Blasting effect analysis method and device based on blasthole scratches and electronic equipment

Abstract

The invention relates to a blasting effect analysis method and device based on blasthole residues and electronic equipment, and the method comprises the steps: obtaining blasthole residue pictures of a plurality of to-be-detected blastholes after single blasting in a current blasting scheme in a target area; inputting the to-be-detected blast hole residue picture into a pre-constructed semantic segmentation model to obtain a segmentation effect picture; and calculating the half-hole rate of the blast holes corresponding to the segmentation effect picture so as to determine the blasting effect of the current blasting scheme based on the statistical result of the half-hole rate. Therefore, the technical problems that in the related technology, the accuracy of field measurement is poor, parameters need to be readjusted when facing different scenes through a three-dimensional laser scanning and clustering combined method, the steps are tedious, the efficiency is low, and follow-up blasting analysis is affected are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image data processing, in particular to a blasting effect analysis method and device based on blasting hole residual marks and electronic equipment. BACKGROUND

[0002] In the excavation and shaping of slopes, underground rock mass, etc., smooth and presplit blasting as contour blasting technology is the main means to form a permanent contour. They involve drilling a row of closely spaced holes along the preset contour line, and then loading the same diameter (50-101mm) explosives as the drilled hole. In contour excavation shaping, the calculation of blasting design parameters depends on empirical formula and engineering analogy methods, and the phenomenon of overbreak and underbreak and arbitrary adjustment of blasting parameters occurs from time to time. More seriously, unreasonable parameter design may cause tunnel collapse, causing personnel and economic property losses. Therefore, after the rock mass is excavated, the tunnel shaping quality needs to be recorded and evaluated in time to optimize the blasting design and adjust the blasting parameters of the next stage, so as to ensure the contour excavation quality and construction safety. The blasting hole residual marks and half-hole rate after smooth and presplit blasting are important basis for optimizing the blasting parameters, and to some extent reflect the excavation quality and overbreak and underbreak of the rock mass.

[0003] In the related art, the calculation of blasting hole residual marks and half-hole rate mainly depends on the measurement by the on-site engineer with a ladder or the digital photography archive method, but the accuracy and efficiency of identification are often low. The method of using three-dimensional laser scanning and combining clustering method for identification is not end-to-end, and needs to be re-scanned and clustered when facing other scenes, which is tedious and inefficient, and affects the subsequent blasting effect analysis results, and needs to be improved. SUMMARY

[0004] The present application provides a blasting effect analysis method and device based on blasting hole residual marks and electronic equipment to solve the technical problems of poor accuracy of on-site measurement, the need to adjust parameters when facing different scenes, tedious steps and low efficiency of the method of three-dimensional laser scanning combined with clustering in the related art, and the impact on subsequent blasting analysis.

[0005] The first aspect embodiment of the present application provides a blasting effect analysis method based on blasting hole residual marks, comprising the following steps: acquiring a blasting hole residual mark picture of a plurality of to-be-detected blasting holes after single blasting of a target area under a current blasting scheme; inputting the to-be-detected blasting hole residual mark picture into a pre-constructed semantic segmentation model to obtain a segmentation effect picture; calculating the half-hole rate of the blasting hole corresponding to the segmentation effect picture to determine the blasting effect of the current blasting scheme based on the statistical result of the half-hole rate.

[0006] Optionally, in an embodiment of the present application, before the blast hole residual mark picture to be detected is input into the pre-constructed semantic segmentation model, further comprising: acquiring a plurality of blast mark images photographed in a sample area, and respectively labeling the plurality of blast mark images to obtain real data of each blast mark image; constructing a blast hole residual mark data set using the plurality of blast mark images and the real data, and dividing the blast hole residual mark data set into a training set, a validation set and a test set; training a plurality of initial semantic segmentation models pre-constructed using the training set to obtain a plurality of trained semantic segmentation models, and verifying the plurality of trained semantic segmentation models using the validation set to obtain a plurality of actual semantic segmentation models; testing the segmentation performance of the plurality of actual semantic segmentation models using the test set, and screening a semantic segmentation model satisfying a preset effective condition from the plurality of actual semantic segmentation models using the segmentation performance.

[0007] Optionally, in an embodiment of the present application, the testing of the segmentation performance of the plurality of actual semantic segmentation models using the test set comprises: acquiring test data output by each actual semantic segmentation model; calculating a quantitative index of each semantic segmentation model based on the test data and corresponding real data to determine the segmentation performance based on the quantitative index.

[0008] Optionally, in an embodiment of the present application, the calculation expression of the quantitative index is:

[0009]

[0010] wherein, IoU represents the intersection over union, P represents any region in the test data, G represents a region corresponding to the any region in the real data, Precision represents the precision, TP represents the true positive, FP represents the false positive, FN represents the false negative, Recall represents the recall, and F1 represents the harmonic mean of the precision and the recall.

[0011] Optionally, in an embodiment of the present application, the calculation expression of the half-hole rate is:

[0012]

[0013] wherein, H a represents the half-hole rate, N represents the actual number of drill holes of the target region, L i / L represents the half-hole rate of a single blast hole.

[0014] The second aspect embodiment of the present application provides a blasting effect analysis device based on blasting hole residual marks, comprising: a first acquisition module configured to acquire blasting hole residual mark pictures of a plurality of to-be-detected blasting holes after a single blasting of a target area under a current blasting scheme; a segmentation module configured to input the to-be-detected blasting hole residual mark pictures into a pre-constructed semantic segmentation model to obtain a segmentation effect map; and an analysis module configured to calculate a half-hole rate of the blasting hole corresponding to the segmentation effect map, and determine a blasting effect of the current blasting scheme based on a statistical result of the half-hole rate.

[0015] Optionally, in an embodiment of the present application, the device further comprises: a second acquisition module configured to acquire a plurality of crater images photographed by a sample area, and label the plurality of crater images respectively to obtain true data of each crater image; a construction module configured to construct a blasting hole residual mark data set by using the plurality of crater images and the true data, and divide the blasting hole residual mark data set into a training set, a validation set and a test set; a training module configured to train a plurality of initial semantic segmentation models pre-constructed by using the training set to obtain a plurality of trained semantic segmentation models, and verify the plurality of trained semantic segmentation models by using the validation set to obtain a plurality of actual semantic segmentation models; and a test module configured to test segmentation performance of the plurality of actual semantic segmentation models by using the test set, and select a semantic segmentation model satisfying a preset effective condition from the plurality of actual semantic segmentation models by using the segmentation performance.

[0016] Optionally, in an embodiment of the present application, the test module comprises: an acquisition unit configured to acquire test data output by each actual semantic segmentation model; and a calculation unit configured to calculate a quantitative index of each semantic segmentation model based on the test data and corresponding true data, and determine the segmentation performance based on the quantitative index.

[0017] Optionally, in an embodiment of the present application, a calculation expression of the quantitative index is:

[0018]

[0019] wherein, IoU represents an intersection over union, P represents any region in the test data, G represents a region corresponding to the any region in the true data, Precision represents an accuracy rate, TP represents a true positive, FP represents a false positive, FN represents a false negative, Recall represents a recall rate, and F1 represents a harmonic mean of the accuracy rate and the recall rate.

[0020] Optionally, in an embodiment of the present application, a calculation expression of the half-hole rate is:

[0021]

[0022] wherein H a represents the half-hole rate, N represents the actual number of drilled holes of the target area, L i represents the half-hole rate of a single blast hole.

[0023] The third aspect of the present application provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the blast effect analysis method based on blast hole residues as described in the above embodiments.

[0024] The fourth aspect of the present application provides a computer readable storage medium, which stores computer instructions for causing the computer to execute the blast effect analysis method based on blast hole residues as described in the above embodiments.

[0025] The fifth aspect of the present application provides a computer program product comprising a computer program, which, when executed, implements the blast effect analysis method based on blast hole residues as described above.

[0026] The embodiments of the present application can use a pre-constructed semantic segmentation model to perform semantic segmentation on the blast hole residue pictures to obtain segmentation effect pictures, and then calculate the half-hole rate of the blast holes corresponding to the segmentation effect pictures to determine the blast effect of the current blasting scheme based on the statistical results of the half-hole rate. The blast hole residues can be quickly identified and the half-hole rate can be calculated to further evaluate the blast effect and optimize the subsequent blasting parameters, which is beneficial to dynamically adjust the blasting design and construction and reduce the construction period and cost. Thus, the technical problem of poor accuracy of on-site measurement in the related art, the need to re-adjust parameters when facing different scenarios by using the three-dimensional laser scanning combined with clustering method, the tedious steps, and the low efficiency, which affect the subsequent blasting analysis, are solved.

[0027] Additional aspects and advantages of the present application will be made apparent by the following description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0028] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description, taken in conjunction with the accompanying drawings, in which:

[0029] Figure 1 A flowchart of a blast effect analysis method based on blast hole residues according to an embodiment of the present application is provided.

[0030] Figure 2 A flowchart of a blast effect analysis method based on blast hole residues according to an embodiment of the present application is provided.

[0031] Figure 3 A schematic diagram of a blast hole residual mark on a slope according to an embodiment of the present application;

[0032] Figure 4 A schematic diagram of a blast hole residual mark on a high side wall according to an embodiment of the present application;

[0033] Figure 5 A schematic diagram of a blast hole residual mark on a tunnel peripheral hole according to an embodiment of the present application;

[0034] Figure 6 A schematic diagram of identifying a minimum circumscribed rectangle of a blast hole residual mark result according to an embodiment of the present application;

[0035] Figure 7 A structural schematic diagram of a blast effect analysis device based on a blast hole residual mark according to an embodiment of the present application;

[0036] Figure 8 A structural schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] Embodiments of the present application are described in detail below with reference to the accompanying drawings, in which the same or similar elements or elements having the same or similar functions are denoted by the same or similar reference numerals throughout. The embodiments described below by reference to the accompanying drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0038] The blast effect analysis method, device and electronic device based on a blast hole residual mark according to the embodiments of the present application are described below with reference to the accompanying drawings. In view of the technical problems in the related art mentioned above, the accuracy of field measurement is poor, the method of three-dimensional laser scanning combined with clustering needs to adjust parameters again when facing different scenes, the steps are complicated, the efficiency is low, and the subsequent blasting analysis is affected. The present application provides a blast effect analysis method based on a blast hole residual mark. In the method, a semantic segmentation model can be used to perform semantic segmentation on a blast hole residual mark picture to obtain a segmentation effect picture, and then the half-hole rate of the blast hole corresponding to the segmentation effect picture is calculated to determine the blast effect of the current blasting scheme based on the statistical result of the half-hole rate. The blast hole residual mark can be quickly identified and the half-hole rate can be calculated to further evaluate the blast effect and optimize the subsequent blasting parameters, which is beneficial to dynamically adjust the blasting design and construction and reduce the construction period and cost. Thus, the technical problems in the related art that the accuracy of field measurement is poor, the method of three-dimensional laser scanning combined with clustering needs to adjust parameters again when facing different scenes, the steps are complicated, the efficiency is low, and the subsequent blasting analysis is affected are solved.

[0039] Specifically, Figure 1A flowchart of a blasting effect analysis method based on blasting hole residual marks provided by an embodiment of the present application.

[0040] As shown in the figure, the blasting effect analysis method based on blasting hole residual marks includes the following steps: Figure 1

[0041] In step S101, the blasting hole residual mark pictures of a plurality of to-be-detected blasting holes after single blasting of a target area under a current blasting scheme are acquired.

[0042] In actual execution, the embodiment of the present application can collect the corresponding blasting hole residual mark pictures after single blasting, wherein the target area is the blasting occurrence area, and the blasting hole residual marks can include the blasting hole residual marks on the slope, the blasting hole residual marks on the high side wall of the underground powerhouse, and the blasting hole residual marks on the surrounding holes of the tunnel, etc. The blasting hole residual marks on the surrounding holes of the tunnel refer to the comprehensive blasting hole residual marks formed along the tunnel face, including the surrounding contour of the traffic hole and the middle adit, and the top contour of the underground powerhouse space.

[0043] By acquiring the blasting hole residual mark pictures of a plurality of to-be-detected blasting holes after single blasting under the current blasting scheme, the embodiment of the present application can perform residual mark semantic segmentation for subsequent blasting analysis.

[0044] In step S102, the to-be-detected blasting hole residual mark pictures are input into the pre-constructed semantic segmentation model to obtain a segmentation effect picture.

[0045] Further, the embodiment of the present application can input the clear to-be-detected blasting hole residual mark pictures into the semantic segmentation model to generate a segmentation effect picture for subsequent half-hole rate calculation.

[0046] Optionally, in an embodiment of the present application, before the to-be-detected blasting hole residual mark pictures are input into the pre-constructed semantic segmentation model, it further includes: acquiring a plurality of blast mark images shot by a sample area, and respectively labeling the plurality of blast mark images to obtain the true data of each blast mark image; constructing a blasting hole residual mark data set using the plurality of blast mark images and the true data, and dividing the blasting hole residual mark data set into a training set, a validation set, and a test set; training the pre-constructed plurality of initial semantic segmentation models using the training set to obtain a plurality of training semantic segmentation models, and verifying the plurality of training semantic segmentation models using the validation set to obtain a plurality of actual semantic segmentation models; testing the segmentation performance of the plurality of actual semantic segmentation models using the test set, and screening a semantic segmentation model meeting a pre-set effective condition from the plurality of actual semantic segmentation models using the segmentation performance.

[0047] ​For example, the embodiment of the present application can use a high-definition camera to shoot a high-definition image of a blast mark on a slope and an underground chamber. The shot blast mark image can include a blast hole residual mark on a slope, a blast hole residual mark on a high side wall of an underground powerhouse, and a blast hole residual mark on a hole around a tunnel.

[0048] The embodiment of the present application can use a data enhancement operation to obtain more blast hole residual mark pictures, use LabelMe to obtain the true situation (real data) of the blast hole residual mark, and use the real data for the input and output of the semantic segmentation model. Then, the blast hole residual mark dataset is constructed by combining a plurality of blast mark images and real data.

[0049] Further, the embodiment of the present application can divide the dataset into a training set, a validation set, and a test set according to a certain proportion.

[0050] In the process of obtaining the semantic segmentation model for identifying the blast hole residual mark, the embodiment of the present application can use the training set and the validation set data to train and verify the model. The optimal semantic segmentation model for identifying the blast hole residual mark is obtained by repeatedly training and verifying the model through optimizing the model hyperparameters and improving the model network structure, that is, the segmentation performance of a plurality of actual semantic segmentation models is tested by using the test set, and the semantic segmentation model with higher accuracy and suitable for blast hole residual mark identification is selected from the plurality of actual semantic segmentation models, that is, the optimal semantic segmentation model.

[0051] Optionally, in an embodiment of the present application, the segmentation performance of a plurality of actual semantic segmentation models is tested by using the test set, including: obtaining test data output by each actual semantic segmentation model; calculating a quantitative index of each semantic segmentation model based on the test data and the corresponding real data, to determine the segmentation performance based on the quantitative index. Wherein, the calculation expression of the quantitative index is:

[0052]

[0053] Wherein, IoU represents the intersection over union, P represents any region in the test data, G represents a region corresponding to any region in the real data, Precision represents the precision, TP represents the true positive, FP represents the false positive, FN represents the false negative, Recall represents the recall, and F1 represents the harmonic mean of the precision and the recall.

[0054] The optimal semantic segmentation model is selected by using the quantitative index and the visualized comparison result on the divided test set. The quantitative features are analyzed by using the intersection over union (IoU), the precision (Precision), the recall (Recall), and the F1-score. The evaluation effect formula of the quantitative index of the segmentation performance is as follows:

[0055]

[0056] Wherein, P represents a predicted sample, i.e. the blast hole residual mark area of the test data (i.e. the model prediction result), G represents a true sample, i.e. the blast hole residual mark area of the true data (i.e. the true situation); TP, FN and FP represent true positive, false positive and false negative respectively, and specifically, TP can represent that the test data (i.e. the model prediction result) is a blast hole residual mark pixel, and the true data (i.e. the true situation) is also the blast hole residual mark pixel; FP represents that the prediction result is a blast hole residual mark pixel, and the true situation is not the blast hole residual mark pixel; FN represents that the true situation is a blast hole residual mark pixel, and the prediction result is not the blast hole residual mark pixel.

[0057] As a possible implementation manner, the three kinds of blast hole residual marks obtained by comparing the true situation with the model recognition on the test set are pixel-level quantified, and the model recognition effect can be explained from the pixel level. The obtained optimal semantic segmentation model can be used to recognize the blast hole residual mark of the test set, and the three kinds of blast hole residual marks of the true situation are pixel counted. Each kind of blast hole residual mark is compared separately, the horizontal coordinate is the true situation, the vertical coordinate is the prediction situation, and the deviation degree between y=x is observed to quantify the pixel situation of the recognized blast hole residual mark.

[0058] The length and width of the blast hole residual mark of the true situation and the recognition result on the test set are compared to determine whether the semantic segmentation result can be used for subsequent half-hole rate analysis. The length and width of the blast hole residual mark are obtained by the minimum circumscribed rectangle. The total width or total length of the blast hole residual mark in each picture recognized by the embodiment of the application is the horizontal coordinate of the true situation, and the vertical coordinate is the prediction situation. The deviation degree between y=x is observed to quantify the length and width of the recognition.

[0059] Through the pixel situation and the length and width situation, it can be comprehensively judged whether the semantic segmentation result can be used for subsequent half-hole rate analysis.

[0060] In step S103, the half-hole rate of each blast hole corresponding to the segmentation effect picture is calculated to determine the blasting effect of the current blasting scheme based on the statistical result of the half-hole rate. The calculation expression of the half-hole rate is:

[0061]

[0062] Wherein, H a represents the half-hole rate, N represents the actual number of drill holes in the target area, L i / L represents the half-hole rate of a single blast hole.

[0063] In actual implementation process, the embodiment of the present application can obtain the length of each identified blast hole residual mark by using the minimum circumscribed rectangle method according to the segmentation effect map; and remove the identification error noise and non-blast hole residual mark area in the picture according to the length-width ratio threshold and the pixel area threshold of the blast hole residual mark.

[0064] Further, in the excavation process, the embodiment of the present application can assume that the designed drilling length of each region is uniform, and the blast hole half-hole rate H a The calculation formula is as follows:

[0065]

[0066] wherein, H a represents the half-hole rate, N represents the actual number of drillings in the target region, L i / L represents the half-hole rate of a single blast hole.

[0067] Further, the embodiment of the present application can combine the blast crater distribution of a single excavation blast area, count and analyze the half-hole rate, evaluate the blast effect this time, and provide reasonable parameters and suggestions for subsequent blast design; wherein, for the detection of blast hole residual marks on the slope and the blast hole residual marks on the high side wall, the photograph is taken vertically to the wall surface; for the detection of blast hole residual marks of peripheral holes such as tunnels, the peripheral holes need to be photographed completely in different regions, and the overlapping area needs to be discarded when calculating the half-hole rate. Thus, the half-hole rate of a single blast hole and the region of a one-time blast area is obtained, wherein, the embodiment of the present application can superimpose the segmentation result and the half-hole rate data on the original picture to generate an intuitive analysis result.

[0068] In combination with Figures 2 to 6 as shown, the working principle of the blast effect analysis method based on blast hole residual marks of the embodiment of the present application is described in detail.

[0069] As shown in Figure 2 , the embodiment of the present application can include the following steps:

[0070] Step S201, constructing a blast hole residual mark data set.

[0071] The embodiment of the present application can use a high-definition camera to shoot the blast crater high-definition image in an underground chamber.

[0072] As shown in Figures 3-5 , the shot blast crater image is generally divided into two categories: blast hole residual marks on the slope, blast hole residual marks on the high side wall of the underground powerhouse, and blast hole residual marks on the peripheral holes of the tunnel. Note that the blast hole residual marks on the peripheral holes of the tunnel refer to a comprehensive blast hole residual mark formed along the tunnel face, including the peripheral contour of the traffic hole and the middle adit, and the top contour of the underground powerhouse space.

[0073] Data augmentation is employed to obtain more images of blast hole remnants. LabelMe is used to obtain the actual condition of the blast hole remnants, resulting in real data. This data is then combined with high-resolution images of blast holes to construct a database of blast hole remnants.

[0074] Step S202: Train and obtain a semantic segmentation model suitable for blast hole remnants.

[0075] The embodiments of this application can train and obtain a semantic segmentation model applicable to blast hole remnants, and divide the dataset into a training set, a validation set, and a test set according to a certain ratio.

[0076] The embodiments of this application can use training set and validation set data to train and validate the model. By optimizing the model hyperparameters and improving the model network structure, the training and validation process can be repeated to obtain the optimal semantic segmentation model for recognizing blast hole residues.

[0077] The optimal semantic segmentation model is selected by applying it to the partitioned test set and through quantitative metrics and visual comparison results. The quantitative metrics include Intersection over Union (IoU), Precision, Recall, and F1 score.

[0078] Step S203: Quantitatively obtain the blast hole remnants and the half-hole length and width of the remnants. Perform pixel-level quantization on the test set, comparing the real-world blast hole remnants with those obtained by the model, to explain the model's recognition performance at the pixel level; for example... Figure 6 As shown, this embodiment of the application can compare the length and width of the blast hole remnants in the actual situation on the test set with the recognition results to determine whether the semantic segmentation results can be used for subsequent half-pore rate analysis. The length and width of the blast hole remnants are obtained through the minimum bounding rectangle.

[0079] Step S204: Calculate the semi-porous ratio and apply it.

[0080] In this embodiment, a clear image of the blast hole remnant to be detected can be input into a semantic segmentation model to generate a segmentation result image. Based on the segmentation result image, the length of each identified blast hole remnant segment is obtained using the minimum bounding rectangle method. Based on the aspect ratio threshold and pixel area threshold of the blast hole remnant, recognition error noise and non-blast hole remnant areas in the image are removed.

[0081] During the excavation process, this embodiment of the application can assume that the design borehole length is uniform in each area, and then calculate the half-hole ratio of the blasting holes.

[0082] Step S205: Practical application of blast hole residues on slopes, high sidewalls, and around tunnels.

[0083] Further, the embodiment of the present application can combine the blast hole distribution of single excavation blasting area, count and analyze the half-hole rate, evaluate the blasting effect of this time, and provide reasonable parameters and suggestions for subsequent blasting design; wherein, for the detection of blast hole residual marks on the slope and the blast hole residual marks on the high side wall, the photograph is taken vertically to the wall surface; for the detection of blast hole residual marks of peripheral holes such as tunnels, the peripheral holes need to be photographed completely in different areas, and the overlapping area needs to be discarded when calculating the half-hole rate. Thus, the single blast hole and the half-hole rate of the area of the first blasting area are obtained, wherein, the embodiment of the present application can superimpose the segmentation result and the half-hole rate data on the original picture to generate an intuitive analysis result.

[0084] In summary, the embodiment of the present application is end-to-end for the detection of blast hole residual marks and the calculation of half-hole rate of the contour blasting excavation of the slope and underground chamber, has the advantages of automation and intelligence, and the obtained result is more objective, reducing the dependence on professionals and equipment. The visual result of the blast hole residual marks based on semantic segmentation can facilitate the detection personnel to intuitively observe the detail situation of the blast hole residual marks. The calculated half-hole rate further reflects the overbreak and underbreak and unevenness, and provides certain reference suggestions for the optimization of the next blasting design. A new idea is provided for the intelligent perception of blast hole residual marks, and a foundation is laid for the promotion of the method based on deep learning.

[0085] According to the blasting effect analysis method based on blast hole residual marks provided in the embodiment of the present application, the semantic segmentation model constructed in advance can be used to perform semantic segmentation on the blast hole residual mark picture to obtain a segmentation effect picture, and then the half-hole rate of the blast hole corresponding to the segmentation effect picture is calculated to determine the blasting effect of the current blasting scheme based on the statistical result of the half-hole rate. The blast hole residual mark can be quickly identified and the half-hole rate can be calculated to further evaluate the blasting effect and optimize the subsequent blasting parameters, which is beneficial to dynamically adjust the blasting design and construction and reduce the construction period and cost. Thus, the technical problem that the accuracy of field measurement is poor in the related art, the method of three-dimensional laser scanning combined with clustering needs to adjust parameters again when facing different scenes, the steps are complicated, and the efficiency is low, and the subsequent blasting analysis is affected is solved.

[0086] Secondly, the blasting effect analysis device based on blast hole residual marks according to the embodiment of the present application is described with reference to the accompanying drawings.

[0087] Figure 7 is a block schematic diagram of the blasting effect analysis device based on blast hole residual marks of the embodiment of the present application.

[0088] As shown in Figure 7 , the blasting effect analysis device 10 based on blast hole residual marks includes a first acquisition module 100, a segmentation module 200, and an analysis module 300.

[0089] Specifically, the first obtaining module 100 is configured to obtain a blast hole residual mark picture of each of the plurality of blast holes to be detected after a single blasting of the target area under the current blasting scheme.

[0090] The segmentation module 200 is configured to input the blast hole residual mark picture to be detected into a pre-constructed semantic segmentation model to obtain a segmentation effect picture.

[0091] The analysis module 300 is configured to calculate a half-hole rate of the blast hole corresponding to the segmentation effect picture, and determine the blasting effect of the current blasting scheme based on a statistical result of the half-hole rate.

[0092] Optionally, in an embodiment of the present application, the blasting effect analysis device 10 based on the blast hole residual mark further comprises a second obtaining module, a construction module, a training module and a testing module.

[0093] The second obtaining module is configured to obtain a plurality of blast hole images of a sample area, and label the plurality of blast hole images respectively to obtain real data of each blast hole image.

[0094] The construction module is configured to construct a blast hole residual mark data set by using the plurality of blast hole images and the real data, and divide the blast hole residual mark data set into a training set, a verification set and a testing set.

[0095] The training module is configured to train a plurality of initial semantic segmentation models by using the training set to obtain a plurality of trained semantic segmentation models, and verify the plurality of trained semantic segmentation models by using the verification set to obtain a plurality of actual semantic segmentation models.

[0096] The testing module is configured to test segmentation performance of the plurality of actual semantic segmentation models by using the testing set, and select a semantic segmentation model satisfying a preset effective condition from the plurality of actual semantic segmentation models based on the segmentation performance.

[0097] Optionally, in an embodiment of the present application, the testing module comprises an obtaining unit and a calculating unit.

[0098] The obtaining unit is configured to obtain testing data output by each actual semantic segmentation model.

[0099] The calculating unit is configured to calculate a quantitative index of each semantic segmentation model based on the testing data and corresponding real data, and determine the segmentation performance based on the quantitative index.

[0100] Optionally, in an embodiment of the present application, a calculation expression of the quantitative index is as follows:

[0101]

[0102]

[0103] wherein, IoU represents the intersection over union, P represents any region in the test data, G represents a region corresponding to any region in the real data, Precision represents the precision, TP represents the true positive, FP represents the false positive, FN represents the false negative, Recall represents the recall, and F1 represents the harmonic mean of the precision and the recall.

[0104] Optionally, in an embodiment of the present application, the calculation expression of the half-hole rate is:

[0105]

[0106] wherein, H a represents the half-hole rate, N represents the actual number of drill holes of the target region, L i / L represents the half-hole rate of a single blast hole.

[0107] It should be noted that the foregoing explanation and description of the embodiment of the blasting effect analysis method based on blast hole residual marks also applies to the embodiment of the blasting effect analysis device based on blast hole residual marks, which will not be described here.

[0108] The blasting effect analysis device based on blast hole residual marks according to the embodiment of the present application can perform semantic segmentation on the blast hole residual mark picture by using the pre-constructed semantic segmentation model to obtain a segmentation effect diagram, and further calculate the half-hole rate of the blast hole corresponding to the segmentation effect diagram, so as to determine the blasting effect of the current blasting scheme based on the statistical result of the half-hole rate. The blasting effect can be quickly identified and the half-hole rate can be calculated, and the blasting effect can be further evaluated, and the subsequent blasting parameters can be optimized, which is beneficial to dynamically adjust the blasting design and construction, and reduces the construction period and cost. Therefore, the technical problems in the related art that the accuracy of the on-site measurement is poor, and the method of three-dimensional laser scanning combined with clustering needs to adjust parameters again when facing different scenes, the steps are complicated, and the efficiency is low, which affects the subsequent blasting analysis are solved.

[0109] Figure 4 The structure schematic diagram of the electronic device provided by the embodiment of the present application is shown. The electronic device can include:

[0110] The memory 401, the processor 402, and the computer program stored in the memory 401 and executable on the processor 402.

[0111] The processor 402 implements the blasting effect analysis method based on blast hole residual marks provided in the above embodiments when executing the program.

[0112] Further, the electronic device further includes:

[0113] The communication interface 403 is used for communication between the memory 401 and the processor 402.

[0114] a memory 401 for storing a computer program executable in the processor 402.

[0115] The memory 401 can include a high-speed RAM memory and can also include a non-volatile memory, for example at least one disk memory.

[0116] If the memory 401, the processor 402 and the communication interface 403 are implemented independently, the communication interface 403, the memory 401 and the processor 402 can be connected to each other through a bus and complete the communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 4 In the figure, only one thick line is used to represent, but it does not mean that there is only one bus or one type of bus.

[0117] Optionally, in a specific implementation, if the memory 401, the processor 402 and the communication interface 403 are integrated on a chip, the memory 401, the processor 402 and the communication interface 403 can complete the communication between each other through an internal interface.

[0118] The processor 402 can be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0119] The embodiment further provides a computer readable storage medium, having stored thereon a computer program, which, when executed by a processor, implements the above blasting effect analysis method based on blasting hole residual marks.

[0120] The embodiment of the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the blasting effect analysis method based on blasting hole residual marks provided by the embodiment of the present application.

[0121] In the description of the application, reference to "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that a particular feature, structure, material, or characteristic being described is included in at least one embodiment or example of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment or example. Furthermore, the described specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. In addition, the usage of "N" means at least two, for example, two, three or the like, unless explicitly stated otherwise.

[0122] Furthermore, the terms "first", "second", or the like, are used merely as a designation of certain elements or features, and do not imply or connote relative importance or a specific order of categorization of the indicated features. Accordingly, features described as "first" or "second" can be explicitly or implicitly included in at least one of the features. In the description of the application, the meaning of "N" is at least two, for example, two, three, etc., unless explicitly specified otherwise.

[0123] Any process or method descriptions or blocks in flow charts or otherwise described herein represent embodiments which can be managed as one or more modules, segments, or portions of code which include one or more executable instructions for implementing specific logic functions or steps, and alternate implementations are possible. In some embodiments, the processes and methods described can be executed by one or more apparatuses or devices, either directly or after conversion to another language. Alternate implementations are possible.

[0124] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered as a sequence of executable instructions stored in a computer readable medium, which can be executed by an instruction execution system, apparatus or device, such as a computer-based system, a processor-based system, or other system that can fetch the instructions from the instruction execution system, apparatus or device and execute the instructions, or a combination of them. For the purposes of this specification, a "computer readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus or device. The computer readable medium can be a computer readable storage medium or a computer readable signal medium. The computer readable storage medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a propagation medium. The computer readable signal medium can include, but is not limited to, a computer readable medium that facilitates transfer of the program from one place to another. A specific example of a computer readable medium is a non-transitory computer-readable storage medium. A specific example of a computer readable signal medium is a source or destination of the computer readable medium. Another specific example of a computer readable signal medium is a computer readable signal travelling through space. Thus, a computer readable medium can take many forms of hardware to carry out the program for use by or in connection with the instruction execution system, apparatus or device.

[0125] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As such, if implemented in hardware, and in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), and the like.

[0126] Those of skill in the art would understand that the steps carried out by the above-mentioned embodiments of the method can be implemented by programs instructing the relevant hardware, and the programs can be stored in a computer readable storage medium. When the programs are executed, they include one or a combination of the steps of the method embodiments.

[0127] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing module, or each of the units can be physically present separately, or two or more units can be integrated in one module. The integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0128] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A blasting effect analysis method based on a blast hole residual mark, characterized by, The method comprises the following steps: obtaining a blast hole residual mark picture of each of a plurality of blast holes to be detected after a single blasting of a target area under a current blasting scheme; inputting the blast hole residual mark picture to be detected into a pre-constructed semantic segmentation model to obtain a segmentation effect picture; calculating a half-hole rate of the blast hole corresponding to the segmentation effect picture to determine a blasting effect of the current blasting scheme based on a statistical result of the half-hole rate.

2. The method of claim 1, wherein, Before the blast hole residual mark picture to be detected is inputted into the pre-constructed semantic segmentation model, the method further comprises the following steps: obtaining a plurality of blast mark images photographed by a sample area and labeling the plurality of blast mark images respectively to obtain true data of each blast mark image; constructing a blast hole residual mark data set by using the plurality of blast mark images and the true data, and dividing the blast hole residual mark data set into a training set, a verification set and a test set; training a plurality of initial semantic segmentation models by using the training set to obtain a plurality of trained semantic segmentation models, and verifying the plurality of trained semantic segmentation models by using the verification set to obtain a plurality of actual semantic segmentation models; testing segmentation performance of the plurality of actual semantic segmentation models by using the test set, and screening a semantic segmentation model meeting a preset effective condition from the plurality of actual semantic segmentation models by using the segmentation performance.

3. The method of claim 2, wherein, The step of testing the segmentation performance of the plurality of actual semantic segmentation models by using the test set comprises the following steps: obtaining test data outputted by each actual semantic segmentation model; calculating a quantitative index of each semantic segmentation model based on the test data and corresponding true data to determine the segmentation performance based on the quantitative index.

4. The method of claim 3, wherein, The calculation expression of the quantitative index is as follows: wherein, IoU represents an intersection over union, P represents any region in the test data, G represents a region corresponding to the any region in the true data, Precision represents an accuracy rate, TP represents a true positive, FP represents a false positive, FN represents a false negative, Recall represents a recall rate, and F1 represents a harmonic mean of the accuracy rate and the recall rate.

5. The method of claim 1, wherein, The calculation expression of the half-hole rate is as follows: where H a represents the half-hole rate, N represents the actual number of drilled holes of the target area, L i / L represents the half-hole rate of a single blast hole.

6. A device for analyzing a blasting effect based on a blast hole residual mark, characterized by, The method comprises the following steps: a first obtaining module, configured to obtain a blast hole residual mark picture of each of a plurality of blast holes to be detected after a single blasting of a target area under a current blasting scheme; a segmentation module, configured to input the blast hole residual mark picture to be detected into a pre-constructed semantic segmentation model to obtain a segmentation effect picture; an analysis module, configured to calculate a half-hole rate of the blast hole corresponding to the segmentation effect picture to determine a blasting effect of the current blasting scheme based on a statistical result of the half-hole rate.

7. The apparatus of claim 6, wherein, The method further comprises the following steps: a second obtaining module, configured to obtain a plurality of blast mark images photographed by a sample area and label the plurality of blast mark images respectively to obtain true data of each blast mark image; a construction module, configured to construct a blast hole residual mark data set by using the plurality of blast mark images and the true data, and divide the blast hole residual mark data set into a training set, a verification set and a test set; The training module is configured to train a plurality of initial semantic segmentation models by using the training set to obtain a plurality of trained semantic segmentation models, and to verify the plurality of trained semantic segmentation models by using the verification set to obtain a plurality of actual semantic segmentation models. The testing module is configured to test segmentation performance of the plurality of actual semantic segmentation models by using the test set, and to screen a semantic segmentation model satisfying a preset effective condition from the plurality of actual semantic segmentation models by using the segmentation performance.

8. An electronic device, comprising: Comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the blasting effect analysis method based on blasting hole residual marks according to any one of claims 1-5.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the blasting effect analysis method based on blasting hole residual marks according to any one of claims 1-5.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed to implement the blasting effect analysis method based on blasting hole residual marks according to any one of claims 1-5.

Images