Method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud

Abstract

The present application discloses a method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud. The method includes: performing coordinate transformation processing on the intensity information of each frame in the initial sonar acquisition data to obtain corresponding cross-sectional point cloud data, and performing central point noise removal processing to obtain overall boundary point cloud data; performing DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain target boundary point cloud data; performing neighborhood analysis on the target boundary point cloud data based on the local density of all points to obtain multiple boundary points; determining target boundary points from all the boundary points based on a preset angle set; and performing linear interpolation processing on all the target boundary points to obtain the target boundary contour for the square culvert. By performing coordinate transformation, denoising, clustering, etc. on the initial sonar data, the present application obtains the target boundary contour of the square culvert, which can effectively improve the accuracy of the culvert boundary contour compared with the manual acquisition solution.

Description

Technical Field

[0001] This application relates to, but is not limited to, the field of point cloud data technology, and in particular to a method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud. Background Art

[0002] With the continuous development of underwater detection technology, underwater sonar devices are increasingly widely used in culvert detection. In related technologies, an underwater unmanned submersible is used to carry a cross-section sonar for the investigation of the current situation of the culvert. By extracting the point cloud boundary from the collected sonar data, it is possible to help accurately identify the structure surface, thereby realizing the real-time evaluation of the underwater environment. The existing method is to obtain a point cloud image through a collection device, and then manually mark the point cloud image to obtain the boundary contour of the culvert, which cannot guarantee the accuracy of the culvert boundary contour. Summary of the Invention

[0003] Embodiments of this application provide a method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud, which can improve the accuracy of the culvert boundary contour.

[0004] In a first aspect, embodiments of this application provide a method for extracting the boundary contour of a square culvert based on sonar point cloud, including:

[0005] Obtain initial sonar acquisition data inside the square culvert, where the initial sonar acquisition data includes intensity information of multiple different scanning cross-sections;

[0006] Perform coordinate transformation processing on each frame of the intensity information to obtain cross-section point cloud data corresponding to each frame of the intensity information;

[0007] Perform central point noise removal processing on the cross-section point cloud data to obtain overall boundary point cloud data;

[0008] Perform DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain target boundary point cloud data;

[0009] Based on the local density of all points in the target boundary point cloud data, perform neighborhood analysis on the target boundary point cloud data to obtain multiple boundary points;

[0010] Determine target boundary points from all the boundary points based on a preset angle set, where the preset angle set includes multiple connection angles, each of the connection angles corresponds to at least one of the boundary points, the target boundary point is the boundary point that is the farthest from the point cloud center in the target boundary point cloud data at the same connection angle, and the connection angle is the connection angle between the boundary point and the point cloud center in the target boundary point cloud data;

[0011] Perform linear interpolation on all the target boundary points to obtain the target boundary contour for the square culvert.

[0012] In some embodiments, perform central point noise removal processing on the cross-sectional point cloud data to obtain overall boundary point cloud data, including:

[0013] Determine the number of point clouds in the cross-sectional point cloud data and the first position of each point;

[0014] Calculate the second position of the point cloud center of the cross-sectional point cloud data based on each of the first positions and the number of point clouds;

[0015] Calculate the Euclidean distance between each point in the cross-sectional point cloud data and the point cloud center based on the second position and all of the first positions;

[0016] Delete the points corresponding to the Euclidean distances less than the first distance threshold from the cross-sectional point cloud data to obtain the overall boundary point cloud data.

[0017] In some embodiments, the second position of the point cloud center of the cross-sectional point cloud data is calculated based on each of the first positions and the number of point clouds, and is obtained according to the following formula:

[0018] ;

[0019] where is the second position, is the first position, is the number of point clouds, i is the index of each point in the cross-sectional point cloud data.

[0020] In some embodiments, perform DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain target boundary point cloud data, including:

[0021] When the neighborhood of a reference point in the overall boundary point cloud data includes at least a preset number of points, determine the points within the neighborhood whose distance from the reference point is less than the preset radius as core points, where the reference point is any point in the overall boundary point cloud data;

[0022] Based on the preset radius, the preset number, and each of the core points, perform neighborhood expansion on each of the core points to obtain corresponding point cloud clusters, where, for each of the point cloud clusters, the distance between the points other than the core points and the core point is less than the preset radius, and the number of points other than the core points in each of the point cloud clusters is greater than or equal to the preset number;

[0023] Delete the points in the overall boundary point cloud data except for the point cloud clusters and the reference points to obtain the target boundary point cloud data.

[0024] In some embodiments, based on the local density of all points in the target boundary point cloud data, perform neighborhood analysis on the target boundary point cloud data to obtain a plurality of boundary points, including:

[0025] Calculate the reference number of adjacent points within the range of a second distance threshold for any point in the target boundary point cloud data;

[0026] Calculate the local density based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth positions of each adjacent point of any point in the target boundary point cloud data, and the reference number;

[0027] Determine the points in the target boundary point cloud data with a local density lower than a preset density threshold as the boundary points.

[0028] In some embodiments, calculate the local density based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth positions of each adjacent point of any point in the target boundary point cloud data, and the reference number, and obtain it according to the following formula:

[0029] ;

[0030] Where, is the position of the point in the target boundary point cloud data, is the local density corresponding to the point in the target boundary point cloud data, Ⅱ() is the indicator function, is the second distance threshold, is the adjacent point corresponding to, is the reference number, i is the index of the point corresponding to, j is the index of the point corresponding to.

[0031] In some embodiments, the connection angle is calculated according to the following formula:

[0032] ;

[0033] Where, is the connection angle, ( ) are the coordinates of the boundary point in the plane, ( ) are the coordinates of the point cloud center in the target boundary point cloud data.

[0034] In some embodiments, linear interpolation is performed on all of the target boundary points to obtain a target boundary contour for the square culvert, which is obtained according to the following formula:

[0035] ;

[0036] where is the position corresponding to the target boundary point, is an interpolation parameter, used to indicate to linear interpolation ratio, is the position of the linearly inserted point in the target boundary contour other than the target boundary point.

[0037] In a second aspect, an embodiment of the present application provides an electronic device, including at least one control processor and a memory communicatively connected to the at least one control processor; the memory stores instructions executable by the at least one control processor, and the instructions are executed by the at least one control processor to enable the at least one control processor to execute the method for extracting the boundary contour of a square culvert based on sonar point cloud as described in the first aspect.

[0038] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium storing computer-executable instructions for executing the method for extracting the boundary contour of a square culvert based on sonar point cloud as described in the first aspect.

[0039] The embodiment of the present application provides a method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud. The method includes: obtaining the initial sonar acquisition data inside the square culvert, where the initial sonar acquisition data includes the intensity information of multiple different scanning cross-sections; performing coordinate transformation processing on each frame of the intensity information to obtain the cross-section point cloud data corresponding to each frame of the intensity information; performing central point noise removal processing on the cross-section point cloud data to obtain the overall boundary point cloud data; performing DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain the target boundary point cloud data; performing neighborhood analysis on the target boundary point cloud data based on the local density of all points in the target boundary point cloud data to obtain multiple boundary points; determining target boundary points from all the boundary points based on a preset angle set, where the preset angle set includes multiple connection angles, each connection angle corresponds to at least one of the boundary points, the target boundary point is the boundary point with the farthest distance from the point cloud center in the target boundary point cloud data on the same connection angle, and the connection angle is the connection angle between the boundary point and the point cloud center in the target boundary point cloud data; performing linear interpolation processing on all the target boundary points to obtain the target boundary contour for the square culvert. According to the solution provided by the embodiment of the present application, the target boundary contour of the square culvert is obtained by performing coordinate transformation, denoising, clustering, etc. on the initial sonar data. Compared with the scheme of manual acquisition, the accuracy of the culvert boundary contour can be effectively improved. Description of the Drawings

[0040] Figure 1 is a flowchart of the steps of a method for extracting the boundary contour of a square culvert based on sonar point cloud provided by an embodiment of the present application;

[0041] Figure 2 is a structural diagram of an electronic device provided by another embodiment of the present application. Detailed Embodiments

[0042] In order to make the purpose, technical solution, and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

[0043] It can be understood that although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the order in the flowchart. Terms such as "first" and "second" in the specification, claims, or the above drawings are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence.

[0044] First, some terms involved in this application are analyzed as follows:

[0045] Sonar (Sound Navigation And Ranging): It is a technology for detecting, locating, and measuring objects by sound waves, widely used in underwater navigation, detection, and communication. It calculates distance, depth, and direction by emitting sound waves and receiving the reflected signals. Sonar can be divided into active sonar and passive sonar. Active sonar actively emits sound waves and analyzes the echoes, while passive sonar only receives external sounds and does not actively emit signals. Sonar has important application values in ocean research, submarine navigation, fishing, and military fields.

[0046] DBSCAN (Density-Based Spatial Clustering of Applications with Noise): It is a density-based clustering algorithm used to divide a dataset into clusters with similar densities and can effectively identify noise points. Different from traditional K-means clustering, DBSCAN does not require presetting the number of clusters but defines the density of a cluster through two parameters: the neighborhood radius eps and the minimum number of samples min_samples. The algorithm determines whether a point is a core point by checking if there are enough points in the neighborhood of each point. If it is a core point, all points in its neighborhood are grouped into the same cluster. If a point cannot be assigned to any cluster, it is marked as a noise point. DBSCAN can discover clusters of arbitrary shapes and has good handling ability for noise, but its performance is sensitive to parameter settings, especially when dealing with clusters of different densities, problems may occur.

[0047] Local density of point cloud: It refers to the distribution of other points in the neighborhood of a given point. Usually, it is measured by calculating the number or distance of points within a certain radius around a point. Points with higher local density are usually located in dense areas, while points with lower density may be located at the boundary or in sparse areas. This metric is often used to distinguish core points, boundary points, and noise points in a point cloud, especially playing an important role in the clustering and denoising processes.

[0048] With the continuous development of underwater detection technology, underwater sonar devices are increasingly widely used in culvert detection. In related technologies, an underwater unmanned submersible is used to carry a cross-section sonar for the current situation investigation of culverts. By extracting the point cloud boundary from the collected sonar data, it can help accurately identify the structure surface, thus realizing the real-time assessment of the underwater environment. The existing method is to obtain the point cloud image through the acquisition device and then manually mark the point cloud image to obtain the boundary contour of the culvert, which cannot guarantee the accuracy of the culvert boundary contour.

[0049] To solve the above problems, the embodiments of the present application provide a method, device, and medium for extracting the boundary contour of a square culvert based on sonar point cloud. The method includes: obtaining the initial sonar acquisition data inside the square culvert, where the initial sonar acquisition data includes the intensity information of multiple different scanning cross-sections; performing coordinate transformation processing on each frame of the intensity information to obtain the cross-section point cloud data corresponding to each frame of the intensity information; performing central point noise removal processing on the cross-section point cloud data to obtain the overall boundary point cloud data; performing DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain the target boundary point cloud data; performing neighborhood analysis on the target boundary point cloud data based on the local density of all points in the target boundary point cloud data to obtain multiple boundary points; determining target boundary points from all the boundary points based on a preset angle set, where the preset angle set includes multiple connection angles, each of the connection angles corresponds to at least one of the boundary points, the target boundary point is the boundary point on the same connection angle that is farthest from the point cloud center in the target boundary point cloud data, and the connection angle is the connection angle between the boundary point and the point cloud center in the target boundary point cloud data; performing linear interpolation processing on all the target boundary points to obtain the target boundary contour for the square culvert. According to the solution provided by the embodiments of the present application, by performing coordinate transformation, denoising, clustering, etc. on the initial sonar data, the target boundary contour of the square culvert is obtained, which can effectively improve the accuracy of the culvert boundary contour compared with the manual acquisition solution.

[0050] The following further elaborates on the embodiments of the present application with reference to the accompanying drawings.

[0051] Refer to Figure 1 , Figure 1 which is the step flow chart of the method for extracting the boundary contour of a square culvert based on sonar point cloud provided by an embodiment of the present application. The embodiments of the present application provide a method for extracting the boundary contour of a square culvert based on sonar point cloud, and the method includes but is not limited to the following steps:

[0052] Step S10: Obtain the initial sonar acquisition data inside the square culvert, where the initial sonar acquisition data includes the intensity information of multiple different scanning cross-sections;

[0053] Step S20: Perform coordinate transformation processing on each frame of the intensity information to obtain the cross-section point cloud data corresponding to each frame of the intensity information;

[0054] Step S30: Perform central point noise removal processing on the cross-section point cloud data to obtain the overall boundary point cloud data;

[0055] Step S40: Perform DBSCAN clustering filtering processing on the overall boundary point cloud data to obtain the target boundary point cloud data;

[0056] Step S50: Based on the local density of all points in the target boundary point cloud data, perform neighborhood analysis on the target boundary point cloud data to obtain multiple boundary points.

[0057] Step S60: Determine target boundary points from all the boundary points based on a preset angle set. The preset angle set includes multiple connection angles, and each connection angle corresponds to at least one boundary point. The target boundary point is the boundary point corresponding to the same connection angle and having the farthest distance from the point cloud center in the target boundary point cloud data. The connection angle is the angle between the boundary point and the point cloud center in the target boundary point cloud data.

[0058] Step S70: Perform linear interpolation processing on all the target boundary points to obtain the target boundary contour for the square culvert.

[0059] It can be understood that after obtaining the initial sonar acquisition data of the square culvert in this application, automated multi-stage processing is performed on the initial sonar acquisition data to obtain the target boundary contour for the square culvert, providing an accurate and effective data basis for underwater culvert detection. Specifically, in this embodiment, the cross-sectional point cloud data after coordinate transformation is subjected to central point noise removal and DBSCAN clustering filtering processing to make the boundary point cloud data more concentrated, obtaining the target boundary point cloud data. Then, the outermost target boundary points are extracted from the target boundary point cloud data by combining the local density of the points and the preset angle set (i.e., angle grouping). Finally, linear interpolation is performed on the target boundary points to smooth the boundary contour formed by the target boundary points. In this way, the point cloud data is gradually simplified and refined to obtain a more accurate target boundary contour. Compared with the manual acquisition scheme, this application can accurately and efficiently extract the pipeline boundary contour, providing accurate and efficient technical support for practical applications such as pipeline detection.

[0060] It can be understood that for step S60 in this embodiment of determining target boundary points from all the boundary points based on a preset angle set, the preset angle set corresponds to multiple connection angles. In this embodiment, all the boundary points are grouped based on the connection angles. Specifically, it can be that the boundary points located on the reference line corresponding to the connection angle are classified into the same angle grouping, or a reference region is established based on the reference line corresponding to the connection angle and the preset angle, and the boundary points located in the reference region are classified into the same angle grouping. It can be known that among any angle grouping, the point with the farthest distance from the point cloud center is the farthest boundary point, and the farthest boundary point is more consistent with the target boundary contour to be obtained in this embodiment. Therefore, determining the point with the farthest distance from the point cloud center among the respective boundary points corresponding to each angle grouping as the target boundary point can provide an effective data basis for subsequent obtaining of an accurate target boundary contour.

[0061] It should be noted that the connection angle in this embodiment is calculated according to the following formula:

[0062] ;

[0063] wherein, is the connection angle, ( ) are the coordinates of the boundary point in the plane, and ( ) are the coordinates of the point cloud center in the target boundary point cloud data.

[0064] Specifically, in some embodiments, Figure 1 step S30 includes but is not limited to the following steps:

[0065] Step S31, determining the number of point clouds in the cross-section point cloud data and the first position of each point;

[0066] Step S32, calculating the second position of the point cloud center of the cross-section point cloud data based on each first position and the number of point clouds;

[0067] Step S33, calculating the Euclidean distance between each point in the cross-section point cloud data and the point cloud center based on the second position and all the first positions;

[0068] Step S34, deleting the points corresponding to the Euclidean distances less than the first distance threshold from the cross-section point cloud data to obtain the overall boundary point cloud data.

[0069] It should be noted that in this embodiment, the second position of the point cloud center of the cross-section point cloud data is calculated based on each first position and the number of point clouds, and is obtained according to the following formula:

[0070] ;

[0071] wherein, is the second position, is the first position, is the number of point clouds, i is the index of each point in the cross-section point cloud data.

[0072] It should be noted that in this embodiment, the Euclidean distance between each point in the cross-section point cloud data and the point cloud center is calculated based on the second position and all the first positions, and is obtained according to the following formula:

[0073]

[0074] wherein, is the Euclidean distance between each point in the cross-section point cloud data and the point cloud center.

[0075] It can be understood that in this embodiment, the position of the point cloud center in the cross-section point cloud data, that is, the second position, is calculated, and the Euclidean distance between each point in the cross-section point cloud data and the point cloud center is calculated. Points corresponding to Euclidean distances less than the first distance threshold are deleted in the cross-section point cloud data, that is, points that are too close to the center are removed. Points that are too close to the center can be considered to be far from the boundary and cannot be points on the boundary contour. Thus, invalid regions can be removed, and then the overall boundary point cloud data can be obtained. It can also be to select the n points with the smallest distance for removal processing to obtain the overall boundary point cloud data, n and the value of

[0076] is 50. Figure 1 Specifically, in some embodiments,

[0077] Step S40 includes but is not limited to the following steps:

[0078] Step S41, when there are at least a preset number of points in the neighborhood of a preset radius of a reference point in the overall boundary point cloud data, points with a distance less than the preset radius from the reference point in the neighborhood are determined as core points, where the reference point is any point in the overall boundary point cloud data;

[0079] Step S42, based on the preset radius, preset number, and each core point, perform neighborhood expansion on each core point to obtain corresponding point cloud clusters, where the distance between points other than the core points in each point cloud cluster and the core point is less than the preset radius, and the number of points other than the core points in each point cloud cluster is greater than or equal to the preset number;

[0080] Step S43, delete points in the overall boundary point cloud data other than the point cloud clusters and the reference points to obtain the target boundary point cloud data.

[0080] It is understandable that in this embodiment, the DBSCAN clustering and filtering algorithm is used to remove noise points, separate discrete points, and extract dense regions from the overall boundary point cloud data. Specifically, a preset radius ε and a preset number min_samples (i.e., the minimum number of points) are used as clustering parameters. Among them, the preset radius ε determines the neighborhood range between points, and min_samples sets the minimum number of points required in a cluster. If the neighborhood of a point p_i contains at least min_samples points and the distance between these points and p_i is less than ε, then this point is considered a core point. Next, based on the preset radius, preset number, and each core point, the neighborhood of each core point is expanded until all neighboring points are assigned to the same cluster. If the number of points in the neighborhood of a point in the overall boundary point cloud data is less than min_samples and this point does not have enough neighbor points to assign it to a certain cluster, it will be marked as a noise point (labeled -1). These points are also the points in the overall boundary point cloud data other than the point cloud clusters and reference points. By deleting the noise points in the overall boundary point cloud data, clearer and more concentrated target boundary point cloud data can be obtained, reducing the interference of noise points while reducing the overall amount of point cloud data and further improving the efficiency of obtaining the target boundary contour.

[0081] It should be noted that in this embodiment, the value of the preset radius ε is 0.5, and the value of the preset number min_samples is 5.

[0082] Specifically, in some embodiments, Figure 1 Step S50 includes but is not limited to the following steps:

[0083] Step S51, calculate the reference number of adjacent points within the range of the second distance threshold for any point in the target boundary point cloud data;

[0084] Step S52, calculate the local density based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth positions of each adjacent point of any point in the target boundary point cloud data, and the reference number;

[0085] Step S53, determine the points with local density lower than the preset density threshold in the target boundary point cloud data as boundary points.

[0086] It should be noted that the local density is calculated based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth positions of each adjacent point of any point in the target boundary point cloud data, and the reference number, and is obtained according to the following formula:

[0087] ;

[0088] Among them, The position of the points in the target boundary point cloud data is the local density corresponding to the points in the target boundary point cloud data, and Ⅱ() is the indicator function is the second distance threshold is the adjacent points of the point corresponding to is the reference quantity i is the index of the point corresponding to j is the index of the point corresponding to

[0089] It should be noted that in this embodiment, the value of the number of adjacent points for calculating the local density, that is, the value of the reference quantity, is 15, and the value of the preset density threshold is 0.08

[0090] It can be understood that in this embodiment, the local density of multiple points in the target boundary point cloud data is calculated, and the boundary points are identified based on whether it is lower than the preset density threshold. It can be known that in the target boundary point cloud data, the points with local density lower than the preset density threshold are more in line with the requirements corresponding to the target boundary contour. Therefore, after calculating the local density based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth positions of the adjacent points of any point in the target boundary point cloud data, and the reference quantity, the points in the target boundary point cloud data with local density lower than the preset density threshold are determined as boundary points

[0091] Specifically, in this embodiment, in step S70, linear interpolation processing is performed on all the target boundary points to obtain the target boundary contour for the square culvert, which is obtained according to the following formula

[0092] ;

[0093] where is the position corresponding to the target boundary point is the interpolation parameter used to indicate from the linear interpolation ratio to is the position of the linearly inserted points in the target boundary contour except for the target boundary points

[0094] It can be understood that after obtaining the target boundary points, linear interpolation processing is performed on the target boundary points, that is, a certain number of points are inserted between adjacent target boundary points to further smooth the contour, and after obtaining the smoothed contour, the boundary point cloud data is obtained, making the outer boundary contour of the culvert smoother and more accurate

[0095] It should be noted that in this embodiment, the number of points inserted between adjacent boundary points is 10

[0096] As Figure 2 shown Figure 2 is a structural diagram of a control device provided by an embodiment of the present application. The present invention also provides a control device 200, including:

[0097] A processor 210, which can be implemented in ways such as a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided by the embodiments of the present application;

[0098] A memory 220, which can be implemented in forms such as a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 220 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 220 and are called by the processor 210 to execute the method for extracting the boundary contour of a square culvert based on sonar point cloud according to the embodiments of the present application;

[0099] An input / output interface 230, which is used to implement information input and output;

[0100] A communication interface 240, which is used to implement communication and interaction between this device and other devices, and can achieve communication through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.);

[0101] A bus 250, which transmits information between various components of the device (such as the processor 210, the memory 220, the input / output interface 230, and the communication interface 240);

[0102] Among them, the processor 210, the memory 220, the input / output interface 230, and the communication interface 240 achieve communication connections with each other inside the device through the bus 250.

[0103] In addition, an embodiment of the present application also provides an electronic device, including the control device 200 of the above embodiment.

[0104] In addition, an embodiment of the present application also provides a storage medium, the storage medium is a computer-readable storage medium, and the storage medium stores a computer program, and when the computer program is executed by a processor, it implements the above-mentioned method for extracting the boundary contour of a square culvert based on sonar point cloud.

[0105] As a non-transitory computer-readable storage medium, the memory can be used to store non-transitory software programs and non-transitory computer-executable programs. In addition, the memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory may optionally include a memory remotely located relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0106] Those of ordinary skill in the art can understand that all or some of the steps and systems disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium generally includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0107] The above is a specific description of the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Those skilled in the art can also make various equivalent deformations or substitutions without departing from the spirit of the present invention, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present invention.

Claims

1. A method for extracting the boundary contour of a square culvert based on sonar point cloud, characterized in that: include: Acquire initial sonar acquisition data in the square culvert, wherein the initial sonar acquisition data includes intensity information of multiple frames of different scanning sections; Performing coordinate conversion processing on the intensity information of each frame to obtain cross-sectional point cloud data corresponding to the intensity information of each frame; The cross-section point cloud data is processed to remove center point noise to obtain overall boundary point cloud data; Performing DBSCAN clustering filtering on the overall boundary point cloud data to obtain target boundary point cloud data; Based on the local density of all points in the target boundary point cloud data, performing neighborhood analysis on the target boundary point cloud data to obtain a plurality of boundary points; Determine a target boundary point from all the boundary points based on a preset angle set, wherein the preset angle set includes a plurality of line angles, each of the line angles corresponds to at least one boundary point, the target boundary point is the boundary point that is farthest from the point cloud center in the target boundary point cloud data corresponding to the same line angle, and the line angle is the line angle between the boundary point and the point cloud center in the target boundary point cloud data; Performing linear interpolation processing on all the target boundary points to obtain a target boundary contour for the square culvert; Wherein, based on the local density of all points in the target boundary point cloud data, the target boundary point cloud data is subjected to neighborhood analysis to obtain a plurality of boundary points, including: Calculate the reference number of neighboring points of any point in the target boundary point cloud data within the range of a second distance threshold; Calculating the local density based on the second distance threshold, a third position of any point in the target boundary point cloud data, fourth positions of each neighboring point of any point in the target boundary point cloud data, and the reference quantity; Determine the point in the target boundary point cloud data whose local density is lower than a preset density threshold as the boundary point; The local density is calculated based on the second distance threshold, the third position of any point in the target boundary point cloud data, the fourth position of each adjacent point of any point in the target boundary point cloud data, and the reference quantity, and is obtained according to the following formula: ； in, is the position of the point in the target boundary point cloud data, is the local density corresponding to the points in the target boundary point cloud data, Ⅱ () is the indicator function, is the second distance threshold, for The adjacent points of the corresponding point, is the reference quantity, i for The index of the corresponding point, j for The index of the corresponding point.

2. The method for extracting the boundary contour of a square culvert based on sonar point cloud according to claim 1 is characterized in that: The cross-section point cloud data is processed to remove the center point noise to obtain the overall boundary point cloud data, including: Determine the number of point cloud points in the cross-section point cloud data and the first position of each point; Calculate the second position of the point cloud center of the cross-section point cloud data based on each of the first positions and the number of point cloud points; Based on the second position and all the first positions, calculating the Euclidean distance between each point in the cross-section point cloud data and the center of the point cloud; Points corresponding to the Euclidean distance less than the first distance threshold are deleted from the cross-section point cloud data to obtain the overall boundary point cloud data.

3. The method for extracting the boundary contour of a square culvert based on sonar point cloud according to claim 2 is characterized in that: The second position of the point cloud center of the cross-section point cloud data is calculated based on each of the first positions and the number of point cloud points, and is obtained according to the following formula: ； in, For the second position, is the first position, is the number of points in the point cloud, i is the index of each point in the cross-section point cloud data.

4. The method for extracting the boundary contour of a square culvert based on sonar point cloud according to claim 3 is characterized in that: The overall boundary point cloud data is subjected to DBSCAN clustering filtering processing to obtain target boundary point cloud data, including: When a neighborhood of a preset radius of a reference point in the overall boundary point cloud data includes at least a preset number of points, a point in the neighborhood whose distance from the reference point is less than the preset radius is determined as a core point, wherein the reference point is any point in the overall boundary point cloud data; Based on the preset radius, the preset number and each of the core points, a neighborhood expansion is performed on each of the core points to obtain a corresponding point cloud cluster, wherein the distance between the points other than the core point in each of the point cloud clusters and the core point is less than the preset radius, and the number of points other than the core point in each of the point cloud clusters is greater than or equal to the preset number; The points other than the point cloud cluster and the reference point in the overall boundary point cloud data are deleted to obtain the target boundary point cloud data.

5. The method for extracting the boundary contour of a square culvert based on sonar point cloud according to claim 1 is characterized in that: The connection angle is calculated according to the following formula: ； in, is the angle of the connecting line, ( ) is the coordinate of the boundary point in the plane, ( ) are the coordinates of the point cloud center in the target boundary point cloud data.

6. The method for extracting the boundary contour of a square culvert based on sonar point cloud according to claim 1, characterized in that: Linear interpolation is performed on all the target boundary points to obtain the target boundary contour for the square culvert, which is obtained according to the following formula: ； in, is the position corresponding to the target boundary point, is the interpolation parameter, Used to indicate arrive The linear interpolation ratio of is the position of a point linearly inserted in the target boundary contour except the target boundary point.

7. An electronic device, characterized in that: It includes at least one control processor and a memory for communicating with the at least one control processor; the memory stores instructions that can be executed by the at least one control processor, and the instructions are executed by the at least one control processor so that the at least one control processor can execute the square culvert boundary contour extraction method based on sonar point cloud as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the method for extracting the boundary contour of a square culvert based on sonar point cloud as described in any one of claims 1 to 6.

Images