Pre-splitting blasting half-wall porosity intelligent identification system and method

Abstract

The invention discloses a pre-splitting blasting half-wall porosity intelligent identification system and method, and the system comprises a front-end display platform and a rear-end operation system which carry out data interaction through a computer, and the operation result information is displayed and checked through the front-end display platform; the front-end display platform consists of a menu bar, a toolbar, a display window and a command line; the rear-end operation system mainly comprises a half-wall hole image database, an image denoising module, an image segmentation module, a data statistical analysis module, an identification calculation module and the like, half-wall hole photo images are collected on steps on the two sides of the platform through the unmanned aerial vehicle oblique photography technology with or without the RTK function, a three-dimensional image database marked with half-wall hole information is established, and the three-dimensional image database is used for storing the half-wall hole information. An intelligent half-wall porosity recognition algorithm based on computer deep learning is developed, accurate recognition of the half-wall porosity is achieved, the traditional presplitting blasting quality evaluation efficiency can be greatly improved, and the method can be widely applied to slope presplitting blasting effect evaluation in the fields of surface mines, water conservancy and hydropower, transportation and the like.

Description

Technical Field

[0001] The invention relates to the technical field of pre-splitting blasting blasthole identification, and in particular to a pre-splitting blasting half-wall porosity intelligent identification system and method. Background Art

[0002] Pre-splitting blasting technology is an important means of controlling the stability of rock and soil slope engineering. It plays an important role in reducing the hazards of production blasting vibration and maintaining slope safety. It has broad development and application scenarios in open-pit mines, water conservancy and hydropower, transportation and other fields.

[0003] Pre-splitting blasting quality evaluation is an important measure to measure the blasting effect and analyze the level of slope stability control. It can also provide guidance for pre-splitting blasting parameter optimization and project settlement. During pre-splitting blasting, forming a well-connected pre-splitting surface along the pre-splitting hole is the most basic requirement for achieving the purpose of pre-splitting blasting. In addition, blasting quality evaluation is also required from aspects such as half-wall porosity, slope surface roughness, crack width of pre-splitting blasting, and slope appearance.

[0004] The half-wall porosity, that is, the preservation rate of residual blasthole traces, refers to the ratio of the total length of blasthole traces preserved on the pre-splitting blasting contour surface to the total length of the pre-splitting hole. The half-wall porosity directly reflects the degree of damage to the rock mass around the pre-splitting hole, and is a key indicator for measuring the effect of pre-splitting blasting. The higher the half-wall porosity, the better the pre-splitting blasting effect, the higher the blasting quality, the more conducive to maintaining slope stability, and the more reasonable the blasting parameters. On the contrary, a low half-porosity indicates a poor pre-splitting blasting effect, reflecting the poor matching of the blasting parameters and the on-site rock mass conditions, and parameter optimization and adjustment are needed to further improve the quality of pre-splitting blasting.

[0005] In terms of pre-splitting blasting quality evaluation, manual observation and ruler measurement are traditionally used to conduct statistical analysis of half-wall porosity. There are problems such as high labor intensity, low operating efficiency, and large measurement errors in statistical analysis work. In addition, when conducting field measurements on high and steep slopes, the operating safety of on-site surveyors cannot be well guaranteed, and the risk of safe operations is high. Summary of the invention

[0006] The purpose of the present invention is to provide a pre-splitting blasting half-wall porosity intelligent identification system and method, which can effectively improve the quality evaluation efficiency and intelligence level of open-pit slope pre-splitting blasting, and overcome the shortcomings and deficiencies of traditional manual statistical evaluation methods.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A pre-splitting blasting half-wall porosity intelligent identification system comprises a front-end display platform and a back-end computing system for data interaction through a computer, wherein the computing result information of the back-end computing system is displayed and viewed through the front-end display platform; the front-end display platform is composed of a menu bar, a tool bar, a display window and a command line; the menu bar comprises an image import module, a recognition computing module and a recognition result output module, wherein the image import module is used for inputting half-wall hole photos, and the recognition computing module is connected with the back-end computing system; the recognition result output module is used for viewing the computing result; the tool bar comprises function buttons related to rotation, dimension measurement, view, bar chart, curve chart, data table, save and print, which are used for operating the chart in the display window; the display window displays the half-wall hole photo image, the half-wall porosity recognition result chart and the half-wall porosity recognition result table in the form of a chart; the command line is used for inputting operation commands and corresponds to the functions of each button of the tool bar; the back-end computing system comprises a three-dimensional image database of the half-wall hole, an image denoising module, an image segmentation module, a data statistical analysis module and a recognition computing module, and the recognition accuracy of the half-wall porosity is gradually improved through continuous correction of the recognition algorithm.

[0009] The present invention provides another technical solution: a pre-splitting blasting half-wall porosity intelligent identification method, which is implemented based on a pre-splitting blasting half-wall porosity intelligent identification system, and includes the following steps:

[0010] S1: Using UAV oblique photography technology with or without RTK real-time differential positioning function, collect half-wall hole photos reflecting the pre-splitting blasting effect of rock slope with centimeter-level positioning accuracy according to the preset UAV flight route;

[0011] S2: Import the collected half-wall hole photos of the on-site pre-splitting blasting effect into the computer 3D image processing software, and use the 3D image processing software to divide the half-wall hole into segment lengths l1, l2, l3...l n Mark the length and inclination information one by one;

[0012] S3: Summarize the annotated half-wall hole photo images to establish a half-wall hole three-dimensional image database containing pre-crack hole annotation information;

[0013] S4: Based on computer deep learning technology, develop half-wall porosity intelligent recognition algorithm and recognition result front-end display platform, combine with manual measurement data comparison and analysis, and gradually correct half-wall porosity intelligent recognition algorithm to improve half-wall porosity recognition accuracy;

[0014] S5: Import the image to be recognized into the front-end display platform. The back-end computing system automatically recognizes and computes the image through the intelligent recognition algorithm of half-wall porosity, and outputs the recognition result of the image in the form of a chart, which can be displayed, viewed, downloaded, and printed on the front-end display platform.

[0015] Furthermore, the overlap between two adjacent photos collected in S1 is not less than 40%. By using the image similarity search and recognition method, the adjacent images are highly fused and regional images of different sizes are synthesized as needed.

[0016] Furthermore, the RTK real-time differential positioning function used in S1 is flexibly selected according to actual needs: if the coordinates of the explosion area need to be located, a method with RTK function is used; if the positioning coordinate information does not need to be reflected, a method without RTK function is used.

[0017] Furthermore, the half-wall hole is located on the upper step and the lower step, and the upper step and the lower step are connected by a platform.

[0018] Furthermore, the half-wall porosity intelligent identification algorithm in S4 includes half-wall hole segment length identification and half-wall porosity statistical analysis. According to the actual needs on site, the half-wall porosity η1 of a single pre-splitting hole or the half-wall porosity η2 of pre-splitting holes within a certain range can be identified, where η1 and η2 reflect the local and overall pre-splitting blasting quality respectively; the calculation methods of η1 and η2 are as follows:

[0019] For the half-wall porosity η1 of a single pre-crack hole: if the half-wall hole after blasting of the same pre-crack hole is divided into n segments, the segment lengths are l1, l2, l3, ...l n , the total length ∑l needs to be accumulated, that is: ∑l=l1+l2+l3+...+l n , n is the number of segments,

[0020] The calculation algorithm of the half-wall porosity η1 is set in the back-end computing system: η1 = L1 / ∑l; L1 is the actual total length of a single pre-crack hole formed by drilling;

[0021] For the half-wall porosity η2 of the pre-cracked holes within a certain range: the sum of the segment lengths of each half-wall hole is accumulated, that is: ∑L=∑l1+∑l2+∑l3+...+∑l m , m is the number of pre-split holes;

[0022] The calculation algorithm of the half-wall porosity η2 is set in the back-end computing system: η2=L0 / ∑L; L0 is the actual total length of all pre-crack holes formed by drilling.

[0023] Furthermore, when identifying and analyzing the half-wall porosity η2 of pre-crack holes within a certain range, the image search, recognition and fusion technology of drone oblique photography is used to splice and synthesize the images taken each time, and synthesize three-dimensional images of different sizes according to actual needs, and then import them into the intelligent recognition system for recognition calculation and result output.

[0024] Compared with the prior art, the present invention has the following beneficial effects:

[0025] 1. Compared with the traditional manual measurement method for evaluating the effect of pre-splitting blasting, the intelligent identification system and method of pre-splitting blasting half-wall porosity of the present invention has the characteristics of high identification accuracy and high analysis efficiency, which can significantly improve the evaluation efficiency of pre-splitting blasting quality of open-pit slopes and realize efficient feedback of pre-splitting blasting parameter regulation.

[0026] 2. The pre-splitting blasting half-wall porosity intelligent identification system and method of the present invention can reduce the labor intensity of pre-splitting blasting effect evaluation, overcome the drawbacks of manual measurement and analysis, and effectively ensure the safety of pre-splitting blasting effect evaluation personnel.

[0027] 3. The intelligent identification system and method of half-wall porosity for pre-splitting blasting of the present invention can well realize the evaluation of the effect of pre-splitting blasting in different zones through the identification of half-wall porosity at different levels or ranges, and provide a basis for the stability analysis of open-pit rock and soil slopes.

[0028] 4. The pre-splitting blasting half-wall porosity intelligent identification system and method of the present invention are simple to operate, safe and reliable, highly practical, easy for on-site technicians to master, and have good prospects for promotion and application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 This is a schematic diagram of the distribution of half-wall holes for pre-splitting blasting of the present invention;

[0030] Figure 2 A schematic diagram of the layout of the front-end display platform of the intelligent identification system of the present invention;

[0031] Figure 3 It is a schematic diagram of the identification process of the present invention.

[0032] In the figure: 1. Up stairs; 2. Down stairs; 3. Half-wall hole; 4. Platform; 5. Menu bar; 6. Tool bar; 7. Display window; 8. Command line. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1-2In order to effectively improve the efficiency and intelligence level of quality evaluation of open-pit slope pre-splitting blasting and overcome the shortcomings and deficiencies of traditional manual statistical evaluation methods, an embodiment of the present invention provides an intelligent identification system for half-wall porosity of pre-splitting blasting, including a front-end display platform and a back-end computing system for data exchange through a computer, and the computing result information of the back-end computing system is displayed and viewed through the front-end display platform.

[0035] Specifically, the front-end display platform is composed of a menu bar 5, a toolbar 6, a display window 7 and a command line 8; the menu bar 5 includes an image import module, a recognition operation module and a recognition result output module, wherein the image import module is used to input photos of the half-wall hole 3, and the photo formats include JPEG, PNG, GIF, BMP, TIFF, etc.; the recognition operation module is used to connect with the back-end operation system; the recognition result output module is used to view the operation results; the toolbar 6 includes function buttons such as rotation, size measurement, view (front view, side view, top view, three-dimensional view), bar graph, curve graph, data table, save, print, etc., which are used to operate the display window The display window 7 displays the half-wall hole 3 photo image, the half-wall porosity recognition result diagram (including bar graph, curve graph), and the half-wall porosity recognition result table in the form of a chart, and displays the half-wall porosity recognition result in the form of a chart; the command line 8 is used to input operation commands and corresponds to the functions of each button of the toolbar 6; the back-end computing system includes a three-dimensional image database of the half-wall hole 3, an image denoising module, an image segmentation module, a data statistical analysis module, and a recognition calculation module. Through the continuous correction of the recognition algorithm, the recognition accuracy of the half-wall porosity is gradually improved.

[0036] See also Figure 3 Based on the above-mentioned pre-splitting blasting half-wall porosity intelligent identification system, the embodiment of the present invention also provides a pre-splitting blasting half-wall porosity intelligent identification method, comprising the following steps:

[0037] S1: Using the drone oblique photography technology with or without RTK (Real-time kinematic) real-time differential positioning function, according to the preset drone flight route, collect centimeter-level positioning accuracy of the half-wall hole 3 reflecting the rock slope pre-splitting blasting effect, wherein the half-wall hole 3 is located on the upper step 1 and the lower step 2, and the upper step 1 and the lower step 2 are connected by the platform 4; there is or is not the RTK function, which can be flexibly selected according to actual needs; if it is necessary to locate the position coordinates of the blasting area, a method with RTK function is used; if it is not necessary to reflect the positioning coordinate information, a method without RTK function is used. The presence or absence of RTK function does not affect the accurate identification of the half-wall porosity of the present invention.

[0038] S2: Import the collected photos of the half-wall hole 3 of the on-site pre-splitting blasting effect into the computer three-dimensional image processing software, and use the three-dimensional image processing software to divide the half-wall hole 3 into segment lengths l1, l2, l3...l n Mark the length and inclination information one by one;

[0039] S3: Summarize the annotated half-wall hole 3 photo images to establish a half-wall hole three-dimensional image database containing pre-crack hole annotation information;

[0040] S4: Based on computer deep learning technology, develop half-wall porosity intelligent recognition algorithm and recognition result front-end display platform, combine with manual measurement data comparison and analysis, and gradually correct half-wall porosity intelligent recognition algorithm to improve half-wall porosity recognition accuracy;

[0041] S5: Import the image to be recognized into the front-end display platform. The back-end computing system automatically recognizes and computes the image through the intelligent recognition algorithm of half-wall porosity, and outputs the recognition result of the image in the form of a chart, which can be displayed, viewed, downloaded, and printed on the front-end display platform.

[0042] In the above steps, the half-wall porosity intelligent identification algorithm includes two main intelligent algorithms: half-wall hole 3 segment length identification and half-wall porosity statistical analysis. According to the actual needs on site, the half-wall porosity η1 of a single pre-splitting hole and the half-wall porosity η2 of pre-splitting holes within a certain range can be identified. The half-wall porosities η1 and η2 here respectively reflect the local and overall pre-splitting blasting quality, and can also explain the rationality of the blasting parameters under the rock mass quality conditions, so as to provide feedback to guide the blasting parameter regulation. The size of a certain range here is flexibly controlled according to the actual situation; the larger the identification range, the more it can reflect the overall pre-splitting blasting effect; the smaller the identification range, the more it can reflect the local pre-splitting blasting effect. For example, a single blasting area, a single step, the east / west / south / north steps, the entire stope, etc. can be divided according to the size of the range, so as to realize the leap-forward pre-splitting blasting quality evaluation of different levels and ranges from local to overall, realize the targeted partition blasting effect judgment, and provide guidance for the pre-splitting blasting parameter regulation of the corresponding partition.

[0043] Among them, the half-wall porosity η1 of a single pre-cracked hole: For a single pre-cracked hole, due to the differences in rock mass conditions, the half-wall hole after blasting of the same pre-cracked hole may have multiple segments (usually the segments are not continuous and there are gaps). If it is divided into n segments, the segment lengths are l1, l2, l3, ...l n , we need to add up the total length ∑l, that is: ∑l = l1 + l2 + l3 + ... + l n , n is the number of segments. The continuity of the half-wall hole determines the number of segments. The better the continuity, the fewer the segments. When the half-wall hole is completely continuous, the number of segments is the least, that is, n = 1.

[0044] The calculation algorithm of the half-wall porosity η1 is set in the back-end computing system: η1=L1 / ∑l; L1 here is the actual total length of a single pre-crack hole formed by drilling, which is based on the on-site measurement acceptance data of the pre-crack hole.

[0045] For the half-wall porosity η2 of the pre-cracked holes within a certain range: the sum of the segment lengths of each half-wall hole 3 is accumulated to obtain the total length ∑L of the half-wall holes within the required identification range, that is: ∑L=∑l1+∑l2+∑l3+...+∑l m , m is the number of pre-crack holes.

[0046] The calculation algorithm of the half-wall porosity η2 is set in the back-end computing system: η2=L0 / ∑L; L0 here is the actual total length of all pre-crack holes formed by drilling, and the pre-crack hole on-site measurement acceptance data shall prevail.

[0047] According to the above calculation method, the half-wall porosities η1 and η2 can be obtained, and then the half-wall porosity results of a single pre-splitting hole and pre-splitting holes in a certain range can be output in the form of graphs and tables respectively; considering the limited range of the image taken each time, in order to better reflect the pre-splitting blasting effect in a large range, it is necessary to adopt the image search, recognition and fusion technology of oblique photography, splice and synthesize the images taken each time, and synthesize images of different sizes according to actual needs, so as to provide solutions for the evaluation of pre-splitting blasting effects in different scenes and different ranges, and use three-dimensional engine software to view the overall pre-splitting blasting effect from different perspectives.

[0048] When implementing:

[0049] (1) Using the UAV oblique photography technology with or without RTK (Real-time kinematic) real-time differential positioning function, according to the preset UAV flight route (flying along the slope direction and vertically), collect three photos of the half-wall hole reflecting the slope pre-splitting blasting effect with centimeter-level positioning accuracy; among them, the overlap between two adjacent photos is not less than 40%. Through the image similarity search and recognition method, a high degree of fusion of adjacent images is achieved, the error of image splicing synthesis is eliminated, and the accuracy and reliability of image synthesis are guaranteed; according to the different numbers of synthesized photos, regional images of different ranges and sizes can be synthesized.

[0050] (2) The collected photos of the on-site pre-splitting blasting effect are imported into the computer three-dimensional image processing software, and the segment lengths (l1, l2, l3...l n ) are marked one by one (length, inclination).

[0051] (3) The annotated half-wall hole 3 images are aggregated to establish a three-dimensional image database, including current high-definition photo images of different blasting positions, sizes, and ranges, and the corresponding blasting parameters, charge structure, and other key information are recorded.

[0052] (4) The traditional manual measurement method is used to statistically analyze the semi-hole ratio of on-site pre-splitting blasting as the basis for intelligent identification and comparative analysis.

[0053] (5) Develop a computer deep learning algorithm for half-wall porosity recognition, combine it with comparative analysis of manual measurement data, and gradually correct and improve the deep learning algorithm for half-wall porosity recognition through autonomous learning and training of big data models to improve the accuracy of half-wall hole 3 length recognition.

[0054] (6) Develop a front-end display platform for the pre-splitting blasting half-wall hole rate intelligent identification system, set up modules such as menu bar 5, toolbar 6, display window 7, command line 8 and corresponding function buttons to meet the functional requirements such as the display of half-wall hole 3 image recognition results.

[0055] (7) In the menu bar 5 of the front-end display platform, first click on image import (or enter the corresponding command in the command line 8) to import the image to be identified into the recognition system; then, click on recognition operation (to connect to the back-end operation system), and the back-end of the system will automatically identify the half-wall hole 3 information of the image in combination with the set half-wall porosity calculation algorithm, so as to realize the automatic and accurate identification of a single pre-crack hole and a certain range of half-wall porosity; finally, click on recognition result output, and the recognition result can be displayed and viewed in the display window 7 in the form of a graph and a table. According to actual needs, the functions of the various buttons of the toolbar 6 can be used to view, download, print, etc. the recognition results.

[0056] The system and method for intelligently identifying the half-wall porosity of pre-splitting blasting provided by the embodiments of the present invention are applied in an actual environment. After verification by on-site pre-splitting blasting effect evaluation in a mine, the half-wall porosity recognition accuracy of the technology of the present invention reaches more than 95%, and the labor intensity of personnel working in the traditional way is greatly reduced, the efficiency of pre-splitting blasting quality evaluation is significantly improved, and good practical application effects are achieved.

[0057] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent identification system for half-wall porosity of pre-splitting blasting, characterized in that: The invention comprises a front-end display platform and a back-end operation system for data exchange through a computer, and the operation result information of the back-end operation system is displayed and viewed through the front-end display platform; the front-end display platform is composed of a menu bar (5), a tool bar (6), a display window (7) and a command line (8); the menu bar (5) comprises an image import module, a recognition operation module and a recognition result output module, the image import module is used for inputting photos of the half-wall hole (3), the recognition operation module is connected to the back-end operation system; the recognition result output module is used for viewing the operation result; the tool bar (6) comprises rotation, size measurement, view, columnar The display window (7) includes a display window (7) for displaying a half-wall hole (3) photo image, a half-wall porosity recognition result diagram, and a half-wall porosity recognition result table in the form of a diagram; the command line (8) is used to input operation commands and corresponds to the functions of each button of the tool bar (6); the back-end computing system includes a three-dimensional image database of the half-wall hole (3), an image denoising module, an image segmentation module, a data statistical analysis module, and a recognition calculation module, and the recognition accuracy of the half-wall porosity is gradually improved by continuously correcting the recognition algorithm.

2. A method for intelligently identifying half-wall porosity of pre-splitting blasting, based on the intelligent identification system for half-wall porosity of pre-splitting blasting according to claim 1, characterized in that: The following steps are involved: S1: Using the UAV oblique photography technology with or without RTK real-time differential positioning function, according to the preset UAV flight route, collect centimeter-level positioning accuracy of the half-wall hole (3) reflecting the pre-splitting blasting effect of the rock slope; S2: Import the collected photos of the half-wall hole (3) of the pre-splitting blasting effect on site into the computer three-dimensional image processing software, and divide the half-wall hole (3) into segments of lengths l1, l2, l3, ... l n Mark the length and inclination information one by one; S3: Summarize the annotated half-wall holes (3) photo images to establish a half-wall hole three-dimensional image database containing pre-crack hole annotation information; S4: Based on computer deep learning technology, develop half-wall porosity intelligent recognition algorithm and recognition result front-end display platform, combine with manual measurement data comparison and analysis, and gradually correct half-wall porosity intelligent recognition algorithm to improve half-wall porosity recognition accuracy; S5: Import the image to be recognized into the front-end display platform. The back-end computing system automatically recognizes and computes the image through the intelligent recognition algorithm of half-wall porosity, and outputs the recognition result of the image in the form of a chart, which can be displayed, viewed, downloaded, and printed on the front-end display platform.

3. The method for intelligently identifying half-wall porosity of pre-splitting blasting according to claim 2, characterized in that: The overlapping degree of two adjacent photos collected in S1 is not less than 40%. The adjacent images are highly fused through the image similarity search and recognition method, and regional images of different sizes are synthesized as needed.

4. The method for intelligently identifying half-wall porosity of pre-splitting blasting according to claim 2, characterized in that: The real-time differential positioning function with or without RTK used in S1 can be flexibly selected according to actual needs: if the coordinates of the explosion area need to be located, a method with RTK function is used; if the positioning coordinate information is not needed, a method without RTK function is used.

5. The method for intelligently identifying half-wall porosity of pre-splitting blasting according to claim 2, characterized in that: The half-wall hole (3) is located on the upper step (1) and the lower step (2), and the upper step (1) and the lower step (2) are connected via a platform (4).

6. The method for intelligently identifying half-wall porosity of pre-splitting blasting according to claim 2, characterized in that: The half-wall porosity intelligent identification algorithm in S4 includes the segment length identification of the half-wall hole (3) and the statistical analysis of the half-wall porosity. According to the actual needs on site, the half-wall porosity η1 of a single pre-splitting hole or the half-wall porosity η2 of pre-splitting holes within a certain range can be identified, where η1 and η2 reflect the local and overall pre-splitting blasting quality respectively; the calculation methods of η1 and η2 are as follows: For the half-wall porosity η1 of a single pre-crack hole: if the half-wall hole (3) of the same pre-crack hole after blasting is divided into n segments, the segment lengths are l1, l2, l3, ...l n , the total length ∑l needs to be accumulated, that is: ∑l=l1+l2+l3+...+l n , n is the number of segments, The calculation algorithm of the half-wall porosity η1 is set in the back-end computing system: η1 = L1 / ∑l; L1 is the actual total length of a single pre-crack hole formed by drilling; For the half-wall porosity η2 of the pre-cracked holes within a certain range: the sum of the segment lengths of each half-wall hole (3) is accumulated, that is: ∑L=∑l1+∑l2+∑l3+...+∑l m , m is the number of pre-splitting holes; The calculation algorithm of the half-wall porosity η2 is set in the back-end computing system: η2=L0 / ∑L; L0 is the actual total length of all pre-crack holes formed by drilling.

7. A method for intelligently identifying half-wall porosity of pre-splitting blasting according to claim 6, characterized in that: When identifying and analyzing the half-wall porosity η2 of pre-crack holes within a certain range, the image search, recognition and fusion technology of drone oblique photography is used to splice and synthesize the images taken each time, and synthesize three-dimensional images of different sizes according to actual needs, and then import them into the intelligent recognition system for recognition calculation and result output.

Images