CN115760766A - Defect identification digital assay receiving method and system based on image identification technology - Google Patents
Defect identification digital assay receiving method and system based on image identification technology Download PDFInfo
- Publication number
- CN115760766A CN115760766A CN202211446026.XA CN202211446026A CN115760766A CN 115760766 A CN115760766 A CN 115760766A CN 202211446026 A CN202211446026 A CN 202211446026A CN 115760766 A CN115760766 A CN 115760766A
- Authority
- CN
- China
- Prior art keywords
- dimensional model
- original
- dimensional
- building
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides a defect identification digital assay receiving method and system based on an image identification technology, and belongs to the technical field of digital assay receiving. The method comprises the following steps: establishing an original three-dimensional model and an original two-dimensional model of a building according to a standard construction drawing and a construction acceptance standard of the building; constructing an actual three-dimensional model of the building based on point cloud data obtained by field scanning of the building by a three-dimensional laser scanning device; constructing an actual two-dimensional model of the building based on image data obtained by shooting the building on site by a camera; and judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model.
Description
Technical Field
The invention relates to the technical field of digital assay collection, in particular to a defect identification digital assay collection method and system based on an image identification technology.
Background
After completion of the project, the quality and achievement of the project construction need to be checked and checked to see whether the design requirements and the project quality are met according to relevant industry standards. Traditional acceptance is manual measurement, uses the measuring tool to the dimensional measurement of each position of completing the facility through the manual work, compares the design size of facility again, judges whether to accord with the design requirement after the facility is completed according to the difference between the two, and the measurement process is loaded down with trivial details, and is long consuming time to the measuring result receives the technological difference influence of acceptance testing personnel easily, leads to the measuring result accuracy to reduce.
The prior art proposes that a digital cloud platform technology is utilized, point cloud data are obtained through laser scanning to establish a three-dimensional model of a building, data check can be carried out on tiny construction positions of the building according to the three-dimensional model, the automation degree is high, however, influences of scanning equipment, surrounding environment, artificial disturbance and even surface materials of a scanned object cannot be avoided in the laser scanning process, the obtained data have more or less noise points, the situation that target point data cannot be correctly expressed exists, the accuracy of actual modeling can be influenced, and therefore acceptance conclusion is influenced.
Disclosure of Invention
In view of the above, the invention provides a defect identification digital test acceptance method and system based on an image identification technology, which improve an automatic acceptance method and overcome the problems of modeling distortion and influence on acceptance conclusion caused by adopting laser scanning to collect source data.
The technical scheme adopted by the embodiment of the invention for solving the technical problem is as follows:
a defect identification digital assay receiving method based on an image identification technology comprises the following steps:
s1, establishing an original three-dimensional model and an original two-dimensional model of a building according to a standard construction drawing and a construction acceptance standard of the building;
s2, constructing an actual three-dimensional model of the building based on point cloud data obtained by scanning the building on site by a three-dimensional laser scanning device;
s3, constructing an actual two-dimensional model of the building based on image data obtained by shooting the building on site by an image acquisition device;
and S4, judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model.
Preferably, the step S1 of establishing the original three-dimensional model and the original two-dimensional model of the building according to the standard drawing of the building includes:
generating an original two-dimensional image and an original three-dimensional image of the building by using drawing software according to the construction acceptance standard and the building construction drawing;
establishing the original two-dimensional model according to the original two-dimensional image;
and establishing the original three-dimensional model according to the original three-dimensional image.
Preferably, the three-dimensional laser scanning equipment the image acquisition equipment is installed on a rotatory cloud platform, rotatory cloud platform is installed on unmanned aerial vehicle equipment, three-dimensional laser scanning equipment with the collection work of image acquisition equipment is in unmanned aerial vehicle equipment flight in-process is executed, three-dimensional laser scanning equipment the image acquisition equipment rotatory cloud platform all with unmanned aerial vehicle equipment's mainboard electric connection, three-dimensional laser scanning equipment the data that image acquisition equipment gathered are unmanned aerial vehicle equipment passes through wireless transmission mode and transmits to the computer.
Preferably, after the step S1 and before the step S2, the method further includes:
s5, on the basis of the three-dimensional laser scanning equipment, the building is scanned on site to obtain point cloud data;
the step S5 includes:
s51, selecting an origin coordinate through a laser pen in the three-dimensional laser scanning equipment, and simultaneously acquiring and recording an initial position coordinate of the unmanned aerial vehicle equipment and initial attitude data of the rotating holder in real time, wherein the attitude data comprises an initial course angle and an initial pitch angle;
step S52, in the process of navigating inside and outside the building, the unmanned aerial vehicle device scans a target position on the building through a laser pen in the three-dimensional laser scanning device to obtain a target point position coordinate, and simultaneously obtains a real-time position coordinate of the unmanned aerial vehicle device and real-time attitude data of the rotating holder in real time;
and S53, correcting the coordinates of the target point by using the coordinates of the origin, the coordinates of the initial position and the data of the initial attitude as references and using the coordinates of the real-time position and the data of the real-time attitude to obtain the coordinates of the point cloud of the target position, wherein the point cloud data is a set of the coordinates of the point clouds.
Preferably, the step S4 of judging whether the building size of the building meets the design standard according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model, and the actual two-dimensional model includes:
cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
cutting the actual three-dimensional model and the actual two-dimensional model according to the construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating a two-dimensional model coincidence rate, wherein the two-dimensional model coincidence rate is the ratio of the coordinate coincidence number of the original two-dimensional coordinate data and the actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
carrying out coordinate origin alignment on the original three-dimensional model and the actual three-dimensional model, comparing the coordinates of the original three-dimensional model and the actual three-dimensional model by taking the original three-dimensional model as a reference, and calculating the three-dimensional model coincidence rate, wherein the three-dimensional model coincidence rate refers to the ratio of the coordinate coincidence number of the original three-dimensional coordinate data and the actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
if the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are higher than the coincidence rate threshold value, confirming that the construction size of the building meets the design standard;
and if the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold, adding error labels to the coordinates of the original two-dimensional model and the original three-dimensional model for which no coincident coordinate data is found.
The invention provides a defect identification digital assay receiving system based on an image identification technology, which comprises: the system comprises unmanned aerial vehicle equipment and a computer, wherein three-dimensional laser scanning equipment, image acquisition equipment and a rotating cloud deck are integrated on the unmanned aerial vehicle equipment, the three-dimensional laser scanning equipment and the image acquisition equipment are installed on the rotating cloud deck, the rotating cloud deck is installed on the unmanned aerial vehicle equipment, and the three-dimensional laser scanning equipment, the image acquisition equipment and the rotating cloud deck are all electrically connected with a main board of the unmanned aerial vehicle equipment;
the three-dimensional laser scanning equipment is used for scanning the building through laser in the flight process of the unmanned aerial vehicle equipment so as to obtain point cloud data of the building and transmit the point cloud data to the main controller of the unmanned aerial vehicle equipment;
the image acquisition equipment is used for shooting the scene of the building in the flight process of the unmanned aerial vehicle equipment so as to acquire image data of the building and transmit the image data to the main controller of the unmanned aerial vehicle equipment, and the image data of the building comprises scene video data and scene image data;
the main controller of the unmanned aerial vehicle equipment is used for transmitting the point cloud data of the building and the image data of the building to the computer through the wireless transceiver of the unmanned aerial vehicle equipment;
the computer includes:
the wireless receiving module is used for receiving the point cloud data of the building and the image data of the building, which are transmitted by the unmanned aerial vehicle device;
the construction original modeling module is used for establishing an original three-dimensional model and an original two-dimensional model of the building according to a standard construction drawing and a construction acceptance standard of the building;
the actual data modeling module is used for constructing an actual three-dimensional model of the building based on the point cloud data of the building; further for constructing an actual two-dimensional model of the building based on the image profile data of the building;
the comparison module is used for judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model;
the central control module is used for controlling the wireless receiving module, the construction original modeling module, the actual data modeling module and the comparison module; the system is also used for storing point cloud data of the building and image data of the building, and also used for storing operation data of the construction original modeling module, the actual data modeling module and the comparison module.
Preferably, the construction primitive modeling module is further configured to:
generating an original two-dimensional image and an original three-dimensional image of the building by using drawing software according to the construction acceptance standard and the building construction drawing; establishing the original two-dimensional model according to the original two-dimensional image; and establishing the original three-dimensional model according to the original three-dimensional image.
Preferably, the method further comprises the following steps:
and the point cloud data acquisition module is used for scanning the building on site based on the three-dimensional laser scanning equipment to obtain the point cloud data.
Preferably, the point cloud data acquisition module comprises:
the datum data acquisition unit is used for acquiring an origin coordinate selected by a laser pen in the three-dimensional laser scanning equipment, and simultaneously acquiring and recording an initial position coordinate of the unmanned aerial vehicle equipment and initial attitude data of the rotating holder in real time, wherein the attitude data comprises an initial course angle and an initial pitch angle;
the real-time data acquisition unit is used for acquiring a target point position coordinate obtained by scanning a target position on the building by a laser pen in the three-dimensional laser scanning device and acquiring a real-time position coordinate of the unmanned aerial vehicle device and real-time attitude data of the rotating holder in real time in the process of navigating inside and outside the building;
and the coordinate correction unit is used for correcting the coordinates of the target point by using the coordinates of the origin, the coordinates of the initial position and the data of the initial posture as reference and by using the coordinates of the real-time position and the data of the real-time posture to obtain the coordinates of the point cloud of the target position, wherein the point cloud data is a set of the coordinates of the point clouds.
Preferably, the comparison module comprises:
the coordinate extraction unit is used for cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
the coordinate extraction unit is also used for cutting the actual three-dimensional model and the actual two-dimensional model according to the construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
the model comparison unit is used for carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating a two-dimensional model coincidence rate, wherein the two-dimensional model coincidence rate refers to the ratio of the number of coincidences of coordinates of the original two-dimensional coordinate data and the actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
the model comparison unit is further configured to perform coordinate origin alignment processing on the original three-dimensional model and the actual three-dimensional model, perform coordinate comparison on the original three-dimensional model and the actual three-dimensional model by using the original three-dimensional model as a reference, and calculate a three-dimensional model coincidence rate, where the three-dimensional model coincidence rate is a ratio of the number of coincidences of coordinates of the original three-dimensional coordinate data and the actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
a confirming unit for confirming that the building size of the building meets the design standard when the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are both higher than a coincidence rate threshold value;
and the defect labeling unit is used for adding error labels to the coordinates of which no coincident coordinate data is found in the original two-dimensional model and the original three-dimensional model when the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold value.
According to the technical scheme, the defect identification digital test acceptance method based on the image identification technology comprises the steps of firstly establishing an original three-dimensional model and an original two-dimensional model of a building according to a standard construction drawing of the building and a construction acceptance standard; constructing an actual three-dimensional model of the building based on point cloud data obtained by scanning the building on site by a three-dimensional laser scanning device; constructing an actual two-dimensional model of the building based on image data obtained by shooting the building on site by an image acquisition device; and judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model. The invention improves the automatic acceptance method, can overcome the problems of modeling distortion and influence on acceptance conclusion caused by adopting laser to scan and collect source data, and can effectively improve the accuracy of identifying defects.
Drawings
FIG. 1 is a flow chart of a defect recognition digital assay receiving method based on image recognition technology.
Fig. 2 is a schematic structural diagram of a defect recognition digital assay receiving system based on an image recognition technology.
Detailed Description
The technical scheme and the technical effect of the invention are further elaborated in the following by combining the drawings of the invention.
As shown in fig. 2, the present invention provides a defect recognition digital assay receiving system based on image recognition technology, the system includes a drone device and a computer, and can be used to execute the method shown in fig. 1. The unmanned aerial vehicle equipment is integrated with three-dimensional laser scanning equipment, image acquisition equipment and a rotating cloud platform, the three-dimensional laser scanning equipment and the image acquisition equipment are installed on the rotating cloud platform, the cloud platform is fixed with a laser pen of the three-dimensional laser scanning equipment through a connecting frame, the rotating cloud platform is installed on the unmanned aerial vehicle equipment, and the three-dimensional laser scanning equipment, the image acquisition equipment and the rotating cloud platform are all electrically connected with a main board of the unmanned aerial vehicle equipment;
the three-dimensional laser scanning equipment is used for scanning the inside and the outside of a building through laser in the flight process of the unmanned aerial vehicle equipment so as to obtain point cloud data of the building and transmit the point cloud data to the main controller of the unmanned aerial vehicle equipment;
the image acquisition equipment is used for shooting the site of the building in the flight process of the unmanned aerial vehicle equipment so as to acquire image data of the building and transmit the image data to the main controller of the unmanned aerial vehicle equipment, and the image data of the building comprises site video data and site image data;
the main controller of the unmanned aerial vehicle equipment is used for transmitting point cloud data of a building and image data of the building to the computer through the wireless transceiving device of the unmanned aerial vehicle equipment, wherein the image data can comprise data such as site videos and site pictures of the building;
the inside of the computer is:
the point cloud data acquisition module is used for scanning the building on site based on the three-dimensional laser scanning equipment to obtain point cloud data;
the wireless receiving module is used for receiving point cloud data of the building and image data of the building, which are transmitted by the unmanned aerial vehicle device;
the construction original modeling module is used for establishing an original three-dimensional model and an original two-dimensional model of the building according to a standard construction drawing and a construction acceptance standard of the building; the construction original modeling module firstly generates an original two-dimensional image and an original three-dimensional image of a building by using drawing software according to a construction acceptance standard and a building construction drawing; then, establishing an original two-dimensional model according to the original two-dimensional image, and establishing an original three-dimensional model according to the original three-dimensional image;
the actual data modeling module is used for constructing an actual three-dimensional model of the building based on the point cloud data of the building; the system is also used for constructing an actual two-dimensional model of the building based on the image data of the building;
the comparison module is used for judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model;
the central control module is used for controlling the wireless receiving module, the construction original modeling module, the actual data modeling module and the comparison module, and the wireless receiving module, the construction original modeling module, the actual data modeling module and the comparison module are electrically connected with the central control module; the system is also used for storing point cloud data of the building and image data of the building, and is also used for storing operation data of a construction original modeling module, an actual data modeling module and a comparison module.
Wherein, point cloud data acquisition module includes:
the system comprises a reference data acquisition unit, a three-dimensional laser scanning device and a control unit, wherein the reference data acquisition unit is used for acquiring an origin coordinate selected by a laser pen in the three-dimensional laser scanning device, and simultaneously acquiring and recording an initial position coordinate of the unmanned aerial vehicle device and initial attitude data of a rotating holder in real time, and the attitude data comprises an initial course angle and an initial pitch angle;
the system comprises a real-time data acquisition unit, a data acquisition unit and a data acquisition unit, wherein the real-time data acquisition unit is used for acquiring target point position coordinates obtained by scanning a target position on a building by a laser pen in three-dimensional laser scanning equipment in the process of navigating inside and outside the building, and simultaneously acquiring real-time position coordinates of the unmanned aerial vehicle equipment and real-time attitude data of a rotating holder in real time;
and the coordinate correction unit is used for correcting the coordinates of the target point by taking the original point coordinates, the initial position coordinates and the initial attitude data as the reference through the real-time position coordinates and the real-time attitude data to obtain the point cloud coordinates of the target position, and the point cloud data is a set of the point cloud coordinates.
The three-dimensional laser scanning equipment focuses on the position of any point far away through the laser pen and sets the position as an original point coordinate, the pitch angle and the course angle of the cradle head are adjusted through adjusting the double-freedom-degree rotary table to simulate the north seeking and tracking adjustment of the cradle head, the cradle head is reset and adjusted after the adjustment is completed, and meanwhile initial data are recorded and transmitted.
Further, the comparison module comprises a coordinate extraction unit, a model comparison unit, a confirmation unit and a defect labeling unit:
the coordinate extraction unit is used for cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
the coordinate extraction unit is also used for cutting the actual three-dimensional model and the actual two-dimensional model according to the construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
the model comparison unit is used for carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating the coincidence rate of the two-dimensional models, wherein the coincidence rate of the two-dimensional models refers to the ratio of the coordinate coincidence number of original two-dimensional coordinate data and actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
the model comparison unit is also used for carrying out coordinate origin alignment processing on the original three-dimensional model and the actual three-dimensional model, carrying out coordinate comparison on the original three-dimensional model and the actual three-dimensional model by taking the original three-dimensional model as a reference, and calculating the three-dimensional model coincidence rate, wherein the three-dimensional model coincidence rate refers to the ratio of the coordinate coincidence number of original three-dimensional coordinate data and actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
the confirming unit is used for confirming that the construction size of the building meets the design standard when the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are higher than the coincidence rate threshold value;
and the defect labeling unit is used for adding error labels to the coordinates of which the coincident coordinate data are not found in the original two-dimensional model and the original three-dimensional model when the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold value.
As shown in fig. 2, the present invention provides a defect identification digital assay receiving method based on image identification technology, which is implemented by a computer in the system shown in fig. 1, and comprises the following steps:
s1, establishing an original three-dimensional model and an original two-dimensional model of a building according to a standard construction drawing and a construction acceptance standard of the building;
s2, constructing an actual three-dimensional model of the building based on point cloud data obtained by the site scanning of the building by the three-dimensional laser scanning equipment;
s3, constructing an actual two-dimensional model of the building based on image data obtained by shooting the building on site by the image acquisition equipment;
and S4, judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model.
The step S1 of establishing the original three-dimensional model and the original two-dimensional model of the building according to the standard drawing of the building comprises the following steps:
step S11, generating an original two-dimensional image and an original three-dimensional image of a building by using drawing software according to a construction acceptance standard and a building construction drawing;
s12, establishing an original two-dimensional model according to the original two-dimensional image;
and S13, establishing an original three-dimensional model according to the original three-dimensional image.
After step S1 and before step S2, the method further includes:
s5, on the basis of the three-dimensional laser scanning equipment, the building is scanned on site to obtain point cloud data;
the step S5 includes:
s51, selecting an origin coordinate through a laser pen in the three-dimensional laser scanning equipment, and simultaneously acquiring and recording an initial position coordinate of the unmanned aerial vehicle equipment and initial attitude data of the rotating holder in real time, wherein the attitude data comprises an initial course angle and an initial pitch angle;
step S52, in the process of navigating inside and outside the building, the unmanned aerial vehicle device scans a target position on the building through a laser pen in the three-dimensional laser scanning device to obtain a target point position coordinate, and simultaneously obtains a real-time position coordinate of the unmanned aerial vehicle device and real-time attitude data of the rotating holder in real time;
and S53, correcting the coordinates of the target point by using the coordinates of the origin, the coordinates of the initial position and the data of the initial attitude as references and using the coordinates of the real-time position and the data of the real-time attitude to obtain the coordinates of the point cloud of the target position, wherein the point cloud data is a set of the coordinates of the point clouds.
The laser pen is used for focusing the position of any point far away and setting the position as an origin coordinate, the pitch angle and the course angle of the cradle head are adjusted by adjusting the double-freedom-degree rotary table to simulate the north seeking and tracking adjustment of the cradle head, the cradle head is reset and adjusted after the adjustment is completed, and angle data are recorded and transmitted simultaneously.
Specifically, the step S4 of judging whether the building size of the building meets the design standard according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model, and the actual two-dimensional model includes:
s41, cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
s42, cutting the actual three-dimensional model and the actual two-dimensional model according to a construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
step S43, carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating the coincidence rate of the two-dimensional models, wherein the coincidence rate of the two-dimensional models refers to the ratio of the number of coincidences of coordinates of original two-dimensional coordinate data and actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
step S44, carrying out coordinate origin alignment processing on the original three-dimensional model and the actual three-dimensional model, carrying out coordinate comparison on the original three-dimensional model and the actual three-dimensional model by taking the original three-dimensional model as a reference, and calculating the three-dimensional model coincidence rate, wherein the three-dimensional model coincidence rate refers to the ratio of the coordinate coincidence number of the original three-dimensional coordinate data and the actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
and S45, if the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are higher than the coincidence rate threshold value, confirming that the building size of the building meets the design standard, wherein the coincidence rate threshold value can be 90%.
And S46, if the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold, adding error labels to the coordinates of the original two-dimensional model and the original three-dimensional model, wherein the coincident coordinate data are not found.
According to the method and the system, a three-dimensional laser scanning point cloud data three-dimensional modeling technology and a field image two-dimensional modeling technology are adopted, the actual three-dimensional model is established for the actual condition of the building by using the point cloud data obtained by laser scanning, the actual two-dimensional model is obtained by using the field image data to perform two-dimensional modeling, then the coordinate coincidence degree calculation is performed on the actual three-dimensional model and the actual two-dimensional model respectively with the original three-dimensional model and the original two-dimensional model obtained by modeling based on the drawing and the construction acceptance standard, the accuracy of defect identification can be improved through two different comparison modes of information sources, the accuracy of acceptance conclusion can be further improved, and the defect part can be identified more strictly. According to the invention, the unmanned aerial vehicle device carries the three-dimensional laser scanning device and the image acquisition device to automatically acquire the field data, and the acquired data is transmitted back to the computer in a wireless transmission manner, so that the system has the characteristics of high automation degree, time saving, energy saving and high working efficiency.
While the invention has been described in connection with what is presently considered to be the most practical and preferred embodiment, it is to be understood that the invention is not to be limited to the disclosed embodiment, but on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims (10)
1. A defect identification digital assay receiving method based on image identification technology is characterized by comprising the following steps:
the method comprises the following steps of S1, establishing an original three-dimensional model and an original two-dimensional model of a building according to a standard construction drawing and a construction acceptance standard of the building;
s2, constructing an actual three-dimensional model of the building based on point cloud data obtained by scanning the building on site by a three-dimensional laser scanning device;
s3, constructing an actual two-dimensional model of the building based on image data obtained by shooting the building on site by an image acquisition device;
and S4, judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model.
2. The method for digitally testing and receiving defect identifications based on image recognition technology according to claim 1, wherein the step S1 of creating an original three-dimensional model and an original two-dimensional model of a building according to a standard drawing of the building comprises:
generating an original two-dimensional image and an original three-dimensional image of the building by using drawing software according to the construction acceptance standard and the building construction drawing;
establishing the original two-dimensional model according to the original two-dimensional image;
and establishing the original three-dimensional model according to the original three-dimensional image.
3. The method for defect recognition digital assay reception based on image recognition technology according to claim 2, wherein the three-dimensional laser scanning device and the image acquisition device are installed on a rotating cradle head, the rotating cradle head is installed on an unmanned aerial vehicle device, the acquisition work of the three-dimensional laser scanning device and the image acquisition device is performed during the flight process of the unmanned aerial vehicle device, the three-dimensional laser scanning device, the image acquisition device and the rotating cradle head are all electrically connected with a main board of the unmanned aerial vehicle device, and the data acquired by the three-dimensional laser scanning device and the image acquisition device are transmitted to a computer by the unmanned aerial vehicle device in a wireless transmission mode.
4. The method for defect recognition digital assay reception based on image recognition technology as claimed in claim 3, wherein after step S1 and before step S2, further comprising:
s5, scanning the building on site based on the three-dimensional laser scanning equipment to obtain point cloud data;
the step S5 includes:
s51, selecting an origin coordinate through a laser pen in the three-dimensional laser scanning equipment, and simultaneously acquiring and recording an initial position coordinate of the unmanned aerial vehicle equipment and initial attitude data of the rotating holder in real time, wherein the attitude data comprises an initial course angle and an initial pitch angle;
step S52, in the process of navigating inside and outside the building, the unmanned aerial vehicle device scans a target position on the building through a laser pen in the three-dimensional laser scanning device to obtain a target point position coordinate, and simultaneously obtains a real-time position coordinate of the unmanned aerial vehicle device and real-time attitude data of the rotating holder in real time;
and S53, correcting the coordinates of the target point by using the coordinates of the origin, the coordinates of the initial position and the data of the initial attitude as references and using the coordinates of the real-time position and the data of the real-time attitude to obtain the coordinates of the point cloud of the target position, wherein the point cloud data is a set of the coordinates of the point clouds.
5. The method for defect recognition digital test reception based on image recognition technology according to claim 4, wherein the step S4 of judging whether the built size of the building meets the design standard according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model comprises:
cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
cutting the actual three-dimensional model and the actual two-dimensional model according to the construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating a two-dimensional model coincidence rate, wherein the two-dimensional model coincidence rate refers to the ratio of the coordinate coincidence number of the original two-dimensional coordinate data and the actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
carrying out coordinate origin alignment processing on the original three-dimensional model and the actual three-dimensional model, carrying out coordinate comparison on the original three-dimensional model and the actual three-dimensional model by taking the original three-dimensional model as a reference, and calculating a three-dimensional model coincidence rate, wherein the three-dimensional model coincidence rate refers to the ratio of the coordinate coincidence number of the original three-dimensional coordinate data and the actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
if the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are higher than the coincidence rate threshold value, confirming that the construction size of the building meets the design standard;
and if the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold, adding error labels to the coordinates of the original two-dimensional model and the original three-dimensional model for which no coincident coordinate data is found.
6. A defect recognition digital assay receiving system based on image recognition technology, comprising: the system comprises unmanned aerial vehicle equipment and a computer, wherein three-dimensional laser scanning equipment, image acquisition equipment and a rotating cloud deck are integrated on the unmanned aerial vehicle equipment, the three-dimensional laser scanning equipment and the image acquisition equipment are installed on the rotating cloud deck, the rotating cloud deck is installed on the unmanned aerial vehicle equipment, and the three-dimensional laser scanning equipment, the image acquisition equipment and the rotating cloud deck are all electrically connected with a main board of the unmanned aerial vehicle equipment;
the three-dimensional laser scanning equipment is used for scanning the building through laser in the flight process of the unmanned aerial vehicle equipment so as to obtain point cloud data of the building and transmit the point cloud data to the main controller of the unmanned aerial vehicle equipment;
the image acquisition equipment is used for shooting the scene of the building in the flight process of the unmanned aerial vehicle equipment so as to acquire image data of the building and transmit the image data to the main controller of the unmanned aerial vehicle equipment, and the image data of the building comprises scene video data and scene image data;
the main controller of the unmanned aerial vehicle device is used for transmitting the point cloud data of the building and the image data of the building to the computer through the wireless transceiver device of the unmanned aerial vehicle device;
the computer includes:
the wireless receiving module is used for receiving the point cloud data of the building and the image data of the building, which are transmitted by the unmanned aerial vehicle device;
the construction original modeling module is used for establishing an original three-dimensional model and an original two-dimensional model of the building according to a standard construction drawing of the building and a construction acceptance standard;
the actual data modeling module is used for constructing an actual three-dimensional model of the building based on the point cloud data of the building; further for constructing an actual two-dimensional model of the building based on the image profile data of the building;
the comparison module is used for judging whether the building size of the building meets the design standard or not according to the original three-dimensional model, the original two-dimensional model, the actual three-dimensional model and the actual two-dimensional model;
the central control module is used for controlling the wireless receiving module, the construction original modeling module, the actual data modeling module and the comparison module; the system is also used for storing point cloud data of the building and image data of the building, and is also used for storing operation data of the construction original modeling module, the actual data modeling module and the comparison module.
7. The image recognition technology-based defect recognition digital laboratory collecting system of claim 6, wherein said construction raw modeling module is further configured to:
generating an original two-dimensional image and an original three-dimensional image of the building by using drawing software according to the construction acceptance standard and the building construction drawing; establishing the original two-dimensional model according to the original two-dimensional image; and establishing the original three-dimensional model according to the original three-dimensional image.
8. The image recognition technology-based defect recognition digital assay receiving system of claim 7, further comprising:
and the point cloud data acquisition module is used for scanning the building on site based on the three-dimensional laser scanning equipment to obtain the point cloud data.
9. The image recognition technology-based defect recognition digital assay collection system of claim 8, wherein the point cloud data collection module comprises:
the datum data acquisition unit is used for acquiring the origin coordinates selected by a laser pen in the three-dimensional laser scanning equipment, and simultaneously acquiring and recording the initial position coordinates of the unmanned aerial vehicle equipment and the initial attitude data of the rotating holder in real time, wherein the attitude data comprises an initial course angle and an initial pitch angle;
the real-time data acquisition unit is used for acquiring a target point position coordinate obtained by scanning a target position on the building by a laser pen in the three-dimensional laser scanning device and acquiring a real-time position coordinate of the unmanned aerial vehicle device and real-time attitude data of the rotating holder in real time in the process of navigating inside and outside the building;
and the coordinate correction unit is used for correcting the coordinates of the target point by using the coordinates of the origin, the coordinates of the initial position and the data of the initial posture as reference and by using the coordinates of the real-time position and the data of the real-time posture to obtain the coordinates of the point cloud of the target position, wherein the point cloud data is a set of the coordinates of the point clouds.
10. The image recognition technology-based defect recognition digital assay receiving system of claim 9, wherein the comparison module comprises:
the coordinate extraction unit is used for cutting the original three-dimensional model and the original two-dimensional model according to a construction acceptance standard to obtain original three-dimensional coordinate data and original two-dimensional coordinate data;
the coordinate extraction unit is also used for cutting the actual three-dimensional model and the actual two-dimensional model according to the construction acceptance standard to obtain actual three-dimensional coordinate data and actual two-dimensional coordinate data;
the model comparison unit is used for carrying out coordinate origin alignment processing on the original two-dimensional model and the actual two-dimensional model, carrying out coordinate comparison on the original two-dimensional model and the actual two-dimensional model by taking the original two-dimensional model as a reference, and calculating a two-dimensional model coincidence rate, wherein the two-dimensional model coincidence rate refers to the ratio of the number of coincidences of coordinates of the original two-dimensional coordinate data and the actual two-dimensional coordinate data to the total number of the original two-dimensional coordinate data;
the model comparison unit is further configured to perform coordinate origin alignment processing on the original three-dimensional model and the actual three-dimensional model, perform coordinate comparison on the original three-dimensional model and the actual three-dimensional model by using the original three-dimensional model as a reference, and calculate a three-dimensional model coincidence rate, where the three-dimensional model coincidence rate is a ratio of the number of coincidences of coordinates of the original three-dimensional coordinate data and the actual three-dimensional coordinate data to the total number of the original three-dimensional coordinate data;
a confirming unit for confirming that the building size of the building meets the design standard when the two-dimensional model coincidence rate and the three-dimensional model coincidence rate are both higher than a coincidence rate threshold value;
and the defect labeling unit is used for adding error labels to the coordinates of which no coincident coordinate data is found in the original two-dimensional model and the original three-dimensional model when the coincidence rate of the two-dimensional model and the coincidence rate of the three-dimensional model are not higher than the coincidence rate threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211446026.XA CN115760766A (en) | 2022-11-18 | 2022-11-18 | Defect identification digital assay receiving method and system based on image identification technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211446026.XA CN115760766A (en) | 2022-11-18 | 2022-11-18 | Defect identification digital assay receiving method and system based on image identification technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115760766A true CN115760766A (en) | 2023-03-07 |
Family
ID=85373219
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211446026.XA Pending CN115760766A (en) | 2022-11-18 | 2022-11-18 | Defect identification digital assay receiving method and system based on image identification technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115760766A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116246295A (en) * | 2023-05-09 | 2023-06-09 | 深圳市嘉闰州生态建工有限公司 | Intelligent analysis device and method for suitability of building templates based on GIS |
-
2022
- 2022-11-18 CN CN202211446026.XA patent/CN115760766A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116246295A (en) * | 2023-05-09 | 2023-06-09 | 深圳市嘉闰州生态建工有限公司 | Intelligent analysis device and method for suitability of building templates based on GIS |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112949478B (en) | Target detection method based on tripod head camera | |
| CN109215111B (en) | Indoor scene three-dimensional modeling method based on laser range finder | |
| WO2018046617A1 (en) | Method and system for calibrating multiple cameras | |
| CN111932609A (en) | Cloud deck calibration method and device for valve hall equipment inspection robot and storage medium | |
| CN105302151A (en) | Aircraft docking guidance and type recognition system and method | |
| CN210451366U (en) | Galvanometer correction system | |
| CN108344693B (en) | Automatic welding-oriented visual measurement method for misalignment of sheet welding seam | |
| CN110322462B (en) | Unmanned aerial vehicle visual landing method and system based on 5G network | |
| CN110533649B (en) | Unmanned aerial vehicle general structure crack identification and detection device and method | |
| CN112257536B (en) | Space and object three-dimensional information acquisition and matching equipment and method | |
| CN115972093B (en) | Workpiece surface measuring method and device and wing wallboard soft mold polishing method | |
| CN110966956A (en) | Binocular vision-based three-dimensional detection device and method | |
| CN113134683A (en) | Laser marking method and device based on machine learning | |
| CN115760766A (en) | Defect identification digital assay receiving method and system based on image identification technology | |
| CN108564628A (en) | A kind of cutterhead vision positioning orientation system towards development machine automation | |
| CN113888712A (en) | Building structure construction deviation analysis method | |
| CN105157564B (en) | A kind of rapid automatized photogrammetric survey method of unit truss splicing accuracy | |
| CN114434036B (en) | Three-dimensional vision system for gantry robot welding of large ship structural member and operation method | |
| CN115965743A (en) | Three-dimensional modeling system and method based on VR and oblique photography collected data | |
| CN116465335A (en) | Automatic thickness measurement method and system based on point cloud matching | |
| CN113109364B (en) | Method and device for detecting chip defects | |
| CN117522830A (en) | Point cloud scanning system for detecting boiler corrosion | |
| CN117006972A (en) | Object detection method, device, equipment, three-dimensional scanning system and storage medium | |
| CN114918723B (en) | Workpiece positioning control system and method based on surface detection | |
| CN112304250B (en) | Three-dimensional matching equipment and method between moving objects |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |