CN112763700B - System and method for quality inspection of finished concrete prefabricated beams and construction of digital solid model - Google Patents

Abstract

The invention relates to a system and method for quality inspection of finished concrete prefabricated beams and construction of a digital solid model. The overall system includes a detection frame, a visual information acquisition array, a sample carrying component and a visual information processing component. The visual information acquisition array is arranged on the detection frame to form a circular visual field along the detection frame. The visual information acquisition array includes a plurality of laser ranging units and optical camera units, and each of the laser ranging units and optical camera units is respectively It is wirelessly or wiredly connected with the visual information processing component. Compared with the prior art, the present invention constructs all the safety information of the beam as a digital entity model through the coupling of the detection frame, visual information acquisition array, sample carrying component and visual information processing component, and realizes the prefabricated beam section to be tested in terms of quality inspection. The detection of cracks, deformation and other apparent defects has the advantages of rapidity, standardization, precision and intelligence.

Description

System and method for quality inspection of finished concrete prefabricated beams and construction of digital solid model

technical field

The invention relates to the field of quality inspection of building materials, in particular to a system and method for quality inspection of finished concrete prefabricated beams and construction of a digital entity model.

Background technique

The prefabrication process of concrete components is the process of pre-processing concrete components of buildings or structures in factories or construction sites. Prefabricated concrete components are used for assembly construction, which has the advantages of saving labor, overcoming seasonal influences, and facilitating construction all year round. Promoting the use of prefabricated concrete components is one of the important ways to realize the industrialization of construction.

Because the construction quality of prefabricated reinforced concrete components and the loss during transportation are difficult to guarantee, cracks, honeycomb pockmarks, segregation, damage and other diseases often occur, which in turn affect the overall safety of the structure. The traditional method uses manual inspection by workers in the factory, which has the disadvantages of time-consuming, labor-intensive, highly subjective, and incomplete inspection results. Therefore, it is urgent to develop a device that can realize the quality control and information management of prefabricated beam sections, so as to effectively support the intelligent operation and maintenance and performance monitoring of the structure during its service period.

Contents of the invention

The purpose of the present invention is to provide concrete prefabricated beam finished product quality inspection and digital solid model construction system and method in order to overcome the defect existing in the above-mentioned prior art, realize crack, deformation and other apparent defects of the prefabricated beam section to be tested in quality inspection Compared with the traditional method, it has the advantages of rapidity, standardization, precision and intelligence, which ensures the reasonable and safe use of the beam end to be tested.

The purpose of the present invention can be achieved through the following technical solutions:

The first purpose of the present invention is to protect a quality inspection system for finished concrete prefabricated beams, including a detection frame, a visual information acquisition array, a sample carrying component, and a visual information processing component;

The visual information acquisition array is arranged on the detection frame to form a ring-shaped visual field along the detection frame. The visual information acquisition array includes a plurality of laser ranging units and optical camera units, and each of the laser ranging units and optical cameras The units are respectively connected to the visual information processing components by wireless or wired communication;

The sample carrying component can transport the finished concrete prefabricated beam through the detection frame at a specific speed;

During the dynamic transportation process of the finished concrete precast beam, the sparse point cloud model of the finished concrete precast beam is obtained through the coupling of the laser ranging unit and the visual information processing component, and the concrete precast beam is obtained through the coupling of the optical camera unit and the visual information processing component The image information of the finished product, the visual information processing component realizes the splicing of image information based on the sparse point cloud model, and obtains the apparent panorama of the finished concrete prefabricated beam. The features are extracted to form the apparent eigenvector of the beam segment, and the quality evaluation of the finished concrete prefabricated beam is realized by the apparent eigenvector of the beam segment.

Further, the detection frame is an annular truss structure.

Further, the sample carrying component is a horizontal transport device that runs through the detection frame.

Further, the sample carrying component is one of rail-type transportation equipment, conveyor belt-type transportation equipment, and industrial handling robots.

Further, the visual information acquisition array further includes a synchronous collector, and the synchronous collector is electrically connected to each laser ranging unit and optical camera unit.

Further, the laser ranging unit is a multi-point infrared ranging sensor, and the optical camera unit is a multi-eye camera.

Further, the visual information processing component includes a microprocessor, a main memory and an auxiliary memory connected to the microprocessor through a main line, a signal transmitter is connected to the main line, and the signal transmitter is wireless with the synchronous collector or a wired connection.

Further, the finished product quality inspection system for prefabricated concrete beams also includes a carrier component control unit and a human-computer interaction interface, and the human-computer interaction interface is electrically connected to the carrier component control unit.

Further, the sample carrying component is electrically connected to the carrying component control unit, so as to realize the control of the conveying path and conveying speed of the finished concrete prefabricated beam.

The second object of the present invention is to protect a construction system for a digital solid model of a prefabricated concrete beam, including the above-mentioned detection frame, visual information acquisition array, sample carrying component and visual information processing component, wherein the optical camera unit and the visual information Process the coupling of components to obtain the image information of the finished concrete prefabricated beam, and further obtain the dense point cloud model;

The visual information processing component combines the sparse point cloud model and the dense point cloud model to obtain the digital entity model of the finished concrete prefabricated beam.

The third object of the present invention is to protect a method for quality inspection of finished concrete prefabricated beams and construction of a digital solid model, including the following steps:

S1: The finished concrete prefabricated beam is transported through the detection frame at a specific speed by the sample carrying component, and at the same time, the laser distance measuring unit and the optical camera unit on the detection frame are coupled with the visual information processing component to obtain the finished product of the concrete prefabricated beam. Sparse point cloud model, dense point cloud model and image information of all parts;

S2: Use deep neural network point cloud segmentation technology to realize the matching of sparse point cloud model and dense point cloud model, and complete the construction of As-Built BIM model. At the same time, through the sparse control points and feature points in the sparse point cloud model, all beam sections The image information of the parts is spliced to obtain the apparent panorama;

S3: Based on the apparent panorama, the deep semantic segmentation network method is used to extract the apparent features in the apparent panorama to form the apparent feature vector of the beam segment;

S4: Based on the internal and external relationship between the optical camera and the point cloud model, the accurate projection and transformation of the two-dimensional image in the panorama data to the As-Built BIM model is realized, and the digital entity model of the beam segment is constructed.

Compared with the prior art, the present invention has the following technical advantages:

1) The present invention proposes a fast and standardized factory inspection system for finished prefabricated beams. Through the coupling of inspection framework, visual information acquisition array, sample carrying components and visual information processing components, all safety information of beams is constructed as a digital entity model, In terms of quality inspection, it realizes the detection of cracks, deformation and other apparent defects on the prefabricated beam section to be tested. Compared with traditional equipment, it has the advantages of rapidity, standardization, precision, and intelligence, ensuring the reasonable and safe use of the beam end to be tested. .

2) This technical solution uses the multi-eye camera and multi-point infrared ranging module carried by the ring-shaped truss-type detection frame. Under the traction of the walking system, the detection frame is driven to scan the prefabricated beam in all directions, and the visual appearance of the prefabricated detection beam is obtained by scanning. View images and 3D point cloud models. Based on technologies such as disease identification and deep learning, the apparent disease in the apparent image is identified and located, and finally the beam section quality inspection results are fed back to the user. At the same time, according to the detection results and the three-dimensional point cloud model, the digital solid model of the test beam section is constructed in the core data console to realize the digital twin of the prefabricated beam section.

3) The digital entity model constructed by this technical solution through the quality inspection system for finished concrete precast beams can directly serve life cycle engineering information management, operation and maintenance evaluation and decision support, and this technology can be used for quality control of finished precast beams in the factory and information management, and effectively support intelligent operation and maintenance and performance monitoring during the service period of the structure.

Description of drawings

Figure 1 is a schematic structural diagram of the detection framework, visual information acquisition array, and sample carrying components in the technical solution;

Fig. 2 is the overall structural diagram of the finished product quality inspection system of concrete prefabricated beams in the technical scheme;

Figure 3 is a flow chart of the technical route of the technical solution.

In the figure: 1. Detection frame, 2. Visual information acquisition array, 3. Sample carrying component, 4. Visual information processing component, 5. Finished concrete prefabricated beam, 6. Carrying component control unit, 7. Database unit, 8. Human machine interface.

Detailed ways

The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

Example

The finished product quality inspection system for prefabricated concrete beams in this embodiment includes a detection frame 1, a visual information acquisition array 2, a sample carrying component 3 and a visual information processing component 4, see Fig. 1 and Fig. 2 .

During specific implementation, the detection frame 1 is an annular truss structure. The visual information acquisition array 2 is arranged on the detection frame 1 to form a circular visual field along the detection frame 1. The visual information acquisition array 2 includes a plurality of laser ranging units and optical camera units, and each laser ranging unit and optical camera unit are respectively It is connected with the visual information processing component 4 by wireless or wired communication.

During the dynamic transportation process of the finished concrete prefabricated beam 5, the sparse point cloud model of the finished concrete prefabricated beam 5 is obtained through the coupling of the laser ranging unit and the visual information processing component 4, and the coupling of the optical camera unit and the visual information processing component 4 The image information of the finished concrete prefabricated beam 5 is obtained, and the visual information processing component 4 realizes splicing of the image information based on the sparse point cloud model to obtain the apparent panorama of the finished concrete prefabricated beam 5. The visual information processing component 4 adopts a deep semantic segmentation network to The apparent features of the structure are extracted to form the apparent eigenvector of the beam segment, and the quality evaluation of the finished concrete prefabricated beam 5 is realized with the apparent eigenvector of the beam segment.

The sample carrying assembly 3 is capable of transporting the finished concrete prefabricated beam 5 through the testing frame 1 at a specific speed. The sample carrying component 3 is a horizontal transport device that runs through the detection frame 1 . The sample carrying component 3 is one of rail-type transportation equipment, conveyor belt-type transportation equipment, and industrial handling robots.

The visual information acquisition array 2 also includes a synchronous collector, which is electrically connected to each laser ranging unit and optical camera unit, so as to ensure time-based synchronous acquisition by the laser ranging unit and the optical camera unit. In the specific selection, the laser ranging unit is a multi-point infrared ranging sensor, and the optical camera unit is a multi-eye camera.

The visual information processing component 4 includes a microprocessor, a main memory and an auxiliary memory connected to the microprocessor through a main line, a signal transmitter is connected to the main line, and the signal transmitter is connected to the synchronous collector wirelessly or by wire. The visual information processing component 4 is a computer component including an x86 or ARM architecture CPU.

The finished product quality inspection system for prefabricated concrete beams also includes a carrier component control unit 6 and a human-computer interaction interface 8 , and the human-computer interaction interface 8 is electrically connected to the carrier component control unit 6 . The sample carrying component 3 is electrically connected with the carrying component control unit 6, so as to realize the control of the conveying path and conveying speed of the finished concrete prefabricated beam 5. During specific implementation, the human-computer interaction interface 8 is a multi-touch touch display screen, which is connected to the I/O interface in the control unit 6 of the carrier component, and the user can select any preset concrete prefabricated beam finished product 5 by touch. The output program corresponds to the specific conveying path and conveying speed. The carrier component control unit 6 is a computer component including an x86 or ARM architecture CPU.

The construction system of the digital solid model of the finished concrete prefabricated beam, including the above-mentioned detection framework 1, visual information acquisition array 2, sample carrying component 3 and visual information processing component 4, wherein the concrete is obtained through the coupling of the optical camera unit and the visual information processing component 4 The image information of the prefabricated beam finished product 5 is further obtained from the dense point cloud model, and the visual information processing component 4 combines the sparse point cloud model and the dense point cloud model to obtain the digital entity model of the concrete prefabricated beam finished product 5 .

Through the multi-eye camera and multi-point infrared ranging module carried by the ring truss detection frame, under the traction of the walking system, the detection frame is driven to scan the prefabricated beam in all directions, and the visual appearance image and three-dimensional image of the prefabricated detection beam can be obtained by scanning. Point cloud model. Based on technologies such as disease identification and deep learning, the apparent disease in the apparent image is identified and located, and finally the beam section quality inspection results are fed back to the user. At the same time, according to the detection results and the three-dimensional point cloud model, the digital solid model of the test beam section is constructed in the core data console to realize the digital twin of the prefabricated beam section.

Through the coupling of detection frame 1, visual information acquisition array 2, sample carrying component 3 and visual information processing component 4, all the safety information of the beam is constructed as a digital solid model, and in terms of quality inspection, cracks, deformation and Compared with traditional equipment, the detection of other apparent defects has the advantages of rapidity, standardization, precision, and intelligence, which ensures the reasonable and safe use of the beam end to be tested.

During specific implementation, the overall system also includes a database unit 7, which is a large-scale storage device with an independent controller and cache, and it is connected with the visual information processing component 4 and the human-computer interaction interface 8 at the same time to realize the visual information processing component. 4 Generate various data dumps, process and integrate the information acquired by the visual information acquisition array 2 , collect all the information required by the user according to actual needs, and display it through the human-computer interaction interface 8 .

In this embodiment, the method for the quality inspection of the finished product of the concrete prefabricated beam and the construction of the digital solid model, see Fig. 3, comprises the following steps:

S1: The finished concrete prefabricated beam 5 is transported through the detection frame 1 at a specific speed by the sample carrying component 3, and at the same time, it is obtained by coupling the laser ranging unit and the optical camera unit on the detection frame 1 with the visual information processing component 4 Sparse point cloud model, dense point cloud model and image information of all parts of the finished concrete prefabricated beam 5.

S2: Use deep neural network point cloud segmentation technology to realize the matching of sparse point cloud model and dense point cloud model, and complete the construction of As-Built BIM model. At the same time, through the sparse control points and feature points in the sparse point cloud model, all beam sections The image information of the parts is spliced to obtain the apparent panorama.

S3: Based on the apparent panorama, the deep semantic segmentation network method is used to extract the apparent features in the apparent panorama to form the apparent feature vector of the beam segment. On the one hand, the vector is used as the characteristic identifier of the beam segment for the construction of the information model ID, which provides support for the later autonomous perceptual operation and maintenance and after-sales tracking. On the other hand, the initial health state of the finished beam section is observed through the standardized analysis of the disease, and the initial parameters of the component are obtained to realize the quality evaluation of the finished product.

S4: Based on the internal and external relationship between the optical camera and the point cloud model, realize the accurate projection and transformation of the 2D image in the panorama data to the As-Built BIM model, construct the digital entity model of the beam segment, and realize the true 3D information of the finished beam segment expression.

The digital entity model constructed by this concrete prefabricated beam finished product quality inspection system can directly serve life cycle engineering information management, operation and maintenance evaluation and decision support. This technology can be used for quality control and information management of prefabricated beam section finished products in the factory. And effectively support the intelligent operation and maintenance and performance monitoring of the structure during its service period.

The above descriptions of the embodiments are for those of ordinary skill in the art to understand and use the invention. It is obvious that those skilled in the art can easily make various modifications to these embodiments, and apply the general principles described here to other embodiments without creative efforts. Therefore, the present invention is not limited to the above-mentioned embodiments. Improvements and modifications made by those skilled in the art according to the disclosure of the present invention without departing from the scope of the present invention should fall within the protection scope of the present invention.

Claims (5)

1. A quality inspection system for finished concrete prefabricated beams, characterized in that it includes a detection frame (1), a visual information acquisition array (2), a sample carrying component (3) and a visual information processing component (4); The visual information acquisition array (2) is set on the detection frame (1), forming a circular visual field along the detection frame (1), and the visual information acquisition array (2) includes a plurality of laser ranging units and optical camera units , each of the laser ranging unit and the optical camera unit is connected to the visual information processing component (4) in wireless or wired communication; The sample carrying component (3) is capable of transporting the finished concrete prefabricated beam (5) through the detection frame (1) at a specific speed; During the dynamic transportation process of the finished concrete prefabricated beam (5), the sparse point cloud model of the finished concrete precast beam (5) is obtained through the coupling of the laser ranging unit and the visual information processing component (4). The coupling of the information processing component (4) obtains the image information of the finished concrete precast beam (5), and the visual information processing component (4) realizes the splicing of image information based on the sparse point cloud model, and obtains the apparent panorama of the finished concrete precast beam (5) Figure, the visual information processing component (4) adopts the deep semantic segmentation network to extract the structural appearance features of the apparent panorama to form the apparent feature vector of the beam segment, and use the apparent feature vector of the beam segment to realize the quality of the finished concrete prefabricated beam (5) evaluate; The detection frame (1) is an annular truss structure; The sample carrying component (3) is a horizontal transport device that runs through the detection frame (1), and the sample carrying component (3) is one of rail-type transport equipment, conveyor belt-type transport equipment, and industrial handling robots. kind; The visual information acquisition array (2) also includes a synchronous collector, and the synchronous collector is electrically connected to each laser ranging unit and optical camera unit; The laser ranging unit is a multi-point infrared ranging sensor, and the optical camera unit is a multi-eye camera. 2. A quality inspection system for finished concrete prefabricated beams according to claim 1, characterized in that the visual information processing component (4) includes a microprocessor and a main memory and an auxiliary memory connected to the microprocessor through a main line , a signal transmitter is connected to the main line, and the signal transmitter is connected to the synchronous collector wirelessly or by wire. 3. The finished product quality inspection system for precast concrete beams according to claim 2, characterized in that, the finished product quality inspection system for precast concrete beams also includes a control unit (6) for carrying components and a man-machine interface (8), The human-computer interaction interface (8) is electrically connected to the carrying component control unit (6); The sample carrying component (3) is electrically connected to the carrying component control unit (6), so as to realize the control of the conveying path and conveying speed of the finished concrete prefabricated beam (5). 4. A system for constructing a digital solid model of a finished concrete prefabricated beam, characterized in that it comprises the quality inspection system for a finished concrete prefabricated beam according to any one of claims 1 to 3, wherein an optical camera unit and a visual information processing component ( 4) The image information of the finished concrete prefabricated beam (5) is obtained by coupling, and the dense point cloud model is further obtained; The visual information processing component (4) combines the sparse point cloud model and the dense point cloud model to obtain a digital entity model of the finished concrete prefabricated beam (5). 5. A method for quality detection of prefabricated concrete beams and construction of a digital solid model, characterized in that, the quality detection system of any one of claims 1 to 3 is used to realize the quality detection of finished concrete prefabricated beams and construction of a digital solid model , including the following steps: S1: The finished concrete prefabricated beam (5) is transported through the detection frame (1) at a specific speed through the sample carrying component (3), and at the same time, the laser ranging unit and the optical camera unit set on the detection frame (1) communicate with the The coupling of the visual information processing component (4) obtains the sparse point cloud model, the dense point cloud model and the image information of all parts of the finished concrete prefabricated beam (5); S2: Use deep neural network point cloud segmentation technology to realize the matching of sparse point cloud model and dense point cloud model, and complete the construction of As-Built BIM model. At the same time, through the sparse control points and feature points in the sparse point cloud model, all The image information of the parts is spliced to obtain the apparent panorama; S3: Based on the apparent panorama, the deep semantic segmentation network method is used to extract the apparent features in the apparent panorama to form the apparent feature vector of the beam segment; S4: Based on the internal and external relationship between the optical camera and the point cloud model, the accurate projection and transformation of the two-dimensional image in the panorama data to the As-BuiltBIM model is realized, and the digital entity model of the beam segment is constructed.