actual needs.

[0041] It should be noted that the lower area inside the concrete pipe is usually eroded by water flow. Compared with the upper dry area, the wet area is darker in color and has a lower grayscale value. The difference in grayscale values ​​can be used to distinguish the two, thereby dividing the pipe into a dry area and a wet area. However, there may be exposed steel bars and damage in the upper area of ​​the pipe wall, and there may be garbage in the lower area. If the distinction is made directly based on the grayscale value, the dry area and the wet area may be misidentified, resulting in errors in the life prediction results of the concrete pipe.

[0042] It should be further explained that due to the effect of gravity, the wet area caused by water scouring is usually in the lower layer of the pipe, while the boundary between the dry area and the wet area is a horizontal straight line edge, and the grayscale difference on both sides of the edge is large.

[0043] Based on this, the embodiment of the present invention constructs an edge window of each edge, shields irrelevant information in the background, and focuses on the grayscale features on both sides of the edge to obtain the position of the boundary between the dry area and the wet area.

[0044] For example, in an embodiment of the present invention, the length of the edge is equal to the length of the corresponding edge window. The edge window acquisition method specifically includes the following two possible implementations:

[0045] In a possible implementation, an embodiment of the present invention can preset the width of the edge window to obtain an edge window that can enclose the current edge, and then repair the defective part in the edge window after obtaining it through a defect recognition model, and finally obtain the edge window of the edge.

[0046] The width of the edge window may be half of the length of the edge, and may be set according to actual needs. The embodiment of the present invention does not impose too many limitations on this.

[0047] For example, a defect recognition model can be obtained by inputting a concrete pipe image into a deep learning module; wherein the deep learning module can be a convolutional neural network or a generative adversarial network, etc., which can be specifically set according to actual needs.

[0048] Among them, the specific steps of pre-training the defect recognition model and repairing the defective part can be implemented by existing technologies, and the embodiments of the present invention will not be described in detail here.

[0049] For example, when constructing an edge window based on the length and width of the edge window of the edge, the edge may be taken as the center line of the edge window, and the width of the edge window may be evenly divided to finally obtain the edge window of the edge.

[0050] In this way, the embodiment of the present invention can ensure that sufficient information of the current edge and its surroundings is included by presetting the width of the edge window and setting a larger value for the width of the edge window, thereby avoiding the loss of important information, and can set the corresponding window width according to different image or data features, with high flexibility.

[0051] In another possible implementation, an embodiment of the present invention can obtain edges parallel to the edge on each side of the edge as parallel lines of the edge, and the length of each parallel line is equal to the length of the corresponding edge; obtain the minimum value of the segmentation index in the area enclosed by the parallel lines on both sides of the edge, and use the area enclosed by the parallel lines on both sides corresponding to the minimum value of the segmentation index as the edge window of the edge.

[0052] It is understandable that the length of the parallel line of the current edge may be greater than or less than the length of the current edge. Therefore, for an edge that is too long, the parallel line of the current edge can be obtained based on the length of the current edge, and the edge that is too short can be extended to be equal to the length of the current edge. In addition, since the edge window is a rectangular window, the parallel lines on both sides obtained based on the above steps are moved according to their vertical distance to the corresponding edge, so that they can overlap with the corresponding edge.

[0053] It should be noted that there may be damaged areas on the pipe wall. In order to reduce the impact of the damaged areas on the edge windows, the damaged areas can be avoided from being divided into the edge windows during the process of constructing the edge windows.

[0054] It should be further explained that the higher the degree of disorder in the edge window, the higher the possibility that a damaged area exists in the edge window. Based on this, the embodiment of the present invention obtains the parallel lines on both sides corresponding to the minimum degree of disorder. At this time, the area surrounded by the parallel lines on both sides has the lowest possibility of a damaged area, which is the edge window of the corresponding edge.

[0055] For example, in the embodiment of the present invention, the segmentation index in the area surrounded by the parallel lines on both sides of the edge is determined, and the details can be referred to the following relationship:

[0056] ;

[0057] In the formula, For the edge One of the parallel lines and the The segmentation index in the area enclosed by the parallel lines on the other side, For the The gray level in the area enclosed by the parallel lines on one side and the edge is The proportion of For the The gray level in the area enclosed by the parallel lines on the other side and the edge is The proportion of is the minimum function.

[0058] Among them, 1 is the parallel lines on one side, 2 is the parallel lines on the other side, and 255 is the maximum grayscale value in the area surrounded by the parallel lines on both sides.

[0059] In the above formula, the segmentation index can be used to characterize the degree of disorder in the area surrounded by the parallel lines on both sides. The farther the parallel lines on one side are from the edge, the drier the concrete is, the closer its color is to gray, and the higher the gray value is; the farther the parallel lines on the other side are from the edge, the wetter the concrete is, the closer its color is to black, and the lower the gray value is. , They respectively represent the degree of chaos in the area enclosed by the parallel lines on one side and the parallel lines on the other side and the edge. The larger the value, the greater the degree of chaos in the area enclosed by the parallel lines on both sides, the smaller the possibility that the parallel lines on both sides are edge lines, and the larger the segmentation index in the area enclosed by the parallel lines on both sides.

[0060] Based on the above formula, the parallel lines on both sides of the edge are arranged and combined, and the grayscale disorder degree in the area surrounded by the parallel lines on both sides is calculated respectively, so that the segmentation index in the area surrounded by the parallel lines on both sides can be obtained. The area surrounded by the parallel lines on both sides is all possible edge windows. The parallel lines on both sides corresponding to the minimum segmentation index in the area surrounded by all the parallel lines on both sides are the edge lines with the greatest possibility, and the sum of the vertical distances from the parallel lines on both sides to the corresponding edge is the width of the edge window, and finally the edge window of the current edge is obtained.

[0061] In this way, the embodiment of the present invention obtains the two parallel lines corresponding to the minimum degree of chaos in the area surrounded by the two parallel lines through iterative calculation, that is, the two side edges of the edge window of the current edge. The final edge window does not contain a damaged area, which is ready for the subsequent boundary line between the wet area and the dry area based on this.

[0062] After obtaining the edge window of each edge based on any of the above embodiments, continue to perform the following steps.

[0063] S2: Edge divides the corresponding edge window into two parts, which are respectively recorded as the first area or the second area, and the absolute value of the difference between the ratio of the gray level mean of the first area and the gray level mean of the second area and 1 is taken as the boundary coefficient of the edge.

[0064] It should be noted that the edge window of each edge obtained based on the above steps does not include defective parts, and the greater the difference in grayscale distribution between the dry area and the wet area.

[0065] Based on this, the embodiment of the present invention can obtain the possibility that the edge is a boundary line based on the grayscale distribution difference between the two sides of the edge in the edge window. The larger the difference, the greater the possibility that the edge is a boundary line.

[0066] Among them, the closer the absolute value of the difference between the ratio of the grayscale mean of the first area and the grayscale mean of the second area is to 1, the greater the difference between the grayscale mean of the first area and the grayscale mean of the second area, and the greater the grayscale difference on both sides of the edge; the closer the absolute value of the difference is to 0, the smaller the grayscale difference on both sides of the edge.

[0067] After obtaining the demarcation coefficient of each edge based on the above steps, continue to perform the following steps.

[0068] S3: The group with the largest boundary coefficient in the segmentation results of the boundary coefficient of all edges by the Otsu threshold method is recorded as the boundary line group, and the edges with the same vector direction in the boundary line group are regarded as the same edge group; the edges in the two edge groups containing the largest number of edges are connected respectively to obtain the boundary lines in the concrete pipe image.

[0069] It should be noted that the boundary between the wet area and the dry area in the concrete pipe image is usually the horizontal plane of the water flowing in the pipe, and the horizontal plane is usually a straight line with a consistent direction and relatively continuous.

[0070] Based on this, the embodiment of the present invention obtains a boundary line group that can be used as a boundary line through the Otsu threshold method, and then takes the edges in the same direction in the boundary line group as the same edge group, and takes the two edge groups containing the largest number of edges as the boundary lines of the pipe walls on both sides in the concrete pipe image, so as to realize the division of dry areas and wet areas.

[0071] For example, in an embodiment of the present invention, when obtaining the boundary coefficient segmentation result of all edges by the Otsu threshold method, all boundary coefficients can be used as possible thresholds, and the boundary coefficients can be divided into a group with a maximum boundary coefficient and a group with a minimum boundary coefficient based on the possible thresholds. The inter-class variance between the group with a maximum boundary coefficient and the group with a minimum boundary coefficient is calculated, and the possible threshold corresponding to the maximum value of the inter-class variance is used as the segmentation threshold. The group with a maximum boundary coefficient and the group with a minimum boundary coefficient obtained based on the segmentation threshold are the final segmentation results.

[0072] The specific step of dividing the boundary coefficients of all edges by the Otsu threshold method can be implemented by the existing technology, and the embodiment of the present invention will not be described in detail here.

[0073] For example, in an embodiment of the present invention, a method for obtaining an edge vector includes: obtaining the edge vector by subtracting the coordinates of two end points of the edge.

[0074] Among them, the left end of the edge can be used as the starting point of the vector of the edge, and the right end of the edge can be used as the end point of the vector of the edge; the right end of the edge can also be used as the starting point of the vector of the edge, and the left end of the edge can be used as the end point of the vector of the edge; the specific settings can be made according to actual needs.

[0075] After the edge vector is obtained based on the above method, the edge vector direction can be obtained by a calculation formula for the vector direction. The calculation formula for the vector direction is a prior art and will not be described in detail in the embodiment of the present invention.

[0076] It can be understood that the two dividing lines finally obtained are the dividing lines on the pipe walls on both sides of the concrete pipe.

[0077] In this way, the embodiment of the present invention can obtain a boundary line in the concrete pipe image by connecting the edges in the two edge groups containing the largest number of edges. The boundary line can be used to divide the dry area and the wet area. Then, the life index of the concrete pipe image can be obtained by respectively calculating the life index of the dry area and the wet area, that is, continue to execute the following steps.

[0078] S4: Divide the concrete pipe image into a dry area and a wet area by a dividing line, and record the maximum value of the life index of the dry area and the life index of the wet area as the life index of the concrete pipe image.

[0079] It should be noted that the dividing line obtained based on the above steps can divide the concrete pipe image into a dry area and a wet area, and the maximum value of the life index of the dry area and the wet area is the life index of the concrete pipe image.

[0080] For example, in the embodiment of the present invention, the life index of the dry area is determined, and the specific relationship can be referred to as follows:

[0081] ;

[0082] In the formula, is the life index of the dry area, is the maximum gray level in the dry area, The gray level in the dry area is proportion.

[0083] in, .

[0084] In the above formula, It is used to characterize the degree of damage in the dry area. The larger the value, the higher the degree of damage in the dry area and the smaller the remaining life of the concrete pipe. Conversely, the smaller the value, the smaller the degree of damage in the dry area and the larger the remaining life of the concrete pipe.

[0085] Similarly, the life index of the wet area can be determined by referring to the calculation formula of the life index of the dry area mentioned above, and the embodiment of the present invention will not be described in detail here.

[0086] In this way, after the embodiment of the present invention divides the dry area and the wet area by the dividing line and obtains their life index, the life index corresponding to the maximum damage degree is used as the life index of the concrete pipe image, so that the life prediction of the concrete pipe can be realized based on this, that is, continue to execute the following steps.

[0087] S5: A concrete pipe life prediction model is constructed according to the concrete pipe image and the corresponding life index to obtain a life prediction result of the concrete pipe.

[0088] It should be noted that the liquid flowing through the concrete pipe is usually consistent, and the material of the concrete pipe is consistent, so the degree of damage changes consistently with the length of use.

[0089] Based on this, the embodiment of the present invention obtains a concrete pipe life prediction model by training multiple collected concrete pipe video streams and corresponding life indexes.

[0090] By way of example, in an embodiment of the present invention, a concrete pipe life prediction model is constructed based on concrete pipe images and corresponding life indexes, including: manually labeling all concrete pipe images in a historically collected concrete pipe video stream using corresponding life indexes to obtain a labeled historical concrete pipe video stream; and using the labeled historical concrete pipe video stream to complete neural network model training to obtain a concrete pipe life prediction model.

[0091] The latest collected concrete pipe video stream is input into the concrete pipe life prediction model to obtain a life index sequence of the concrete pipe video stream; the maximum life index in the life index sequence is used as the predicted life index of the latest collected concrete pipe video stream.

[0092] Among them, the specific steps of training the concrete pipe life prediction model can be implemented by existing technologies, and the embodiments of the present invention will not be described in detail here.

[0093] For example, in the embodiment of the present invention, a life prediction result of a concrete pipe is obtained, and then the method further includes: judging the necessity of maintenance of the concrete pipe by comparing the life prediction result with a preset threshold value.

[0094] The preset threshold value may be set according to actual needs, and the embodiment of the present invention does not impose too many limitations on this.

[0095] Specifically, if the predicted life index is greater than the preset threshold, it means that the degree of damage of the concrete pipe is high and it is necessary to maintain the concrete pipe; if the predicted life index is not greater than the preset threshold, it means that the degree of damage of the concrete pipe is low.

[0096] It can be seen that in the embodiment of the present invention, when determining the life prediction result based on the concrete pipe detection data, an edge window of each edge in the concrete pipe image can be preset, wherein the length of the edge is equal to the length of the corresponding edge window; the edge divides the corresponding edge window into two parts, which are respectively recorded as the first area or the second area, and the absolute value of the difference between the ratio of the grayscale mean of the first area and the grayscale mean of the second area and 1 is used as the boundary coefficient of the edge; the group with the largest boundary coefficient in the boundary coefficient segmentation result of all edges by the Otsu threshold method is recorded as the boundary line group, and the edges with the same vector direction in the boundary line group are taken as the same edge group; the boundary line in the concrete pipe image is obtained by the two edge groups containing the largest number of edges, and the concrete pipe image is divided into a dry area and a wet area; the maximum value of the life index of the dry area and the wet area is recorded as the life index of the concrete pipe image, and a concrete pipe life prediction model is constructed according to the concrete pipe image and the corresponding life index to obtain the life prediction result of the concrete pipe.

[0097] In this way, the embodiment of the present invention analyzes the grayscale representation of all edges in the concrete pipe image in its edge window to obtain the possibility that the edge is the boundary line between the dry area and the wet area, and combines the vector direction of the edge to finally accurately obtain the boundary line that can distinguish the dry area from the wet area; on this basis, the embodiment of the present invention calculates the life index of the dry area and the wet area respectively, and can accurately obtain the life index of the concrete pipe image, thereby effectively improving the accuracy of the life prediction of the concrete pipe.

[0098] An embodiment of the present invention also discloses a life prediction system based on concrete pipe detection data, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a life prediction method based on concrete pipe detection data provided by the present invention is implemented.

[0099] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0100] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM, a dynamic random access memory DRAM, a static random access memory SRAM, an enhanced dynamic random access memory EDRAM, a high bandwidth memory HBM, a hybrid memory cube HMC, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0101] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0102] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A life prediction method based on concrete pipe inspection data, characterized in that: include: Presetting an edge window of each edge in the concrete pipe image, wherein the length of the edge is equal to the length of the corresponding edge window; The edge window acquisition method of an edge comprises: acquiring an edge parallel to the edge on each side of the edge as a parallel line of the edge, wherein the length of each parallel line is equal to the length of the corresponding edge; Obtain the minimum value of the segmentation index in the area surrounded by the parallel lines on both sides of the edge, and use the area surrounded by the parallel lines on both sides corresponding to the minimum value of the segmentation index as the edge window of the edge; The segmentation index in the area enclosed by the parallel lines on both sides of the edge satisfies the relationship: ; For the edge One of the parallel lines and the The segmentation index in the area enclosed by the parallel lines on the other side, For the The gray level in the area enclosed by the parallel lines on one side and the edge is The proportion of For the The gray level in the area enclosed by the parallel lines on the other side and the edge is The proportion of is the minimum function; The edge divides the corresponding edge window into two parts, which are respectively recorded as the first area or the second area, and the absolute value of the difference between the ratio of the gray level mean of the first area and the gray level mean of the second area and 1 is taken as the boundary coefficient of the edge; The group with the largest boundary coefficient in the segmentation results of all edges by the Otsu threshold method is recorded as the boundary line group, and the edges with the same vector direction in the boundary line group are regarded as the same edge group; the boundary line in the concrete pipe image is obtained by the two edge groups containing the largest number of edges, and the concrete pipe image is divided into a dry area and a wet area; The maximum value of the life index of the dry area and the wet area is recorded as the life index of the concrete pipe image. A concrete pipe life prediction model is constructed according to the concrete pipe image and the corresponding life index to obtain the life prediction result of the concrete pipe.

2. The life prediction method based on concrete pipe detection data according to claim 1 is characterized in that: The edge window of each edge in the preset concrete pipe image also includes: The concrete pipe video stream is obtained through the acquisition device, and the concrete pipe video stream is pre-processed after frame extraction to obtain the concrete pipe image.

3. The life prediction method based on concrete pipe detection data according to claim 1 is characterized in that: The edge vector acquisition method includes: The difference between the coordinates of the two end points of the edge is used to obtain the edge vector.

4. The life prediction method based on concrete pipe detection data according to claim 1 is characterized in that: The method for obtaining the lifespan index of the dry area and the wet area includes: ; is the life index of the dry area, is the maximum gray level in the dry area, The gray level in the dry area is proportion; Determine the life index for wet areas.

5. The life prediction method based on concrete pipe detection data according to claim 1 is characterized in that: The method of obtaining the life prediction result of the concrete pipe further includes: The necessity of maintenance of concrete pipes is determined by comparing the life prediction results with the preset threshold.

6. The life prediction system based on concrete pipe inspection data is characterized by: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the life prediction method based on concrete pipe detection data according to any one of claims 1 to 5 is implemented.

Images