CN115731431A - Tunnel blast hole identification method, device, equipment and storage medium - Google Patents
Tunnel blast hole identification method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN115731431A CN115731431A CN202211485799.9A CN202211485799A CN115731431A CN 115731431 A CN115731431 A CN 115731431A CN 202211485799 A CN202211485799 A CN 202211485799A CN 115731431 A CN115731431 A CN 115731431A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- target detection
- detection model
- rock wall
- training
- 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
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a tunnel blasthole identification method, which comprises the following steps of S1, acquiring a tunnel rock wall image marked with a blasthole; s2, dividing all tunnel rock wall images marked with blastholes into a training set and a verification set according to a set proportion; s3, inputting the data in the training set into a target detection model for training; s4, after iterative training for a set number of times, obtaining a target detection model which is best represented in the verification set; s5, acquiring a real-time tunnel rock wall image, and inputting the real-time tunnel rock wall image into a target detection model which is best represented in a verification set; and S6, judging whether the image output by the target detection model contains a labeling frame and a label or not, and obtaining a recognition result. The invention can improve the efficiency of identifying the blast hole. The invention also provides a tunnel blasthole identification method, a device, equipment and a storage medium, which have corresponding effects.
Description
Technical Field
The invention relates to the field of image recognition, in particular to a tunnel blasthole recognition method, a tunnel blasthole recognition device, tunnel blasthole recognition equipment and a storage medium.
Background
In tunnel engineering, a multi-arm drill jumbo and manual auxiliary drilling are usually adopted for blasting and drilling, the workload is high, and a large amount of manpower and material resources are usually consumed for drilling and charging. The number of blastholes and the flatness of the contour of surrounding rocks directly influence the blasting effect of rock drilling. Adopting a manual handheld air drill to divide a pilot tunnel drill hole into small sections, adopting a cut with a certain inclination angle for partial drill holes, and enabling the dip angle of the cut to be between 55 and 75 degrees; in the prior art, the problem of low identification efficiency exists when a shot hole is identified in a manual mode.
Disclosure of Invention
In view of the above, it is desirable to provide a tunnel borehole identification method, apparatus, device, and storage medium that can improve the efficiency of borehole identification.
In a first aspect, the present invention provides a tunnel blasthole identification method, including:
s1, acquiring a tunnel rock wall image marked with a blast hole;
s2, dividing all tunnel rock wall images marked with blastholes into a training set and a verification set according to a set proportion;
s3, inputting the data in the training set into a target detection model for training;
s4, after iterative training for a set number of times, obtaining a target detection model which is best represented in the verification set;
s5, acquiring a real-time tunnel rock wall image, and inputting the real-time tunnel rock wall image into a target detection model which is best represented in a verification set;
and S6, judging whether the image output by the target detection model contains a labeling frame and a label or not, and obtaining a recognition result.
In one embodiment, the image of the tunnel rock wall in S1 is labeled by labelImg, and the borehole is labeled as hole.
In one embodiment, in S2, the set ratio is 9.
In one embodiment, the target detection model comprises a Yolov5-s target detection model.
In one embodiment, the S4 includes:
and comparing the training set with the verification set through the mAP to obtain the score of the target detection model, and obtaining the target detection model with the highest score.
In one embodiment, the S6 includes:
if the image output by the target detection model comprises the labeling frame and the label, indicating that the real-time tunnel rock wall image comprises the blast hole;
and if the image output by the target detection model does not contain the labeling frame and the label, indicating that the real-time tunnel rock wall image does not contain the blast hole.
In a second aspect, the invention provides a tunnel blasthole identification device, which comprises a data acquisition module, a division module, a training module, an optimal model acquisition module, a real-time detection module and a judgment module;
the data acquisition module is used for acquiring the tunnel rock wall image marked with the blast hole;
the dividing module is used for dividing all tunnel rock wall images marked with blast holes into a training set and a verification set according to a set proportion;
the training module is used for inputting the data in the training set into the target detection model for training;
the optimal model acquisition module is used for acquiring a target detection model which is best represented in the verification set after iterative training for a set number of times;
the real-time detection module is used for acquiring a real-time tunnel rock wall image and inputting the real-time tunnel rock wall image into a target detection model which is best represented in the verification set;
the judging module is used for judging whether the image output by the target detection model contains the labeling frame and the label or not to obtain an identification result.
In a third aspect, the present invention provides a tunnel blasthole identification device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the tunnel blasthole identification method according to any one of the above items when executing the computer program.
In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of the tunnel bore identification method of any one of the preceding claims.
The invention has the beneficial effects that:
the invention greatly improves the efficiency of tunnel face blasthole identification and charging in the tunnel engineering construction process, realizes the automation of tunnel blasting construction, improves the tunnel face construction safety and improves the drilling accuracy.
Drawings
The invention is further illustrated by means of the attached drawings, but the embodiments in the drawings do not constitute any limitation to the invention, and for a person skilled in the art, without inventive effort, further drawings may be derived from the following figures.
Fig. 1 is a diagram of an embodiment of a tunnel blasthole identification method according to the invention.
FIG. 2 is a block diagram of the FCOS network model of the present invention.
FIG. 3 is a diagram of an embodiment of a Yolov5-s target detection model according to the present invention.
FIG. 4 is a diagram of PR-Curve trained in the present invention.
FIG. 5 is a schematic diagram of the model accuracy detection of the present invention.
Fig. 6 is a diagram of an embodiment of a tunnel blasthole identification device according to the invention.
Detailed Description
In order that those skilled in the art will better understand the disclosure, the following detailed description is given with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In a first aspect, referring to fig. 1, the present invention provides a tunnel blasthole identification method, including:
s1, acquiring a tunnel rock wall image marked with a blast hole;
in this step, a large number of images of the tunnel wall containing the blastholes are collected in advance and then manually labeled.
S2, dividing all tunnel rock wall images marked with blastholes into a training set and a verification set according to a set proportion;
the feature graph is introduced into an FCOS network model, which is based on an Anchor-free method, and the biggest difference is that the bounding box can be quickly regressed without thousands of prior frames. The network structure is shown in FIG. 2, and because the Center-new branch is introduced, a target cannot predict a plurality of frames, and the deviation caused by too close drilling positions is effectively avoided.
S3, inputting the data in the training set into a target detection model for training;
s4, after iterative training for a set number of times, obtaining a target detection model which is best represented in the verification set;
after training is completed, model weight of image recognition can be obtained. The method comprises the steps of acquiring images by using an intelligent sensing terminal, reasoning position information of a drill hole by using model weights trained by an FCOS target detector, and accordingly segmenting the drill hole and transmitting the segmented drill hole into the model weights trained by the Yolov5-s detector.
Alternatively, the set number of times may be 3000 times.
S5, acquiring a real-time tunnel rock wall image, and inputting the real-time tunnel rock wall image into a target detection model which is best represented in a verification set;
the method comprises the steps that a camera device loaded on a tunnel excavating machine obtains a real-time tunnel rock wall image, the image is input into a target detection model, the target detection model obtains position information of a shot hole, the shot hole is marked with a rectangular frame, and the position information is marked as 'the shot hole is contained in the image and the position is marked in the image', and if the image is judged not to contain the shot hole, no mark is made.
And S6, judging whether the image output by the target detection model contains a labeling frame and a label or not, and obtaining a recognition result.
In one embodiment, the image of the tunnel rock wall in S1 is labeled by labelImg, and the borehole is labeled as hole.
In one embodiment, in S2, the set ratio is 9.
In one embodiment, the target detection model comprises a Yolov5-s target detection model. The structure of the Yolov5-s target detection model is shown in fig. 3.
By adopting the Yolov5-s target detection, each batch of training data can be transmitted through the data loader, and meanwhile, the training data is enhanced, so that the iteration efficiency is higher. The data loader performs three types of data enhancement: zooming, color space adjustment and mosaic enhancement; the method can effectively solve the problem of the most painful small object in model training.
yolov5 uses the Pythroch frame, is very friendly to the user, can conveniently train own dataset, can be light turn into the ONXX format that the android used with Pythroch weight file, then can convert the use format into OpenCV, directly deploy to the cell-phone application end.
In one embodiment, the S4 includes:
and comparing the training set with the verification set through the mAP to obtain the score of the target detection model, and obtaining the target detection model with the highest score.
Specifically, a PR-Curve graph as shown in fig. 4 is generated in a file after training is finished in the yolov5 algorithm, a training set and a verification set are compared through an mAP (mean Average Precision), and a score (score) is returned, wherein the higher the score is, the higher the detection Precision of the model is. Through tests, the IoU threshold value of the mAP is 0.5, and the tunnel blasthole identification is effective and well implemented, namely mAP50; ioU is used for filtering redundant frames, and detection frames with detection frame overlapping parts larger than an IoU threshold value are filtered, so that the detection frame with the highest confidence coefficient is left. The schematic diagram of the accuracy detection is shown in fig. 5.
If:
(1) the IOU of the prediction box and the real box is greater than the IOU threshold (= 0.5), then this sample is the true instance TP =1;
(2) if less than the threshold, no label box on the box results in FN =1 and a wrong prediction box results in FP =1.
On this basis, the confidence Score is used to draw the number of prediction boxes, calculate the point on the PR-Curve graph, and calculate the AP value according to the area under the Curve.
The accuracy is as follows: p = TP/P = correct recognition/total prediction
And (4) recall rate: r = TP/TP + FN = correct recognition/total true
In one embodiment, the S6 includes:
if the image output by the target detection model contains the labeling frame and the label, indicating that the real-time tunnel rock wall image contains the blast hole;
and if the image output by the target detection model does not contain the labeling frame and the label, indicating that the real-time tunnel rock wall image does not contain the blast hole.
In a second aspect, the present invention provides a tunnel blasthole identification apparatus, as shown in fig. 6, the apparatus includes a data acquisition module, a division module, a training module, an optimal model acquisition module, a real-time detection module, and a judgment module;
the data acquisition module is used for acquiring the tunnel rock wall image marked with the blast hole;
the dividing module is used for dividing all the tunnel rock wall images marked with the blastholes into a training set and a verification set according to a set proportion;
the training module is used for inputting the data in the training set into the target detection model for training;
the optimal model acquisition module is used for acquiring a target detection model which is best represented in the verification set after iterative training for a set number of times;
the real-time detection module is used for acquiring a real-time tunnel rock wall image and inputting the real-time tunnel rock wall image into a target detection model which is best represented in the verification set;
the judging module is used for judging whether the image output by the target detection model contains the labeling frame and the label or not to obtain an identification result.
In a third aspect, the present invention provides a tunnel blasthole identification device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the tunnel blasthole identification method according to any one of the above items when executing the computer program.
In a fourth aspect, the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, carries out the steps of a tunnel borehole identification method as defined in any one of the preceding claims.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above may be implemented by hardware instructions of a computer program, and the computer program may be stored in a non-volatile computer-readable storage medium, and when executed, may include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), rambus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not to be construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, and these are all within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (9)
1. A tunnel blasthole identification method is characterized by comprising the following steps:
s1, acquiring a tunnel rock wall image marked with a blast hole;
s2, dividing all tunnel rock wall images marked with blastholes into a training set and a verification set according to a set proportion;
s3, inputting the data in the training set into a target detection model for training;
s4, after iterative training for a set number of times, obtaining a target detection model which is best represented in the verification set;
s5, acquiring a real-time tunnel rock wall image, and inputting the real-time tunnel rock wall image into a target detection model which is best represented in a verification set;
and S6, judging whether the image output by the target detection model contains the labeling frame and the label or not, and obtaining a recognition result.
2. The system for identifying tunnel blastholes based on deep learning as claimed in claim 1, wherein the image of the tunnel rock wall in S1 is labeled by labelImg, and the blastholes are labeled as holes.
3. The deep learning-based tunnel borehole identification system according to claim 1, wherein in S2, the set ratio is 9.
4. The deep learning-based tunnel borehole identification system according to claim 1, wherein the target detection model comprises a Yolov5-s target detection model.
5. The deep learning-based tunnel borehole identification system according to claim 1, wherein said S4 comprises:
and comparing the training set with the verification set through the mAP to obtain the score of the target detection model, and obtaining the target detection model with the highest score.
6. The deep learning-based tunnel borehole identification system according to claim 1, wherein said S6 comprises:
if the image output by the target detection model contains the labeling frame and the label, indicating that the real-time tunnel rock wall image contains the blast hole;
and if the image output by the target detection model does not contain the labeling frame and the label, indicating that the real-time tunnel rock wall image does not contain the blast hole.
7. A tunnel blasthole recognition device is characterized by comprising a data acquisition module, a division module, a training module, an optimal model acquisition module, a real-time detection module and a judgment module;
the data acquisition module is used for acquiring the tunnel rock wall image marked with the blast hole;
the dividing module is used for dividing all tunnel rock wall images marked with blast holes into a training set and a verification set according to a set proportion;
the training module is used for inputting the data in the training set into the target detection model for training;
the optimal model acquisition module is used for acquiring a target detection model which is best represented in the verification set after iterative training for a set number of times;
the real-time detection module is used for acquiring a real-time tunnel rock wall image and inputting the real-time tunnel rock wall image into a target detection model which is best represented in the verification set;
the judging module is used for judging whether the image output by the target detection model contains the labeling frame and the label or not and obtaining the identification result.
8. A tunnel borehole identification device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method according to any one of claims 1 to 6.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211485799.9A CN115731431A (en) | 2022-11-24 | 2022-11-24 | Tunnel blast hole identification method, device, equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211485799.9A CN115731431A (en) | 2022-11-24 | 2022-11-24 | Tunnel blast hole identification method, device, equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115731431A true CN115731431A (en) | 2023-03-03 |
Family
ID=85298106
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211485799.9A Pending CN115731431A (en) | 2022-11-24 | 2022-11-24 | Tunnel blast hole identification method, device, equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115731431A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116721355A (en) * | 2023-08-09 | 2023-09-08 | 江西云眼视界科技股份有限公司 | Billboard detection method, billboard detection system, readable storage medium and computer equipment |
-
2022
- 2022-11-24 CN CN202211485799.9A patent/CN115731431A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116721355A (en) * | 2023-08-09 | 2023-09-08 | 江西云眼视界科技股份有限公司 | Billboard detection method, billboard detection system, readable storage medium and computer equipment |
CN116721355B (en) * | 2023-08-09 | 2023-10-24 | 江西云眼视界科技股份有限公司 | Billboard detection method, billboard detection system, readable storage medium and computer equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR102645202B1 (en) | Method and apparatus for machine learning | |
US20190234745A1 (en) | Method for estimating position of ego vehicle for autonomous driving and autonomous driving apparatus | |
JP6908944B2 (en) | Leverage target object integration networks and target object prediction networks to learn CNN-based object detectors applicable to user requirements such as key performance indicators for use in multiple camera or surround view monitoring Method and learning device, and testing method and testing device using this | |
CN114981840A (en) | Ground segmentation method and device based on point cloud data and computer equipment | |
CN109858476B (en) | Tag expansion method and electronic equipment | |
CN108038544B (en) | Neural network deep learning method and system based on big data and deep learning | |
JP2019056668A (en) | Information processor, information processing system, method for processing information, and information processing program | |
CN112561859B (en) | Monocular vision-based steel belt drilling and anchor net identification method and device for anchoring and protecting | |
CN115731431A (en) | Tunnel blast hole identification method, device, equipment and storage medium | |
CN114781272A (en) | Carbon emission prediction method, device, equipment and storage medium | |
CN113505261B (en) | Data labeling method and device and data labeling model training method and device | |
CN112037180B (en) | Chromosome segmentation method and device | |
CN110580507B (en) | City texture classification and identification method | |
CN110097077B (en) | Point cloud data classification method and device, computer equipment and storage medium | |
CN113792718B (en) | Method for positioning face area in depth map, electronic device and storage medium | |
CN117765708A (en) | Slope instability prediction method and device based on map image analysis technology | |
US10311338B1 (en) | Learning method, learning device for detecting lanes on the basis of CNN and testing method, testing device using the same | |
CN116363319B (en) | Modeling method, modeling device, equipment and medium for building roof | |
CN112241705A (en) | Target detection model training method and target detection method based on classification regression | |
CN114580044A (en) | Building outer surface data acquisition method, building outer surface data acquisition device, computer equipment and medium | |
CN115620083B (en) | Model training method, face image quality evaluation method, equipment and medium | |
CN115965787A (en) | Remote sensing image interactive segmentation method, computer equipment and storage medium | |
CN112199975A (en) | Identity verification method and device based on human face features | |
CN115392407A (en) | Method, device, equipment and medium for warning danger source based on unsupervised learning | |
CN116129158A (en) | Power transmission line iron tower small part image recognition method and device |
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 |