A method and system for quantitative automatic identification of three-dimensional point cloud of flat steel deformation on the surface of earth-rock dams

Abstract

This invention discloses a method and system for the quantitative automatic identification of three-dimensional point cloud deformation of flat steel on the surface of an earth-rock dam: Based on the overall three-dimensional laser point cloud data of the earth-rock dam, three-dimensional laser point cloud data of the flat steel on the dam surface is extracted; the three-dimensional laser point cloud data of the flat steel is projected from a three-dimensional plane to a two-dimensional plane to obtain the projected two-dimensional point cloud data of the flat steel; based on the principle of the limit circle, boundary point clouds are extracted from the projected two-dimensional point cloud data of the flat steel to form a directed two-dimensional boundary line; according to the projection mapping relationship, the three-dimensional points in the three-dimensional laser point cloud data of the flat steel corresponding to each two-dimensional point on the two-dimensional boundary line are obtained to form the directed three-dimensional flat steel point cloud boundary line; the magnitude of the deformation is obtained; and the direction of the deformation is obtained. This invention extracts the point cloud data of the flat steel entity from the overall point cloud data, automatically identifies the deformed flat steel region, and quantitatively analyzes the magnitude and direction of the flat steel deformation.

Description

Technical Field

[0001] This invention relates to the field of deformation monitoring technology, and in particular to a three-dimensional point cloud quantitative automatic identification method and system for the deformation of flat steel on the surface of earth-rock dams based on the limit circle. Background Technology

[0002] Conventional deformation monitoring technology for earth-rock dams: The dam surface deformation is monitored primarily using a measurement robot monitoring system, supplemented by a dam GNSS monitoring system. For monitoring the bulge of the flat steel on the dam surface, relying solely on traditional measurement methods—measuring tapes or rangefinders—results in low accuracy, a small measurement range, and an inability to directly reflect the bulge status of the flat steel. Therefore, 3D laser scanning technology demonstrates its advantages. However, in specific projects, given the massive amount of point cloud data, the rapid, automated, and quantitative identification and analysis of the deformation of the flat steel point cloud attached to the dam surface is an urgent problem to be solved.

[0003] The object of this invention (a 10cm wide flat steel bar on the surface of an earth-rock dam) is quite unique, being an unconventional monitoring object. Furthermore, the classification and identification of the 3D laser point cloud is quite challenging. Besides studying relevant theoretical knowledge on deformation monitoring, there are currently no suitable papers or monographs specifically addressing the extraction of this 10cm wide flat steel bar from the surface of earth-rock dams, while simultaneously quantitatively analyzing the magnitude and direction of deformation. Because the point cloud data is massive, current methods typically involve a significant amount of time spent on manual extraction followed by comparative analysis, which is time-consuming, labor-intensive, and highly impractical. Summary of the Invention

[0004] This invention addresses the problems and shortcomings of existing technologies by providing a novel method and system for quantitative automatic identification of three-dimensional point cloud deformation of flat steel on the surface of earth-rock dams.

[0005] The present invention solves the above-mentioned technical problems through the following technical solution:

[0006] This invention provides a method for quantitative automatic identification of three-dimensional point cloud deformation of flat steel on the surface of earth-rock dams, characterized by comprising the following methods:

[0007] S1. Based on the overall three-dimensional laser point cloud data of the earth-rock dam surface, extract the three-dimensional laser point cloud data of the flat steel on the earth-rock dam surface;

[0008] S2. Project the extracted three-dimensional laser point cloud data of the flat steel from the three-dimensional plane to the two-dimensional XOY plane to obtain the projected two-dimensional point cloud data of the flat steel.

[0009] S3. Based on the principle of the limit circle, the boundary point cloud is extracted from the two-dimensional point cloud data of the projected flat steel to form a directed two-dimensional boundary line.

[0010] S4. According to the projection mapping relationship, obtain the three-dimensional points in the three-dimensional laser point cloud data of the flat steel corresponding to each two-dimensional point on the two-dimensional boundary line. These three-dimensional points are connected in sequence to form the directed three-dimensional flat steel point cloud boundary line.

[0011] S5. Obtain the magnitude of deformation: Compare the boundary line of the directional 3D flat steel point cloud with the outer dimensions of the flat steel when it is not deformed to obtain the amount of deformation caused by the stretching of the flat steel. Compare the actual projected length of the flat steel with the projected length of the flat steel when it is not deformed to obtain the bending deformation of the flat steel.

[0012] S6. Obtain the deformation direction: Compare the line segment projected from the centerline of the flat steel onto the plane with the line segment projected from the centerline of the flat steel onto the plane when it is not deformed. Based on the relative positional relationship of the two line segments on the projection plane, determine the deformation direction of the flat steel.

[0013] Preferably, in step S1, the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam is extracted sequentially using coarse extraction, fine extraction, and optimized extraction methods. The coarse extraction method involves setting a threshold of 6cm on both sides of the center line of the flat steel at the initial installation coordinate position of the flat steel for buffer analysis to initially extract the three-dimensional laser point cloud data of the flat steel. The fine extraction method involves performing a fine extraction based on the color white of the flat steel on the basis of the first coarse extraction. The optimized extraction method involves performing an optimized extraction based on the reflection intensity on the basis of the second coarse extraction.

[0014] The present invention also provides a three-dimensional point cloud quantitative automatic recognition system for the deformation of flat steel on the surface of earth-rock dams, characterized in that it includes a first extraction module, a projection module, a second extraction module, a mapping module, a first acquisition module, and a second acquisition module;

[0015] The first extraction module is used to extract the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam based on the overall three-dimensional laser point cloud data of the earth-rock dam surface;

[0016] The projection module is used to project the extracted three-dimensional laser point cloud data of the flat steel from three dimensions to a two-dimensional XOY plane to obtain the projected two-dimensional point cloud data of the flat steel.

[0017] The second extraction module is used to extract the boundary point cloud from the projected two-dimensional point cloud data of flat steel based on the principle of the limit circle to form a directed two-dimensional boundary line;

[0018] The mapping module is used to obtain the three-dimensional points in the three-dimensional laser point cloud data of the flat steel corresponding to each two-dimensional point on the two-dimensional boundary line according to the projection mapping relationship. These three-dimensional points are connected in sequence to form the directed three-dimensional flat steel point cloud boundary line.

[0019] The first acquisition module is used to compare the boundary line of the directional three-dimensional flat steel point cloud with the outer dimensions of the flat steel when it is not deformed, to obtain the amount of deformation of the flat steel due to stretching, and to compare the actual projected length of the flat steel with the projected length of the flat steel when it is not deformed, to obtain the amount of bending deformation of the flat steel.

[0020] The second acquisition module is used to compare the line segment of the center line of the flat steel three-dimensional laser point cloud data projected onto the plane with the line segment of the center line of the flat steel when it is not deformed, and to determine the deformation direction of the flat steel based on the relative positional relationship of the two line segments on the projection plane.

[0021] Preferably, the first extraction module is used to extract the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam by sequentially using coarse extraction, fine extraction and optimized extraction methods. The coarse extraction method is to perform buffer analysis by setting a threshold of 6cm on both sides of the center line of the flat steel at the initial installation coordinate position of the flat steel, and initially extract the three-dimensional laser point cloud data of the flat steel. The fine extraction method is to perform a fine extraction based on the color white of the flat steel on the basis of the first coarse extraction. The optimized extraction method is to perform an optimized extraction based on the reflection intensity on the basis of the second coarse extraction.

[0022] Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of the present invention.

[0023] The positive and progressive effects of this invention are as follows:

[0024] (1) In terms of time efficiency, the existing methods, which involve manual extraction and analysis, are inefficient. Multiple experiments have shown that a single data extraction process takes approximately 10 hours for one person familiar with the process. Using this patented technology, while ensuring accuracy and completeness, the process is completed in less than 5 minutes using multi-threading. Therefore, this invention significantly improves efficiency compared to existing technologies. In terms of accuracy, the data results fully meet the needs of actual production, in accordance with relevant engineering measurement standards.

[0025] (2) From an economic perspective, since it serves the deformation monitoring of earth-rock dam hydropower stations, it overcomes the predicament of conventional deformation monitoring which only covers the whole area. By analyzing the deformation of the dam as a whole, the magnitude and direction of deformation at any location at multiple time points can be obtained. The data of the flat steel on the dam surface (which strengthens the stability of the dam) can also be efficiently extracted using this patented technology, and the magnitude and direction of the deformation of the flat steel can be obtained. The data can be fed back to the power plant in a timely manner, and timely countermeasures can be taken to ensure the safe and stable operation of the power plant and the continuous and stable output of power generation. This indirectly brings huge safety guarantees and economic benefits to the power plant and society.

[0026] (3) From the perspective of environmental protection, the use of this patented technology and the use of three-dimensional laser scanning technology to monitor the deformation of earth-rock dam surface will not generate additional garbage or waste at the construction site. One data acquisition device can be connected to a laptop computer, which is efficient and fast. The environmental protection concept runs through the entire operation process. Attached Figure Description

[0027] Figure 1 This is a flowchart of a method for quantitative automatic identification of three-dimensional point cloud deformation of flat steel on the surface of an earth-rock dam, according to a preferred embodiment of the present invention.

[0028] Figure 2 This is a schematic diagram of the three-dimensional laser point cloud data of the flat steel according to a preferred embodiment of the present invention.

[0029] Figure 3 This is a schematic diagram of the boundary line extracted after the flat steel is projected according to a preferred embodiment of the present invention.

[0030] Figure 4 This is a schematic diagram of the structure of a three-dimensional point cloud quantitative automatic identification system for deformation of flat steel on the surface of an earth-rock dam, according to a preferred embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] like Figure 1 As shown in the figure, this embodiment provides a method for quantitative automatic identification of three-dimensional point cloud deformation of flat steel on the surface of earth-rock dam, which includes the following methods:

[0033] Step 101: Based on the overall three-dimensional laser point cloud data of the earth-rock dam surface, extract the three-dimensional laser point cloud data of the flat steel on the earth-rock dam surface (see...). Figure 2 ).

[0034] In step 101, the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam is extracted by coarse extraction, fine extraction and optimized extraction in sequence. The overall three-dimensional laser point cloud data of the earth-rock dam surface includes information such as coordinate position, true color RGB, and reflection intensity.

[0035] The first coarse extraction is performed. Because the position of the flat steel on the dam surface is fixed, although there will be positional shifts due to deformation, a preliminary extraction can be conducted based on the initial installation coordinates. The specific steps are as follows: The actual dimensions of the flat steel are a long strip-shaped entity 10cm wide and 1cm thick, with a fixed initial installation position. The actual collected 3D point cloud data can be analyzed using a buffer zone with a 6cm threshold set to the left and right sides of the flat steel's centerline at the designed installation position, allowing for the preliminary extraction of the flat steel data.

[0036] Second fine extraction. Since the surface of the flat steel on the dam face is coated with a layer of white pigment, the point cloud data of the flat steel can be finely extracted according to true color RGB. The specific steps are: based on the first coarse extraction, a fine extraction is performed according to the white color of the flat steel.

[0037] The third round of optimization extraction can be performed. Based on the characteristic that the reflection intensity of the flat steel point cloud data is slightly stronger than that of the surrounding stones, the extraction accuracy of the flat steel point cloud can be further optimized. The specific steps are: based on the second coarse extraction, perform a second round of optimization extraction according to the reflection intensity (the reflection intensity of the flat steel point cloud data is slightly stronger than that of the surrounding stones).

[0038] Step 102: Project the extracted three-dimensional laser point cloud data of the flat steel from the three-dimensional plane to the two-dimensional XOY plane to obtain the projected two-dimensional point cloud data of the flat steel.

[0039] Step 103: Based on the principle of the limiting circle, extract the boundary point cloud from the projected two-dimensional point cloud data of the flat steel to form a directed two-dimensional boundary line (see...). Figure 3 ).

[0040] Step 104: Obtain the three-dimensional points in the three-dimensional laser point cloud data of the flat steel corresponding to each two-dimensional point on the two-dimensional boundary line according to the projection mapping relationship. These three-dimensional points are connected in sequence to form the directed three-dimensional flat steel point cloud boundary line.

[0041] Step 105: Obtain the magnitude of deformation: Compare the boundary line of the directional 3D flat steel point cloud with the outer dimensions of the flat steel when it is not deformed to obtain the amount of deformation caused by the stretching of the flat steel. Compare the actual projected length of the flat steel with the projected length of the flat steel when it is not deformed to obtain the bending deformation of the flat steel.

[0042] Step 106: Obtain the deformation direction: Compare the line segment projected from the centerline of the flat steel onto the plane with the line segment projected from the centerline of the flat steel onto the plane when it is not deformed. Based on the relative positional relationship of the two line segments on the projection plane, determine the deformation direction of the flat steel.

[0043] like Figure 4As shown in the figure, this embodiment also provides a three-dimensional point cloud quantitative automatic recognition system for the deformation of flat steel on the surface of earth-rock dam, which includes a first extraction module 1, a projection module 2, a second extraction module 3, a mapping module 4, a first acquisition module 5, and a second acquisition module 6.

[0044] The first extraction module 1 is used to extract the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam based on the overall three-dimensional laser point cloud data of the earth-rock dam surface, and to extract the three-dimensional laser point cloud data of the flat steel on the earth-rock dam surface by using coarse extraction, fine extraction and optimized extraction methods in sequence.

[0045] The coarse extraction method involves setting a 6cm threshold on both sides of the center line of the flat steel at the initial installation coordinate position for buffer analysis, and initially extracting the three-dimensional laser point cloud data of the flat steel. The fine extraction method involves performing a fine extraction based on the color white of the flat steel, based on the first coarse extraction. The optimized extraction method involves performing an optimized extraction based on the reflection intensity, based on the second coarse extraction.

[0046] The projection module 2 is used to project the extracted three-dimensional laser point cloud data of the flat steel from three dimensions to a two-dimensional XOY plane to obtain the projected two-dimensional point cloud data of the flat steel.

[0047] The second extraction module 3 is used to extract the boundary point cloud from the projected two-dimensional point cloud data of the flat steel based on the principle of the limit circle to form a directed two-dimensional boundary line.

[0048] The mapping module 4 is used to obtain the three-dimensional points in the three-dimensional laser point cloud data of the flat steel corresponding to each two-dimensional point on the two-dimensional boundary line according to the projection mapping relationship. These three-dimensional points are connected in sequence to form the directed three-dimensional flat steel point cloud boundary line.

[0049] The first acquisition module 5 is used to compare the boundary line of the directional three-dimensional flat steel point cloud with the outer dimensions of the flat steel when it is not deformed, to obtain the amount of deformation of the flat steel due to stretching, and to compare the actual projected length of the flat steel with the projected length of the flat steel when it is not deformed, to obtain the amount of bending deformation of the flat steel.

[0050] The second acquisition module 6 is used to compare the line segment of the center line of the flat steel three-dimensional laser point cloud data projected onto the plane with the line segment of the center line of the flat steel when it is not deformed, and to determine the deformation direction of the flat steel based on the relative position relationship of the two line segments on the projection plane.

[0051] Regarding the necessary experimental conditions and methods: ① The collected point cloud data of the earth-rock dam surface must be complete, and true color data must be collected on a clear day; ② There are requirements for data density: the point cloud interval must be less than or equal to 5cm, and it is necessary to ensure that the point cloud density is high enough and there is enough data; ③ The computer should run in a multi-core, multi-threaded mode, and the memory should be more than 4GB.

[0052] This invention, guided by the basic theory of conventional deformation monitoring methods, and through mathematical derivation and computer technology, extracts point cloud data of flat steel entities from a massive amount of point cloud data, automatically identifies deformed flat steel regions, and quantitatively analyzes the magnitude and direction of the deformation of the flat steel.

[0053] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A method for quantitatively and automatically identifying three-dimensional point cloud deformation of earth-rockfill dam surface flat steel, characterized in that, The method comprises the following steps: S1, based on the three-dimensional laser point cloud data of the overall surface of the earth-rock dam, three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam is extracted; S2, the three-dimensional laser point cloud data of the flat steel extracted is projected from three-dimensional to two-dimensional XOY plane to obtain the two-dimensional point cloud data of the flat steel after projection; S3, based on the principle of limit circle, the boundary point cloud is extracted from the two-dimensional point cloud data of the flat steel after projection to form a directed two-dimensional boundary line; S4, according to the projection mapping relationship, the three-dimensional point corresponding to each two-dimensional point on the two-dimensional boundary line in the three-dimensional laser point cloud data of the flat steel is obtained, and the three-dimensional points are sequentially connected to form a directed three-dimensional flat steel point cloud boundary line; S5, the deformation amount is obtained: by comparing the directed three-dimensional flat steel point cloud boundary line with the peripheral size of the flat steel before deformation, the deformation amount of the flat steel stretched due to deformation is obtained, and by comparing the actual projection length of the flat steel with the projection length of the flat steel before deformation, the bending deformation amount of the flat steel is obtained; S6, the deformation direction is obtained: by comparing the line segment of the center line projection of the three-dimensional laser point cloud data of the flat steel to the plane with the line segment of the center line projection of the flat steel before deformation to the plane, the deformation direction of the flat steel is judged according to the relative position relationship of the two line segments on the projection plane; In the step S1, the three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam is extracted by sequentially adopting rough extraction, fine extraction and optimization extraction methods, the rough extraction method is to set a threshold of 6cm on the left and right sides of the flat steel center line according to the coordinate position of the initial installation of the flat steel for buffer zone analysis, the three-dimensional laser point cloud data of the flat steel is preliminarily extracted, the fine extraction method is to carry out fine extraction once according to the color white of the flat steel on the basis of rough extraction, and the optimization extraction method is to carry out optimization extraction once according to the reflection intensity on the basis of fine extraction.

2. A quantitative automatic identification system for three-dimensional point cloud of earth-rockfill dam surface flat steel deformation, characterized in that, It comprises a first extraction module, a projection module, a second extraction module, a mapping module, a first acquisition module and a second acquisition module; The first extraction module is used for extracting three-dimensional laser point cloud data of the flat steel on the surface of the earth-rock dam based on the three-dimensional laser point cloud data of the overall surface of the earth-rock dam; The projection module is used for projecting the three-dimensional laser point cloud data of the flat steel extracted from three-dimensional to two-dimensional XOY plane to obtain the two-dimensional point cloud data of the flat steel after projection; The second extraction module is used for extracting the boundary point cloud from the two-dimensional point cloud data of the flat steel after projection to form a directed two-dimensional boundary line based on the principle of limit circle; The mapping module is used for obtaining the three-dimensional point corresponding to each two-dimensional point on the two-dimensional boundary line in the three-dimensional laser point cloud data of the flat steel according to the projection mapping relationship, and the three-dimensional points are sequentially connected to form a directed three-dimensional flat steel point cloud boundary line; The first acquisition module is used for comparing the directed three-dimensional flat steel point cloud boundary line with the peripheral size of the flat steel before deformation to obtain the deformation amount of the flat steel stretched due to deformation, and comparing the actual projection length of the flat steel with the projection length of the flat steel before deformation to obtain the bending deformation amount of the flat steel; The second acquisition module is used for comparing the line segment of the center line projection of the three-dimensional laser point cloud data of the flat steel to the plane with the line segment of the center line projection of the flat steel before deformation to the plane, and judging the deformation direction of the flat steel according to the relative position relationship of the two line segments on the projection plane. The first extraction module is used for extracting the three-dimensional laser point cloud data of the earth and rockfill dam surface flat steel in turn by using rough extraction, fine extraction and optimization extraction modes, the rough extraction mode is to set a 6cm threshold for buffer zone analysis on the left and right sides of the flat steel center line according to the coordinate position of the initial installation of the flat steel, and the three-dimensional laser point cloud data of the flat steel is preliminarily extracted, the fine extraction mode is to carry out fine extraction once according to the color white of the flat steel on the basis of rough extraction, and the optimization extraction mode is to carry out optimization extraction once according to the reflection intensity on the basis of fine extraction.

Images