Machine learning based automatic blind shot identification method and device for open pit mine

Abstract

The present disclosure relates to the technical field of computer technology and intelligent mining of coal mines, and particularly relates to a method and device for automatic identification of blind blasting in open-pit mines based on machine learning. The method comprises: acquiring video data transmitted by a UAV, wherein the UAV is provided with a camera device, and the video data is obtained by data collection of the UAV on the blasting process of a target blasting area of the open-pit mine; acquiring a key frame set corresponding to the video data; acquiring charging parameters corresponding to the blasting process of the target blasting area; acquiring image features of the key frame set and parameter features of the charging parameters; identifying the image features and the parameter features by using an automatic identification model of blind blasting in open-pit mines established based on a target machine learning algorithm, and acquiring an identification result of unblasted blast holes corresponding to the target blasting area. The present disclosure can reduce the duration of blind blasting detection and improve the accuracy of blind blasting detection, thereby improving the work efficiency of blind blasting detection.

Description

Technical Field

[0001] This disclosure relates to the fields of computer technology and intelligent coal mining technology, and in particular to a method and device for automatic identification of blind shots in open-pit mines based on machine learning. Background Technology

[0002] With the development of science and technology, blasting is a crucial and highly dangerous process in open-pit mining. Misfires can occur due to various reasons, making their detection a key concern. For example, during blasting, hundreds or even thousands of blast holes can detonate in a very short time, making misfires difficult to detect. Once detonated, misfires can cause casualties and increase safety hazards. For instance, misfire detection in open-pit mines requires manual entry into the blasting area after blasting is complete and the fumes have dissipated, resulting in low efficiency and high risk. Summary of the Invention

[0003] This disclosure provides a machine learning-based automatic identification method and apparatus for misfire detection in open-pit mines, which can reduce the detection time and improve the accuracy of misfire detection, thereby increasing the efficiency of misfire detection. The technical solution of this disclosure is as follows:

[0004] According to a first aspect of the present disclosure, a method for automatic identification of blind shots in open-pit mines based on machine learning is provided, the method comprising:

[0005] Acquire video data transmitted by a drone, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area in the open-pit mine;

[0006] Obtain the set of keyframes corresponding to the video data;

[0007] Obtain the charge parameters corresponding to the blasting process in the target blasting area;

[0008] Obtain the image features of the keyframe set and the parameter features of the charge parameters;

[0009] An automatic identification model for blind blasting in open-pit mines, based on a target machine learning algorithm, is used to identify the image features and parameter features, thereby obtaining the identification results of the unexploded blast holes corresponding to the target blasting area.

[0010] According to some embodiments, the automatic identification model for open-pit mine blind blasting established based on a target machine learning algorithm is used to identify the image features and the parameter features, and to obtain the identification results of the unexploded blast holes corresponding to the target blasting area, including:

[0011] The keyframe set is processed using the gray-level co-occurrence matrix algorithm to obtain the texture features of each blast hole corresponding to the keyframe set;

[0012] An edge detection algorithm is used to process the keyframe set to obtain the edge features of each blast hole corresponding to the keyframe set;

[0013] Based on the charge quantity in the charge parameters and the rock mass blastability coefficient corresponding to the target blasting area, the parameter characteristics of the charge parameters are obtained;

[0014] The texture features, edge features, and parameter features of each blast hole are input into the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm for identification, and the identification results of the unblasted blast holes corresponding to the target blasting area are obtained.

[0015] According to some embodiments, the method further includes:

[0016] Based on the charge quantity in the charge parameters, the rock mass blastability coefficient corresponding to the target blasting area, the edge gradient intensity, and the texture energy value, the first weight corresponding to the texture features of each blast hole, the second weight corresponding to the edge features of each blast hole, and the third weight corresponding to the parameter features of the charge parameters are obtained in the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm.

[0017] According to some embodiments, obtaining the charge parameters corresponding to the blasting process of the target blasting area includes:

[0018] Based on the blasting design information and the rock strata structure corresponding to the target blasting area, the charge parameters corresponding to the blasting process of the target blasting area are obtained.

[0019] According to some embodiments, obtaining the set of keyframes corresponding to the video data includes:

[0020] Obtain the blasting information corresponding to the target blasting area;

[0021] Based on the explosion information, keyframes are extracted from the video data to obtain a keyframe set.

[0022] According to some embodiments, each keyframe in the keyframe set corresponds to a gun hole, and the method further includes:

[0023] Based on the borehole number information corresponding to each keyframe in the keyframe set and the identification result of the unexploded borehole, the unexploded borehole is processed as a blind shot.

[0024] According to some embodiments, the method further includes:

[0025] Obtain the set of historical keyframes corresponding to the historical blasting process in the open-pit mine;

[0026] The borehole regions of each historical keyframe in the historical keyframe set are cropped to obtain the borehole images corresponding to each historical borehole, and the annotation information corresponding to the borehole images corresponding to each historical borehole is obtained.

[0027] The gray-level co-occurrence matrix algorithm is used to process the images of the boreholes corresponding to each historical borehole to obtain the historical texture features of each historical borehole.

[0028] An edge detection algorithm is used to process the images of the blast holes corresponding to each historical blast hole to obtain the historical edge features of each historical blast hole;

[0029] Obtain the historical charge parameters corresponding to the historical blasting process;

[0030] Based on the charge quantity in the historical charge parameters and the rock mass blastability coefficient corresponding to the historical blasting process, the characteristics of the historical charge parameters are obtained.

[0031] The initial machine learning algorithm is trained based on the texture features of each historical blast hole, the edge features of each historical blast hole, the features of the historical charge parameters, and the annotation information to establish the automatic identification model for open-pit mine blind blasts based on the target machine learning algorithm.

[0032] According to a second aspect of the present disclosure, a machine learning-based automatic identification device for blind shots in open-pit mines is provided, comprising:

[0033] A data acquisition unit is used to acquire video data transmitted by a drone, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area of ​​the open-pit mine.

[0034] A set acquisition unit is used to acquire the set of keyframes corresponding to the video data;

[0035] The parameter acquisition unit is used to acquire the charge parameters corresponding to the blasting process of the target blasting area;

[0036] The parameter acquisition unit is also used to acquire the image features of the keyframe set and the parameter features of the charge parameters;

[0037] The blast hole identification unit is used to identify the image features and parameter features by using an open-pit mine blind blasting automatic identification model established based on a target machine learning algorithm, and to obtain the identification results of the unblasted blast holes corresponding to the target blasting area.

[0038] According to a third aspect of the present disclosure, an electronic device is provided, comprising:

[0039] processor;

[0040] Memory used to store the processor's executable instructions;

[0041] The processor is configured to execute the instructions to implement the machine learning-based automatic identification method for blind shots in open-pit mines, as described in any one of the preceding aspects.

[0042] According to a fourth aspect of the present disclosure, a storage medium is provided that, when instructions in the storage medium are executed by a processor of an electronic device, enables the electronic device to perform the machine learning-based automatic identification method for blind shots in open-pit mines as described in any of the preceding aspects.

[0043] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described in any one of the preceding aspects.

[0044] The technical solutions provided by the embodiments of this disclosure have at least the following beneficial effects:

[0045] In some or related embodiments, video data transmitted by a drone is acquired. The drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of a target blasting area in an open-pit mine. A set of keyframes corresponding to the video data is acquired; the charge parameters corresponding to the blasting process in the target blasting area are acquired; the image features of the keyframe set and the parameter features of the charge parameters are acquired; and an automatic open-pit mine misfire identification model based on a target machine learning algorithm is used to identify the image features and parameter features, thereby obtaining the identification results of unblasted blast holes corresponding to the target blasting area. Therefore, video data of the blasting process in the target blasting area can be acquired by a drone, and identification can be performed based on the keyframe set and charge parameters to obtain the identification results of unblasted blast holes. This eliminates the need for manual entry into the blasting area (target blasting area) for inspection, reducing the inefficiency and high risk of manual inspection. Real-time detection and positioning of misfires in the blasting area can be performed at the moment of blasting, reducing the detection time and improving the accuracy of misfire detection, thus improving the efficiency of misfire detection and enhancing personnel safety.

[0046] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0047] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure, and are not intended to unduly limit this disclosure.

[0048] Figure 1 This is a flowchart of the first machine learning-based automatic identification method for blind shots in open-pit mines provided in this disclosure embodiment;

[0049] Figure 2 This is a flowchart of the second machine learning-based automatic identification method for blind shots in open-pit mines provided in this disclosure embodiment;

[0050] Figure 3a This is an example schematic diagram of an image frame acquired by a drone according to an embodiment of this disclosure;

[0051] Figure 3b This is an example schematic diagram of an image frame acquired by a drone according to an embodiment of this disclosure;

[0052] Figure 4 This is a block diagram illustrating an automatic identification device for blind shots in open-pit mines based on machine learning, according to an exemplary embodiment.

[0053] Figure 5 This is a block diagram illustrating an electronic device according to an exemplary embodiment. Detailed Implementation

[0054] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings.

[0055] This disclosure presents a method and apparatus for automatic identification of misfires in open-pit mines based on machine learning. In some embodiments, the terms "automatic identification method for misfires in open-pit mines based on machine learning" can be substituted for terms such as "information processing method" and "communication method," and the terms "automatic identification apparatus for misfires in open-pit mines based on machine learning" can be substituted for terms such as "information processing apparatus" and "communication apparatus," and the terms "information processing system" and "communication system" can be substituted for terms such as "information processing system."

[0056] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0057] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0058] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0059] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0060] In the embodiments disclosed herein, "multiple" refers to two or more.

[0061] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0062] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0063] In some embodiments, "terminal" or "terminal device" may be referred to as "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," etc.

[0064] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0065] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0066] Figure 1 This is a flowchart of the first machine learning-based automatic identification method for blind shots in open-pit mines provided in this disclosure, as shown in the embodiments. Figure 1 As shown, this machine learning-based automatic identification method for misfires in open-pit mines can be used to detect misfires after blasting in open-pit mines, and includes the following steps:

[0067] In step S11, video data transmitted by the UAV is acquired. The UAV is equipped with a camera device, and the video data is obtained by the UAV collecting data on the blasting process of the target blasting area of ​​the open-pit mine.

[0068] In some embodiments, the executing entity of this disclosure may be, for example, an electronic device. This electronic device does not specifically refer to a particular fixed electronic device. For example, when the device identifier changes, the electronic device may also change accordingly. For example, when the structure of the electronic device changes, the electronic device may also change accordingly. Furthermore, the executing entity of this disclosure may also be a server, which may be a single server or a server cluster; this disclosure does not limit this.

[0069] In some embodiments, the electronic device disclosed herein may be, for example, a device that communicates with a drone and a signal acquisition device. The communication method is not limited thereto.

[0070] According to some embodiments, a drone is, for example, an aircraft that can be controlled remotely or autonomously without a human pilot. This drone does not specifically refer to a particular fixed drone. For example, the drone may change as its components change. For example, the drone's identifier may change as its identification changes.

[0071] According to some embodiments, the drones in the embodiments of this disclosure can be a collection of drones or a single drone. The embodiments of this disclosure do not limit this. The number of drones can be determined, for example, based on the area of ​​the target blasting zone.

[0072] The drone is equipped with a camera device, but there are no limitations on the type of camera device. For example, the camera device can be a webcam. The frame rate of the webcam can be, for example, a webcam with a frame rate greater than a preset frame rate threshold.

[0073] In some embodiments, an open-pit mine may refer, for example, to a mine that extracts mineral resources through open-pit mining. This open-pit mine does not specifically refer to a particular fixed open-pit mine. For example, when the type of mineral corresponding to the open-pit mine changes, the open-pit mine may also change accordingly. For example, when the area corresponding to the open-pit mine changes, the open-pit mine may also change accordingly.

[0074] In some embodiments, the target blasting area may be, for example, the area where the blast hole was detonated. This target blasting area may be, for example, a portion of the first open-pit mine, or the entire area; this disclosure does not limit the scope of the embodiments.

[0075] According to some embodiments, the video data is data collected by a drone during the blasting process of the target blasting area in the first open-pit mine. This video data is not specifically defined as any fixed data point. For example, the video data may change if the target blasting area changes. Similarly, the video data may change if the video duration changes, or if the acquisition time point changes.

[0076] According to some embodiments, video data transmitted by a drone is acquired, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area in the open-pit mine.

[0077] In step S12, the set of keyframes corresponding to the video data is obtained;

[0078] In some embodiments, keyframes may include key information frames. In this disclosure, keyframes may be key image frames before and after each blasting time point. The keyframe set may be a collection of at least one keyframe. This keyframe set does not specifically refer to a fixed set. For example, the keyframe set may change when the method of acquiring it changes. For example, the keyframe set may also change when the number of keyframes corresponding to it changes.

[0079] According to some embodiments, the method for obtaining the keyframe set is not limited. For example, the keyframe set can be obtained through a keyframe recognition model and explosion data, or it can be directly extracted from video data based on explosion information.

[0080] According to some embodiments, a set of keyframes corresponding to video data is obtained.

[0081] In step S13, the charge parameters corresponding to the blasting process of the target blasting area are obtained;

[0082] According to some embodiments, the charge parameters may be, for example, parameters related to the blasting materials added to each borehole before blasting in the target blasting area. These charge parameters may include, for example, the charge amount and the rock mass explosiveness factor. These charge parameters are not specifically defined as a single fixed parameter. For example, when the type of parameter corresponding to a charge parameter changes, the charge parameter may also change accordingly. For example, when the value of the parameter corresponding to a charge parameter changes, the charge parameter may also change accordingly.

[0083] According to some embodiments, the method for obtaining the charge parameters is not limited. For example, it can be determined based on the regional information of the target blasting area. This regional information may include, for example, the rock type and the rock hardness.

[0084] According to some embodiments, the charge parameters corresponding to the blasting process of the target blasting area can be obtained.

[0085] In step S14, the image features of the keyframe set and the parameter features of the charge parameters are obtained;

[0086] In some embodiments, the image features may be features obtained by feature extraction based on a set of keyframes. For example, these may be image features obtained by feature extraction from each keyframe in the keyframe set.

[0087] In some embodiments, the parameter feature may be a feature corresponding to the charge parameters, and the parameter feature does not specifically refer to a certain fixed feature. For example, when the charge parameters change, the parameter feature may also change accordingly.

[0088] According to some embodiments, image features of keyframe sets and parametric features of charge parameters can be obtained.

[0089] In step S15, an automatic identification model for open-pit mine blind blasting based on a target machine learning algorithm is used to identify image features and parameter features, and to obtain the identification results of unblasted blast holes corresponding to the target blasting area.

[0090] According to some embodiments, the target machine learning algorithm may be, for example, a trained algorithm that can be used for unexploded ordnance hole identification. This target machine learning algorithm is not specifically defined by any particular algorithm. For example, the target machine learning algorithm may change accordingly when the algorithm type corresponding to it changes. Similarly, the target machine learning algorithm may change accordingly when the parameters within it change.

[0091] According to some embodiments, the automatic identification model for misfires in open-pit mines, established based on a target machine learning algorithm, can be any automatic identification model for misfires in open-pit mines established based on a target machine learning algorithm. This automatic identification model for misfires in open-pit mines does not specifically refer to a particular fixed model.

[0092] In some embodiments, the identification results of unexploded ordnance holes can be used, for example, to indicate the unexploded ordnance hole information corresponding to the target blasting area. The identification result of unexploded ordnance holes is not specifically defined by a fixed result. For example, when the number of unexploded ordnance holes changes, the identification result of unexploded ordnance holes can also change accordingly. For example, when the target blasting area changes, the identification result of unexploded ordnance holes can also change accordingly. Unexploded ordnance holes can also be referred to as misfires.

[0093] In some embodiments, an automatic identification model for open-pit mine blind blasting based on a target machine learning algorithm can be used to identify image features and parameter features, and obtain the identification results of unblasted blast holes corresponding to the target blasting area.

[0094] In some or related embodiments, video data transmitted by a drone is acquired. The drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area in an open-pit mine. The process involves acquiring a set of keyframes corresponding to the video data; acquiring the charge parameters corresponding to the blasting process in the target blasting area; acquiring the image features of the keyframe set and the parameter features of the charge parameters; and using an open-pit mine misfire automatic identification model based on a target machine learning algorithm to identify the image features and parameter features, thereby obtaining the identification results of unblasted blast holes corresponding to the target blasting area. Therefore, by acquiring video data of the blasting process in the target blasting area using a drone, identification can be performed based on the keyframe set and charge parameters to obtain the identification results of unblasted blast holes. This eliminates the need for manual entry into the blasting area (target blasting area) for inspection, reducing the inefficiency and high risk of manual inspection. It allows for real-time detection and positioning of misfires in the blasting area at the moment of blasting, reducing the detection time and improving the accuracy of misfire detection, thus improving the efficiency of misfire detection and enhancing personnel safety.

[0095] Figure 2 This is a flowchart of the second machine learning-based automatic identification method for blind shots in open-pit mines provided in this disclosure, as shown in the embodiments. Figure 2 As shown, it includes the following steps:

[0096] In step S21, video data transmitted by the UAV is acquired. The UAV is equipped with a camera device, and the video data is obtained by the UAV collecting data on the blasting process of the target blasting area of ​​the open-pit mine.

[0097] The relevant descriptions can be as described above, and will not be repeated here.

[0098] According to some embodiments, the blasting process of the target blasting area in an open-pit mine can be, for example, a process that starts from the first blast hole and continues until no blast hole is detonated within a preset time.

[0099] According to some embodiments, the drone can hover at a safe distance above the center of the blasting face, i.e., the center of the target blasting area. The drone's hovering position ensures that all blast holes have a resolution greater than a resolution threshold in the image and are located in the central region, acquiring video data of the entire blasting process. The hovering position and number of drones can be determined, for example, based on the area of ​​the target blasting area and the resolution of the blast holes in the image. For example, the camera device can be a webcam. The frame rate of this webcam can be, for example, a camera with a frame rate greater than a preset frame rate threshold. This preset frame rate can be, for example, 60 FPS. The image frames acquired by the drone can be, for example, as shown in the image... Figure 3a and 3b As shown.

[0100] In step S22, the set of keyframes corresponding to the video data is obtained;

[0101] The relevant descriptions can be as described above, and will not be repeated here.

[0102] According to some embodiments, when video data is acquired, it can be preprocessed. This image preprocessing may include grayscale conversion, contrast enhancement, filtering and noise reduction, and data augmentation.

[0103] According to some embodiments, the keyframe set corresponding to the video data is obtained, including:

[0104] Obtain the blasting information corresponding to the target blasting area;

[0105] Based on the explosion information, keyframes are extracted from the video data to obtain a set of keyframes.

[0106] Based on some examples, drones can use cameras with a frame rate greater than 60 FPS to shoot. The frame rate is high enough to ensure that the fast-delayed blasting process can be captured. Among the many adjacent frames captured, the key frames are the ones that can distinguish between the detonated and undetonated blast holes.

[0107] According to some embodiments, for example, open-pit mines can adopt inter-row delayed blasting, where boreholes in the same row are detonated simultaneously, and boreholes in adjacent rows are detonated sequentially at the same time interval. Open-pit mines can also use grouped delayed blasting, where all boreholes are divided into several groups as needed, and each group is detonated at the same time interval. Given a fixed blasting method in an open-pit mine, a row / group of detonated boreholes will exist in a given image frame. For example, in the same image frame, corresponding rows / groups should be detonated simultaneously or in the same state; if there are different boreholes, then that borehole may be a misfire.

[0108] According to some embodiments, the UAV has edge computing capabilities. The UAV's onboard computer needs to have a pre-installed Sobel operator FPGA module for real-time calculation of G value, a dedicated computing chip (such as Movidius) for real-time calculation of texture features, and a wireless transmission module for acquiring charge parameters.

[0109] In step S23, the charge parameters corresponding to the blasting process of the target blasting area are obtained;

[0110] The relevant descriptions can be as described above, and will not be repeated here.

[0111] According to some implementation examples, since each open-pit mine and each blasting area has a different structure and blasting requirements, each blasting is professionally designed according to its rock strata structure and blasting requirements, including the location of the blast holes, the amount of explosives, the hole depth, etc.

[0112] According to some embodiments, the charge parameters corresponding to the blasting process in an open-pit mine are obtained, including:

[0113] Based on the blasting requirements and rock strata structure of the target blasting area, the charge parameters for the blasting process in the target blasting area are obtained.

[0114] The blasting requirements information may include, for example, the blasting range and blasting depth.

[0115] In step S24, the gray-level co-occurrence matrix algorithm is used to process the keyframe set to obtain the texture features of each blast hole corresponding to the keyframe set;

[0116] The relevant descriptions can be as described above, and will not be repeated here.

[0117] According to some embodiments, the Gray-Level Co-occurrence Matrix (GLCM) algorithm is an algorithm for analyzing image texture features. It can be used in image processing and computer vision fields to describe the texture characteristics of an image by statistically analyzing the spatial dependencies of pixel gray values.

[0118] In step S25, an edge detection algorithm is used to process the keyframe set to obtain the edge features of each borehole corresponding to the keyframe set.

[0119] The relevant descriptions can be as described above, and will not be repeated here.

[0120] According to some embodiments, the edge detection algorithm is not specifically defined by a single fixed algorithm. For example, the edge detection algorithm may include Canny, Sobel, etc. For example, when the algorithm type corresponding to the edge detection algorithm changes, the edge detection algorithm may also change accordingly.

[0121] In step S26, the parameter characteristics of the charge parameters are obtained based on the charge amount and the rock mass blastability coefficient corresponding to the target blasting area in the charge parameters.

[0122] The relevant descriptions can be as described above, and will not be repeated here.

[0123] In step S27, the texture features, edge features, and charge parameters of each blast hole are input into the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm for identification, and the identification results of the unblasted blast holes corresponding to the target blasting area are obtained.

[0124] The relevant descriptions can be as described above, and will not be repeated here.

[0125] According to some embodiments, after obtaining the borehole image, i.e., the keyframe image, through data processing, two types of image features can be extracted from the borehole image: texture features and edge features. For example, the image frame can be input into a texture feature extraction algorithm and an edge detection algorithm, and the output is the corresponding feature value.

[0126] According to some embodiments, G can represent the edge gradient intensity, which is obtained by calculating the mean magnitude (px) of the Sobel operator, and the value range can be, for example, 20-80; E_t represents the texture energy value, which can be obtained through the energy feature of the gray-level co-occurrence matrix, and the value range can be, for example, 5-15.

[0127] Among them, the openings of the detonated blast holes are completely destroyed, with visible radial cracks and loose rock at the edges. The openings of the undetonated blast holes are intact, retaining blast mud or plugs. For example, the crater after the detonation of the detonated blast holes is a regular funnel shape with sharp edges, while the undetonated crater is either a non-blast crater or only a shallow depression.

[0128] According to some embodiments, the method further includes:

[0129] Based on the charge quantity, rock mass blastability coefficient, edge gradient intensity, and texture energy value in the charge parameters, the first weight corresponding to the texture features of each borehole, the second weight corresponding to the edge features of each borehole, and the third weight corresponding to the parameter features of the charge parameters are obtained in the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm.

[0130] The feature dynamic weighting formulas based on three features—charge parameters, edge, and texture—are shown in (1) to (3) below:

[0131]

[0132]

[0133] Where Q represents the charge amount (kg), k represents the rock mass blastability coefficient, and Q / k represents the charge amount-rock mass hardness ratio, which can be obtained through design parameters / rock tests. The charge amount-rock mass hardness ratio can range from 2 to 10. G represents the edge gradient intensity, obtained by calculating the average amplitude (px) of the Sobel operator, with a value range of 20 to 80. E_t represents the texture energy value, obtained through the energy features of the gray-level co-occurrence matrix, with a value range of 5 to 15. Therefore, based on the charge amount-rock mass hardness ratio Q / k, the edge gradient intensity G, and the texture energy value E_t, the weight coefficients w_morph, w_edge, and w_texture of the charge parameters, edge, and texture features can be calculated using the normalized square ratio formula. This allows for better adaptation to lithological variations and improves the robustness of the model.

[0134] According to some embodiments, under the hard rock large charge condition (Q / k>6), the morphology of the crater after detonation differs significantly from that of the undetonated blast hole. In this case, w_morph approaches 1, and the feature extraction result is dominated by the charge parameters. Under the soft rock small charge condition (Q / k<3), the texture features of the residue after blasting differ significantly between detonated and undetonated blast holes. In this case, w_texture approaches 1, and the feature extraction result is dominated by the texture features of the blast hole image. Under the critical state (the critical state can be, for example, when an electronic detonator or explosive is not completely detonated, which is the most dangerous in the production process), there are differences in the explosion traces around the blast hole after blasting. In this case, w_edge increases significantly, and the feature extraction result is dominated by the edge gradient of the blast hole image.

[0135] According to some implementations, when w_edge>0.6 and w_morph<0.2, the model directly identifies the borehole corresponding to the keyframe as an undetonated borehole.

[0136] According to some embodiments, where each keyframe in the keyframe set corresponds to a gun hole, the method further includes:

[0137] Based on the borehole number information corresponding to each keyframe in the keyframe set and the identification results of unexploded boreholes, misfire processing is performed on the unexploded boreholes. Therefore, the accuracy of unexploded borehole acquisition can be improved, the accuracy of subsequent misfire processing can be improved, and the inaccuracy and complexity of manual inspection can be reduced. This can improve both the accuracy and efficiency of misfire identification.

[0138] In some embodiments, keyframes can be labeled, with detonated blast holes labeled as 1 and undetonated blast holes labeled as 2. This disclosure does not limit the labeling method.

[0139] According to some embodiments, the method further includes:

[0140] Obtain the set of historical keyframes corresponding to the historical blasting process in the open-pit mine;

[0141] The borehole regions of each historical keyframe in the historical keyframe set are cropped to obtain the borehole images corresponding to each historical borehole, and the annotation information corresponding to the borehole images corresponding to each historical borehole is obtained.

[0142] The gray-level co-occurrence matrix algorithm is used to process the images of the boreholes corresponding to each historical borehole to obtain the historical texture features of each historical borehole.

[0143] An edge detection algorithm is used to process the images of the blast holes corresponding to each historical blast hole to obtain the historical edge features of each historical blast hole;

[0144] Obtain historical charge parameters corresponding to historical blasting processes;

[0145] Based on the charge quantity and rock mass blastability coefficient corresponding to the historical blasting process in the historical charge parameters, the characteristics of the historical charge parameters are obtained.

[0146] Based on the texture features, edge features, charge parameters, and annotation information of each historical blast hole, an initial machine learning algorithm is trained to establish an automatic identification model for undetonated blast holes in open-pit mines. Therefore, the machine learning algorithm can be trained using data from historical blasting processes, improving its accuracy and thus enhancing the identification of undetonated blast holes.

[0147] The machine learning algorithm includes, but is not limited to, support vector machines, random forests, and neural networks.

[0148] According to some embodiments, video data of a new blasting process, i.e., a test set, is acquired. Keyframes of each group of blast holes are extracted and cropped according to the blasting design. Each cropped image contains one blast hole. The cropped images are numbered, for example, xy, indicating that the image represents the y-th blast hole in the x-th group, thus creating an open-pit mine blast hole blasting test set. After processing the blast hole images in the test set through grayscale conversion, contrast enhancement, filtering and denoising, and data augmentation, texture features, edge features, and charge parameter features of the blast hole images in the test set can be extracted. The extracted features are input into a trained open-pit mine blast hole classification model to achieve automatic identification and classification of detonated and undetonated blast hole images in the test set. Blades identified as undetonated in the test set are misfires that failed to detonate during the blasting process and can be treated as misfires based on their corresponding numbers.

[0149] In one or related embodiments, a gray-level co-occurrence matrix algorithm can be used to process the keyframe set to obtain the texture features of each blast hole corresponding to the keyframe set; an edge detection algorithm can be used to process the keyframe set to obtain the edge features of each blast hole corresponding to the keyframe set; based on the charge amount and the rock mass blastability coefficient corresponding to the target blasting area in the charge parameters, the parameter features of the charge parameters can be obtained; the texture features, edge features, and parameter features of each blast hole are input into an open-pit mine blind blasting automatic identification model established based on a target machine learning algorithm for identification, and the identification results of the unblasted blast holes corresponding to the target blasting area are obtained. Therefore, the corresponding features can be obtained through the algorithm corresponding to each feature, which can improve the accuracy of each feature acquisition and improve the accuracy of the identification results of the unblasted blast holes.

[0150] A block diagram of an automatic identification device for blind shots in open-pit mines based on machine learning is shown according to an exemplary embodiment. (Refer to...) Figure 4 The device 400 includes:

[0151] The data acquisition unit 401 is used to acquire video data transmitted by the UAV. The UAV is equipped with a camera device, and the video data is obtained by the UAV collecting data on the blasting process of the target blasting area of ​​the open-pit mine.

[0152] The set acquisition unit 402 is used to acquire the set of keyframes corresponding to the video data;

[0153] The parameter acquisition unit 403 is used to acquire the charge parameters corresponding to the blasting process of the target blasting area;

[0154] The parameter acquisition unit 403 is also used to acquire the image features of the keyframe set and the parameter features of the charge parameters;

[0155] The blast hole identification unit 404 is used to identify image features and parameter features by adopting an open-pit mine blind blast automatic identification model based on target machine learning algorithm, and to obtain the identification results of unblasted blast holes corresponding to the target blasting area.

[0156] According to some embodiments, the borehole identification unit 404 is used to identify image features and parameter features using an open-pit mine blind blasting automatic identification model established based on a target machine learning algorithm, and to obtain the identification results of unblasted boreholes corresponding to the target blasting area. Specifically, it is used for:

[0157] The gray-level co-occurrence matrix algorithm is used to process the keyframe set to obtain the texture features of each gun hole corresponding to the keyframe set;

[0158] An edge detection algorithm is used to process the keyframe set to obtain the edge features of each borehole corresponding to the keyframe set;

[0159] Based on the charge quantity and the rock mass blastability coefficient corresponding to the target blasting area in the charge parameters, the parameter characteristics of the charge parameters are obtained;

[0160] The texture features, edge features, and charge parameters of each blast hole are input into an open-pit mine blind blasting automatic identification model based on a target machine learning algorithm for identification, and the identification results of the unblasted blast holes corresponding to the target blasting area are obtained.

[0161] According to some embodiments, the borehole identification unit 404 is also specifically used for:

[0162] Based on the charge quantity, rock mass blastability coefficient, edge gradient intensity, and texture energy value in the charge parameters, the first weight corresponding to the texture features of each borehole, the second weight corresponding to the edge features of each borehole, and the third weight corresponding to the parameter features of the charge parameters are obtained in the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm.

[0163] According to some embodiments, when the borehole identification unit 404 is used to acquire the charge parameters corresponding to the blasting process of the target blasting area, it is specifically used for:

[0164] Based on the blasting design information and rock strata structure corresponding to the target blasting area, the charge parameters corresponding to the blasting process of the target blasting area are obtained.

[0165] According to some embodiments, when the set acquisition unit acquires the set of keyframes corresponding to the video data, it is specifically used for:

[0166] Obtain the blasting information corresponding to the target blasting area;

[0167] Based on the explosion information, keyframes are extracted from the video data to obtain a set of keyframes.

[0168] According to some embodiments, where each keyframe in the keyframe set corresponds to a gun hole, the gun hole identification unit 404 is further specifically used for:

[0169] Based on the borehole number information corresponding to each keyframe in the keyframe set and the identification results of unexploded boreholes, blind firing processing is performed on the unexploded boreholes.

[0170] According to some embodiments, the borehole identification unit 404 is also specifically used for:

[0171] Obtain the set of historical keyframes corresponding to the historical blasting process in the open-pit mine;

[0172] The borehole regions of each historical keyframe in the historical keyframe set are cropped to obtain the borehole images corresponding to each historical borehole, and the annotation information corresponding to the borehole images corresponding to each historical borehole is obtained.

[0173] The gray-level co-occurrence matrix algorithm is used to process the images of the boreholes corresponding to each historical borehole to obtain the historical texture features of each historical borehole.

[0174] An edge detection algorithm is used to process the images of the blast holes corresponding to each historical blast hole to obtain the historical edge features of each historical blast hole;

[0175] Obtain historical charge parameters corresponding to historical blasting processes;

[0176] Based on the charge quantity and rock mass blastability coefficient corresponding to the historical blasting process in the historical charge parameters, the characteristics of the historical charge parameters are obtained.

[0177] The initial machine learning algorithm is trained based on the texture features, edge features, charge parameters, and annotation information of each historical blast hole to establish an automatic identification model for blind blasts in open-pit mines based on the target machine learning algorithm.

[0178] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0179] In some or related embodiments, a data acquisition unit is used to acquire video data transmitted by a drone, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area in the open-pit mine; a set acquisition unit is used to acquire the set of key frames corresponding to the video data; a parameter acquisition unit is used to acquire the charge parameters corresponding to the blasting process in the target blasting area; the parameter acquisition unit is also used to acquire the image features of the key frame set and the parameter features of the charge parameters; a borehole identification unit is used to identify the image features and parameter features using an open-pit mine blind blasting automatic identification model established based on a target machine learning algorithm, and to obtain the identification results of the unblasted boreholes corresponding to the target blasting area. Therefore, video data of the blasting process in the target blasting area can be acquired by a drone, and identification can be performed based on the set of key frames and charge parameters to obtain the identification results of the unblasted boreholes. This eliminates the need for manual entry into the blasting area (i.e., the target blasting area) for inspection, reducing the inefficiency and high risk of manual inspection. It allows for real-time detection and positioning of blind blasts in the blasting area at the moment of blasting, reducing the detection time and improving the accuracy of blind blasting detection, thereby improving the efficiency of blind blasting detection and enhancing personnel safety.

[0180] Figure 5 A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device 500 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0181] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0182] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, such as keyboard, mouse, etc.; output unit 507, such as various types of monitors, speakers, etc.; storage unit 508, such as disk, optical disk, etc.; and communication unit 509, such as network card, modem, wireless transceiver, etc. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0183] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the license plate recognition method. For example, in some embodiments, the license plate recognition method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the license plate recognition method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the license plate recognition method by any other suitable means (e.g., by means of firmware).

[0184] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0185] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0186] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0187] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0188] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), the Internet, and blockchain networks.

[0189] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service system that addresses the management difficulties and weak business scalability inherent in traditional physical hosts and VPS (Virtual Private Server) services. Servers can also be servers for distributed systems or servers integrated with blockchain technology.

[0190] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0191] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. An automatic identification method for blind shots in open-pit mines based on machine learning, characterized in that, include: Acquire video data transmitted by a drone, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area in the open-pit mine; Obtain the set of keyframes corresponding to the video data; Obtain the charge parameters corresponding to the blasting process in the target blasting area; Obtain the image features of the keyframe set and the parameter features of the charge parameters; An automatic open-pit mine blasting identification model based on a target machine learning algorithm is used to identify the image features and parameter features, obtaining the identification results of unexploded blast holes corresponding to the target blasting area. Specifically, a gray-level co-occurrence matrix algorithm is used to process the keyframe set to obtain the texture features of each blast hole corresponding to the keyframe set; an edge detection algorithm is used to process the keyframe set to obtain the edge features of each blast hole corresponding to the keyframe set; parameter features of the charge parameters are obtained based on the charge amount and the rock mass blastability coefficient corresponding to the target blasting area; and the texture features, edge features, and parameter features of each blast hole are input into the automatic open-pit mine blasting identification model based on the target machine learning algorithm for identification, obtaining the identification results of unexploded blast holes corresponding to the target blasting area.

2. The method according to claim 1, characterized in that, The method further includes: Based on the charge quantity in the charge parameters, the rock mass blastability coefficient corresponding to the target blasting area, the edge gradient intensity, and the texture energy value, the first weight corresponding to the texture features of each blast hole, the second weight corresponding to the edge features of each blast hole, and the third weight corresponding to the parameter features of the charge parameters are obtained in the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm.

3. The method according to any one of claims 1 to 2, characterized in that, The process of obtaining the charge parameters corresponding to the blasting process of the target blasting area includes: Based on the blasting design information and the rock strata structure corresponding to the target blasting area, the charge parameters corresponding to the blasting process of the target blasting area are obtained.

4. The method according to claim 1, characterized in that, The step of obtaining the set of keyframes corresponding to the video data includes: Obtain the blasting information corresponding to the target blasting area; Based on the explosion information, keyframes are extracted from the video data to obtain a keyframe set.

5. The method according to claim 1, characterized in that, in, Each keyframe in the keyframe set corresponds to a gun hole, and the method further includes: Based on the borehole number information corresponding to each keyframe in the keyframe set and the identification result of the unexploded borehole, the unexploded borehole is processed as a blind shot.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the set of historical keyframes corresponding to the historical blasting process in the open-pit mine; The borehole regions of each historical keyframe in the historical keyframe set are cropped to obtain the borehole images corresponding to each historical borehole, and the annotation information corresponding to the borehole images corresponding to each historical borehole is obtained. The gray-level co-occurrence matrix algorithm is used to process the images of the boreholes corresponding to each historical borehole to obtain the historical texture features of each historical borehole. An edge detection algorithm is used to process the images of the blast holes corresponding to each historical blast hole to obtain the historical edge features of each historical blast hole; Obtain the historical charge parameters corresponding to the historical blasting process; Based on the charge quantity in the historical charge parameters and the rock mass blastability coefficient corresponding to the historical blasting process, the characteristics of the historical charge parameters are obtained. The initial machine learning algorithm is trained based on the texture features, edge features, historical charge parameter features, and annotation information of each historical blast hole to establish the automatic identification model for open-pit mine blind blasts based on the target machine learning algorithm.

7. An automatic identification device for blind shots in open-pit mines based on machine learning, characterized in that, include: A data acquisition unit is used to acquire video data transmitted by a drone, wherein the drone is equipped with a camera device, and the video data is obtained by the drone collecting data on the blasting process of the target blasting area of ​​the open-pit mine. A set acquisition unit is used to acquire the set of keyframes corresponding to the video data; The parameter acquisition unit is used to acquire the charge parameters corresponding to the blasting process of the target blasting area; The parameter acquisition unit is also used to acquire the image features of the keyframe set and the parameter features of the charge parameters; The blast hole identification unit is used to identify the image features and parameter features using an open-pit mine blind blasting automatic identification model established based on a target machine learning algorithm, and to obtain the identification results of the unblasted blast holes corresponding to the target blasting area. Specifically, the unit uses a gray-level co-occurrence matrix algorithm to process the keyframe set to obtain the texture features of each blast hole corresponding to the keyframe set; it uses an edge detection algorithm to process the keyframe set to obtain the edge features of each blast hole corresponding to the keyframe set; it obtains the parameter features of the charge parameters based on the charge amount and the rock mass blastability coefficient corresponding to the target blasting area; and it inputs the texture features, edge features, and parameter features of each blast hole into the open-pit mine blind blasting automatic identification model established based on the target machine learning algorithm for identification, thereby obtaining the identification results of the unblasted blast holes corresponding to the target blasting area.

8. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the machine learning-based automatic identification method for blind shots in open-pit mines as described in any one of claims 1 to 6.

9. A storage medium storing instructions, characterized in that, When the instructions are executed on an electronic device, the electronic device performs the machine learning-based automatic identification method for open-pit mine blasting as described in any one of claims 1 to 6.

Images