Tunnel operator risk behavior automatic identification method and system, equipment and medium

Abstract

The invention discloses a tunnel operator risk behavior automatic identification method and system, equipment and a medium, and the method comprises the steps: obtaining the pixel coordinate of each operator in a monitoring image when the existence of a risk behavior in the monitoring image is identified, and constructing a pixel coordinate array of the operators in a camera monitoring range; positioning coordinates and personal information of all operating personnel in the tunnel are obtained, the positioning coordinates of each operating personnel are associated with the personal information of the operating personnel, the positioning coordinates of the operating personnel in the monitoring range of the camera are screened out, and a positioning coordinate array of the operating personnel in the monitoring range is constructed. And finally, performing automatic matching according to the pixel coordinate array and the positioning coordinate array of the operating personnel in the monitoring range of the camera to realize corresponding matching between the pixel coordinates and the positioning coordinates, and combining the relevance between the positioning coordinates and the personal information to obtain the personal information of the risk actor. Therefore, the risk behavior personnel can be accurately assessed and evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel construction safety monitoring, and in particular to a method and system for automatically identifying risky behaviors of tunnel workers, electronic equipment, and a computer-readable storage medium. Background Art

[0002] During the tunnel construction process, it is necessary to assess and evaluate the risk behaviors of TBM tunnel operators. The assessment content generally includes not wearing a safety helmet, not wearing reflective clothing, smoking in illegal areas, etc., which are generally inspected by safety officers. However, since the safety officer's evidence collection process is time-consuming and labor-intensive, there is an urgent need for a method that can effectively automatically identify the risk behaviors of TBM tunnel operators. At present, the existing technology usually uses machine vision technology to automatically identify the risk behaviors of operators. For example, patent CN116311497A discloses a method for detecting abnormal behaviors of tunnel workers based on machine vision. It accurately detects abnormal behaviors of tunnel workers online and in real time based on machine vision theory and issues warnings. It can detect abnormal behaviors of workers. However, this method does not obtain the identity information of the risk behavior person, so it is impossible to automatically match risk behaviors with personal information, and thus it is impossible to conduct personnel assessment and evaluation. Summary of the invention

[0003] The present invention provides a method and system for automatically identifying risky behaviors of tunnel workers, electronic equipment, and a computer-readable storage medium, which can automatically match the information of risky behavior persons while automatically identifying risky behaviors, thereby accurately evaluating and assessing risky behavior persons.

[0004] According to one aspect of the present invention, a method for automatically identifying risky behaviors of tunnel workers is provided, comprising the following contents:

[0005] Obtain the pixel coordinates of each operator in the monitoring image with risky behavior, and construct a pixel coordinate array of the operators within the monitoring range;

[0006] Obtain the location coordinates and personal information of the workers in the tunnel and associate them to build a location coordinate array of the workers within the monitoring range;

[0007] Automatic matching is performed based on the pixel coordinate array and the positioning coordinate array, and the personal information of the risk behavior person is obtained according to the matching result.

[0008] Furthermore, the process of automatically matching based on the pixel coordinate array and the positioning coordinate array and obtaining the personal information of the risk behavior person according to the matching result includes the following:

[0009] Determine whether the number of pixel coordinates in the pixel coordinate array is consistent with the number of positioning coordinates in the positioning coordinate array. If they are consistent, proceed to the subsequent steps.

[0010] Arranging the multiple pixel coordinates in the pixel coordinate array and the multiple positioning coordinates in the positioning coordinate array in position order, and matching the two arrays one by one in order to obtain the corresponding relationship between the pixel coordinates and the positioning coordinates;

[0011] The personal information of the risky behavior person is obtained based on the corresponding relationship between the pixel coordinates of the risky behavior person and the positioning coordinates and the association relationship between the positioning coordinates and the personal information.

[0012] Furthermore, after obtaining the corresponding relationship between the pixel coordinates and the positioning coordinates, the following contents are also included:

[0013] Based on the corresponding relationship between the pixel coordinates and the positioning coordinates, an initial point pair between the pixel coordinate array and the positioning coordinate array is obtained;

[0014] The pixel coordinates and positioning coordinates of each pair of initial points are fused to obtain the fused new coordinates, and a fused coordinate array is constructed;

[0015] Based on the fusion coordinate array and the pixel coordinate array, the closest distance matching is performed to obtain new point pairs, and the coordinates of the new point pairs are fused again;

[0016] The above steps are continuously iterated until the iteration termination condition is met, and the matching relationship between the pixel coordinate array and the positioning coordinate array is obtained.

[0017] Furthermore, the process of fusing the pixel coordinates and the positioning coordinates of each pair of initial point pairs to obtain the fused new coordinates includes the following:

[0018] The pixel coordinates and positioning coordinates are transformed by UT respectively, the positioning coordinates after UT transformation are used as the prior estimation value, and the pixel coordinates after UT transformation are used as the observation value, and the coordinate fusion is performed using the unscented Kalman filter algorithm.

[0019] Furthermore, when the number of pixel coordinates in the pixel coordinate array is inconsistent with the number of positioning coordinates in the positioning coordinate array, the size of the monitoring range is continuously adjusted until the numbers of the two are consistent.

[0020] Furthermore, the iteration loss is calculated based on the following formula:

[0021]

[0022] Among them, Loss represents the iterative loss, n represents the number of pixel coordinates in the pixel coordinate array, and p iRepresents the i-th pixel coordinate in the pixel coordinate array, q′ i represents the jth fused coordinate in the fused coordinate array that is closest to the i-th pixel coordinate, |p i -q′ j | 2 Represents the distance value between the i-th pixel coordinate and the j-th fusion coordinate. When Loss ≤ the preset threshold t, the iteration terminates.

[0023] Furthermore, coordinate fusion is performed based on the following formula:

[0024]

[0025] in, represents the new coordinates after fusion, represents the prior estimate, z represents the observed value, K represents the Kalman gain, S xy represents the covariance matrix between the prior estimates and the observed values, S y Represents the observation covariance matrix, and the superscript T represents the transpose.

[0026] In addition, the present invention also provides a system for automatically identifying risk behaviors of tunnel workers, comprising:

[0027] A pixel coordinate array building module is used to obtain the pixel coordinates of each operator in the monitoring image with risky behavior and build a pixel coordinate array of the operators within the monitoring range;

[0028] A positioning coordinate array construction module is used to obtain the positioning coordinates and personal information of the workers in the tunnel and associate them to construct a positioning coordinate array of the workers within the monitoring range;

[0029] The array automatic matching module is used to automatically match based on the pixel coordinate array and the positioning coordinate array, and obtain the personal information of the risk behavior person according to the matching results.

[0030] In addition, the present invention also provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0031] In addition, the present invention also provides a computer-readable storage medium for storing a computer program for automatically identifying risky behaviors of tunnel workers, wherein the computer program executes the steps of the method described above when running on a computer.

[0032] The present invention has the following beneficial effects:

[0033] The method for automatically identifying risky behaviors of tunnel workers of the present invention first uses machine vision technology to automatically identify risky behaviors in monitoring images. When risky behaviors are identified in monitoring images, the pixel coordinates of each worker in the monitoring image are obtained, thereby constructing a pixel coordinate array of the workers within the camera monitoring range. Then, the positioning coordinates and personal information of all workers in the tunnel are obtained, and the positioning coordinates of each worker are associated with their personal information. At the same time, the positioning coordinates of the workers within the camera monitoring range are screened out, thereby constructing a positioning coordinate array of the workers within the monitoring range. Finally, the pixel coordinate array and the positioning coordinate array of the workers within the camera monitoring range are automatically matched to achieve the corresponding match between the pixel coordinates and the positioning coordinates. Combined with the correlation between the positioning coordinates and the personal information, the personal information of the risky behavior person can be obtained, so that the risky behavior person assessment can be accurately performed.

[0034] In addition, the automatic identification system for risky behaviors of tunnel workers of the present invention also has the above advantages.

[0035] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0037] Figure 1 It is a flow chart of a method for automatically identifying risky behaviors of tunnel workers according to a preferred embodiment of the present application.

[0038] Figure 2 yes Figure 1 Schematic diagram of the sub-process of step S3 in FIG.

[0039] Figure 3 yes Figure 2 Schematic diagram of the sub-process of step S32.

[0040] Figure 4 It is a schematic diagram of the module structure of a system for automatically identifying risky behaviors of tunnel workers in another embodiment of the present application. DETAILED DESCRIPTION

[0041] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0042] Reference Figure 1The preferred embodiment of the present application provides a method for automatically identifying risk behaviors of tunnel workers, comprising the following contents:

[0043] Step S1: Obtain the pixel coordinates of each operator in the monitoring image with risky behavior, and construct a pixel coordinate array of the operators within the monitoring range;

[0044] Step S2: Acquire the location coordinates and personal information of the workers in the tunnel and associate them to construct a location coordinate array of the workers within the monitoring range;

[0045] Step S3: Automatically match based on the pixel coordinate array and the positioning coordinate array, and obtain personal information of the risk behavior person according to the matching result.

[0046] It can be understood that the method for automatically identifying risky behaviors of tunnel workers in this embodiment first uses machine vision technology to automatically identify risky behaviors in monitoring images. When risky behaviors are identified in monitoring images, the pixel coordinates of each worker in the monitoring image are obtained, thereby constructing a pixel coordinate array of the workers within the camera monitoring range. Then, the positioning coordinates and personal information of all workers in the tunnel are obtained, and the positioning coordinates of each worker are associated with their personal information. At the same time, the positioning coordinates of the workers within the camera monitoring range are screened out, thereby constructing a positioning coordinate array of the workers within the monitoring range. Finally, according to the pixel coordinate array and the positioning coordinate array of the workers within the camera monitoring range, automatic matching is performed to achieve the corresponding matching between the pixel coordinates and the positioning coordinates. Combined with the correlation between the positioning coordinates and the personal information, the personal information of the risky behavior person can be obtained, so that the risky behavior person assessment can be accurately carried out.

[0047] It can be understood that before the automatic identification of risk behaviors, it is necessary to deploy optical fiber, switches, wireless bridges and other hardware equipment in the tunnel to build network transmission links in construction areas such as TBM equipment and secondary lining areas. Specifically, since TBM tunnels generally have long construction mileages and the cost of using wireless transmission modules in the entire tunnel is high, the local area network from the information room outside the tunnel to the TBM power distribution room is constructed by fiber-optic access. The coverage of the local area network is further expanded through switches, wireless bridges and other devices, including key monitoring areas of TBM, secondary lining areas, etc., to provide a good network transmission link for UWB data, video monitoring images and other data. In addition, according to the information coverage capability of the UWB base station, base stations are deployed at the TBM equipment, secondary lining area, and tunnel entrance respectively. According to the field of view parameters of the camera and the tunnel environment, AI cameras are deployed in key control areas such as the TBM support area and main beam, and channel gates with face recognition terminals are deployed at the entrance of the tunnel.

[0048] It can be understood that in step S1, each operating area in the tunnel is monitored in real time by an AI camera, and machine vision technologies such as a neural network target detection algorithm are used to automatically identify whether there is risky behavior in the camera monitoring image, and can automatically detect risky behaviors such as wearing a helmet, wearing a reflective vest, smoking detection, abnormal gathering, etc., and frame the pixel coordinate position of the risky behavior person in the image. Among them, the specific detection algorithm belongs to the prior art and will not be repeated here. When it is determined that there is a risky behavior in the monitoring image, the pixel coordinates of each operator in the monitoring image at this time are obtained, so as to construct a pixel coordinate array of the operators within the camera monitoring range. Among them, the pixel coordinate selection criteria of each operator are the same, for example, the face pixel position, hand pixel position, helmet pixel position, etc. of each operator are selected, which can be set according to needs and are not specifically limited here.

[0049] It can be understood that in step S2, the TOF algorithm is used to calculate the positioning mileage of each operator in the tunnel, and the positioning coordinates of each operator can be obtained. At the same time, when the operator performs face recognition at the channel gate, the UWB tag information closest to the gate is obtained at the same time, so that the association between the face information and the tag information can be realized. In addition, the gate is opened only when the face recognition result is consistent with the tag personnel information, ensuring the correctness of the identity of the personnel when entering the site, and ensuring that the UWB positioning tag is worn in place. Then, the positioning tags within the monitoring range of the camera are screened out, so as to construct the positioning coordinate array of the operators within the monitoring range of the camera. Among them, the TOF algorithm belongs to the existing algorithm, and the specific tag positioning process is not repeated here. In addition, the monitoring range of the camera can be converted according to the parameters such as the field of view angle parameters and the installation position of the camera. The specific conversion process belongs to the prior art and is not repeated here.

[0050] It is understandable that Figure 2 As shown, in step S3, the process of automatically matching based on the pixel coordinate array and the positioning coordinate array and obtaining the personal information of the risk behavior person according to the matching result includes the following:

[0051] Step S31: determining whether the number of pixel coordinates in the pixel coordinate array is consistent with the number of positioning coordinates in the positioning coordinate array, if they are consistent, proceeding to the subsequent steps;

[0052] Step S32: Arrange the multiple pixel coordinates in the pixel coordinate array and the multiple positioning coordinates in the positioning coordinate array in position order, and match the two arrays one by one in order to obtain the corresponding relationship between the pixel coordinates and the positioning coordinates;

[0053] Step S33: obtaining the personal information of the risky behavior person according to the corresponding relationship between the pixel coordinates of the risky behavior person and the positioning coordinates and the association relationship between the positioning coordinates and the personal information.

[0054] Specifically, assuming that the pixel coordinate array is P and the positioning coordinate array is Q, first determine whether the number of coordinate points in P and Q is consistent, and only if they are consistent will the subsequent matching steps be performed. When the number of coordinate points in P and Q is consistent, the pixel coordinates in the pixel coordinate array P are arranged in order from left to right according to the position, and the positioning coordinates in the positioning coordinate array Q are arranged in order from small to large according to the positioning mileage, and then P and Q are matched one by one in order, and the automatic matching between the pixel coordinate array P and the positioning coordinate array Q can be completed, thereby obtaining the corresponding relationship between the pixel coordinates and the positioning coordinates. Finally, according to the positioning coordinates corresponding to the pixel coordinates of the risk behavior personnel, and the association relationship between the positioning coordinates and the personal information, the personal information of the risk behavior personnel can be obtained. Among them, in other embodiments of the present invention, the pixel coordinates in the pixel coordinate array P can also be arranged in order from right to left according to the position, and the positioning coordinates in the positioning coordinate array Q can be arranged in order from large to small according to the positioning mileage, and then matched one by one.

[0055] Optionally, when the number of pixel coordinates in the pixel coordinate array is inconsistent with the number of positioning coordinates in the positioning coordinate array, the size of the monitoring range is continuously adjusted until the number of the two is consistent. It can be understood that the monitoring range of the camera along the tunnel axis direction is D1~D2, which is converted according to the field of view parameters and installation position parameters of the camera, and the monitoring range is corrected to (D1-a)~(D2+a), where a represents a correction parameter, which is a calibration value set according to the camera range error. The correction parameter a is adjusted until the number of pixel coordinates in the pixel coordinate array is consistent with the number of positioning coordinates in the positioning coordinate array. For example, if the number of pixel coordinates in the pixel coordinate array P is greater than the number of positioning coordinates in the positioning coordinate array Q, the correction parameter a is increased until the number of the two is consistent; if the number of pixel coordinates in the pixel coordinate array P is less than the number of positioning coordinates in the positioning coordinate array Q, the correction parameter a is reduced until the number of the two is consistent.

[0056] It is understandable that due to the influence of factors such as UWB positioning accuracy, overlapping of operator positions, pixel coordinate position selection, etc., the matching accuracy of the above matching method may have errors, which may lead to errors in obtaining personal information of risky behavior personnel. Figure 3 As shown, in the step S32, after obtaining the corresponding relationship between the pixel coordinates and the positioning coordinates, the following contents are also included:

[0057] Step S321: obtaining an initial point pair between the pixel coordinate array and the positioning coordinate array based on the corresponding relationship between the pixel coordinates and the positioning coordinates;

[0058] Step S322: coordinate fusion is performed on the pixel coordinates and the positioning coordinates of each pair of initial point pairs to obtain fused new coordinates, and a fused coordinate array is constructed;

[0059] Step S323: performing the closest distance matching based on the fused coordinate array and the pixel coordinate array to obtain a new point pair, and performing coordinate fusion on the new point pair again;

[0060] Step S324: continuously iterate the above steps until the iteration termination condition is met, and obtain the matching relationship between the pixel coordinate array and the positioning coordinate array.

[0061] Specifically, in step S321, the initial point pair between the pixel coordinate array P and the positioning coordinate array Q is obtained according to the correspondence between the pixel coordinates and the positioning coordinates in the pixel coordinate array P and the positioning coordinate array Q, wherein a pixel coordinate and the corresponding UWB positioning coordinate constitute a point pair, for example, i and q j Form a pair of points, p i represents the i-th pixel coordinate in the pixel coordinate array P, q j represents the j-th pixel coordinate in the positioning coordinate array Q. After the sequential matching in step S32, p i and q j Correspondingly, record (i, j) at this time.

[0062] In step S322, coordinate fusion is performed on the pixel coordinates and positioning coordinates of each pair of initial point pairs to obtain fused coordinates, and a fused coordinate array is constructed. The process of fusion of the pixel coordinates and positioning coordinates of each pair of initial point pairs to obtain fused new coordinates includes the following:

[0063] The pixel coordinates and positioning coordinates are transformed by UT respectively, the positioning coordinates after UT transformation are used as the prior estimation value, and the pixel coordinates after UT transformation are used as the observation value, and the coordinate fusion is performed using the unscented Kalman filter algorithm.

[0064] Specifically, UT transformation is performed on the pixel coordinates and the positioning coordinates respectively, where the UT transformation formula is: u0 represents the pixel coordinates or positioning coordinates before UT transformation, which is a two-dimensional vector. u represents the pixel coordinates or positioning coordinates after UT transformation. N represents the Gaussian distribution dimension. E represents an adjustable parameter, which is usually set to 3-N. L represents the covariance matrix of u. Represents (N+E)L iThe i-th column of the square root of the lower triangular decomposition of . Then, the UWB positioning coordinates after UT transformation are used as the prior estimate The pixel coordinates after UT transformation are taken as the observation value z, the unscented Kalman gain is calculated, and the two coordinate information are fused using the unscented Kalman gain. The specific fusion formula is as follows:

[0065]

[0066] in, represents the new coordinates after fusion, represents the prior estimate, i.e., the UWB positioning coordinate after UT transformation, z represents the observation value, i.e., the pixel coordinate after UT transformation, K represents the unscented Kalman gain, S xy represents the covariance matrix between the prior estimates and the observed values, S y represents the observation covariance matrix, S y = cov(z,z), where the superscript T indicates transposition. After that, a new fusion coordinate array Q′ can be constructed.

[0067] In step S323, the nearest distance matching is performed based on the fused coordinate array Q′ and the pixel coordinate array P. The distance calculation can use the Euclidean distance or the Manhattan distance to obtain the matching relationship between each fused coordinate in the fused coordinate array Q′ and each pixel coordinate in the pixel coordinate array P, construct a new point pair, and use the new point pair as the input for the next round of iteration. In addition, when performing the nearest distance calculation, only the 6 positioning coordinates around each pixel coordinate are calculated instead of traversing all positioning coordinates, so as to improve the calculation speed.

[0068] In step S324, the above steps S322 and S323 are continuously iterated until the iteration termination condition is met, and the matching relationship between the pixel coordinate array P and the final fusion coordinate array Q', that is, the corresponding relationship between the pixel coordinates and the fusion coordinates, can be obtained. In each iteration process, the matching relationship between the fusion coordinate array Q' and the positioning coordinate array Q, that is, the corresponding relationship between the fusion coordinates and the positioning coordinates, is recorded, so that the final matching relationship between the pixel coordinate array P and the positioning coordinate array Q can be obtained. Optionally, the iteration loss is calculated based on the following formula:

[0069]

[0070] Among them, Loss represents the iterative loss, n represents the number of pixel coordinates in the pixel coordinate array, and p i Represents the i-th pixel coordinate in the pixel coordinate array, q′ irepresents the jth fused coordinate in the fused coordinate array that is closest to the i-th pixel coordinate, |p i -q′ j | 2 Represents the distance value between the i-th pixel coordinate and the j-th fusion coordinate. When Loss ≤ the preset threshold t, the iteration terminates. When Loss > t or the number of iterations is less than 10, the iteration continues.

[0071] It can be understood that the present invention adopts the point pair results of the previous sequential matching as the initial input of the iteration, which is conducive to improving the iteration efficiency, and uses the unscented Kalman filter algorithm to perform iterative matching of the two-dimensional point cloud, which can more accurately achieve precise matching between the pixel coordinate array and the positioning coordinate array, thereby improving the matching accuracy.

[0072] Optionally, before obtaining the initial point pair, the coordinate data of the pixel coordinate array P and the positioning coordinate array Q need to be normalized. Specifically, the center point coordinates of the pixel coordinate array P and the positioning coordinate array Q are first calculated respectively, and the calculation formula is: Among them, p center Represents the pixel coordinates of the center point of the pixel coordinate array P, p i represents the pixel coordinates of the i-th position in the pixel coordinate array P, n represents the number of coordinate points in the pixel coordinate array P and the positioning coordinate array Q, q center Represents the center point positioning coordinates of the positioning coordinate array Q, q i represents the location coordinates of the i-th position point in the location coordinate array Q. Then, based on the center point coordinates p center and q center Normalization is performed separately, and the specific normalization formula is: p′ i =p i -p center ,q′ i =q i -q center , p′ i and q′ i They represent the normalized i-th pixel coordinate and i-th positioning coordinate respectively.

[0073] In addition, optionally, in order to reduce the amount of calculation, the coordinate values ​​such as pixel coordinates and positioning coordinates in the present invention are calculated using only the coordinate values ​​in the x direction (ie, the tunnel axis direction).

[0074] In addition, if Figure 4 As shown, another embodiment of the present invention further provides a system for automatically identifying risk behaviors of tunnel workers, preferably using the method as described above, including:

[0075] A pixel coordinate array building module is used to obtain the pixel coordinates of each operator in the monitoring image with risky behavior and build a pixel coordinate array of the operators within the monitoring range;

[0076] A positioning coordinate array construction module is used to obtain the positioning coordinates and personal information of the workers in the tunnel and associate them to construct a positioning coordinate array of the workers within the monitoring range;

[0077] The array automatic matching module is used to automatically match based on the pixel coordinate array and the positioning coordinate array, and obtain the personal information of the risk behavior person according to the matching results.

[0078] It can be understood that the automatic identification system for risky behaviors of tunnel workers in this embodiment first uses machine vision technology to automatically identify risky behaviors in monitoring images. When risky behaviors are identified in monitoring images, the pixel coordinates of each worker in the monitoring image are obtained, thereby constructing a pixel coordinate array of the workers within the camera monitoring range. Then, the positioning coordinates and personal information of all workers in the tunnel are obtained, and the positioning coordinates of each worker are associated with their personal information. At the same time, the positioning coordinates of the workers within the camera monitoring range are screened out, thereby constructing a positioning coordinate array of the workers within the monitoring range. Finally, the pixel coordinate array and the positioning coordinate array of the workers within the camera monitoring range are automatically matched to achieve the corresponding match between the pixel coordinates and the positioning coordinates. Combined with the correlation between the positioning coordinates and the personal information, the personal information of the risky behavior person can be obtained, so that the risky behavior person assessment can be accurately carried out.

[0079] In addition, another embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the above method by calling the computer program stored in the memory.

[0080] In addition, another embodiment of the present invention further provides a computer-readable storage medium for storing a computer program for automatically identifying risky behaviors of tunnel workers, wherein the computer program executes the steps of the method described above when running on a computer.

[0081] The general form of computer readable storage media includes: floppy disk, flexible disk, hard disk, magnetic tape, any other magnetic medium, CD-ROM, any other optical medium, punch cards, paper tape, any other physical medium with a pattern of holes, random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), flash erasable programmable read-only memory (FLASH-EPROM), any other memory chip or cartridge, or any other medium that can be read by a computer. The instructions can further be transmitted or received by a transmission medium. The term transmission medium can include any tangible or intangible medium that can be used to store, encode or carry instructions for execution by a machine, and includes digital or analog communication signals or other intangible media that facilitate the communication of the above instructions. Transmission media include coaxial cables, copper wires and optical fibers, which include the wires of a bus used to transmit a computer data signal.

[0082] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0083] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0084] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0085] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0086] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0087] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0088] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for automatically identifying risky behaviors of tunnel workers, characterized in that: Includes the following: Obtain the pixel coordinates of each operator in the monitoring image with risky behavior, and construct a pixel coordinate array of the operators within the monitoring range; Obtain the location coordinates and personal information of the workers in the tunnel and associate them to build a location coordinate array of the workers within the monitoring range; Automatic matching is performed based on the pixel coordinate array and the positioning coordinate array, and the personal information of the risk behavior person is obtained according to the matching result.

2. The method for automatically identifying risk behaviors of tunnel workers according to claim 1, characterized in that: The process of automatically matching based on the pixel coordinate array and the positioning coordinate array and obtaining the personal information of the risk behavior person according to the matching result includes the following: Determine whether the number of pixel coordinates in the pixel coordinate array is consistent with the number of positioning coordinates in the positioning coordinate array. If they are consistent, proceed to the subsequent steps. Arranging the multiple pixel coordinates in the pixel coordinate array and the multiple positioning coordinates in the positioning coordinate array in position order, and matching the two arrays one by one in order to obtain the corresponding relationship between the pixel coordinates and the positioning coordinates; The personal information of the risky behavior person is obtained based on the corresponding relationship between the pixel coordinates of the risky behavior person and the positioning coordinates and the association relationship between the positioning coordinates and the personal information.

3. The method for automatically identifying risk behaviors of tunnel workers according to claim 2, characterized in that: After obtaining the correspondence between pixel coordinates and positioning coordinates, the following contents are also included: Based on the corresponding relationship between the pixel coordinates and the positioning coordinates, an initial point pair between the pixel coordinate array and the positioning coordinate array is obtained; The pixel coordinates and positioning coordinates of each pair of initial points are fused to obtain the fused new coordinates, and a fused coordinate array is constructed; Based on the fusion coordinate array and the pixel coordinate array, the closest distance matching is performed to obtain new point pairs, and the coordinates of the new point pairs are fused again; The above steps are continuously iterated until the iteration termination condition is met, and the matching relationship between the pixel coordinate array and the positioning coordinate array is obtained.

4. The method for automatically identifying risk behaviors of tunnel workers according to claim 3, characterized in that: The process of fusing the pixel coordinates and the positioning coordinates of each pair of initial points to obtain the fused new coordinates includes the following: The pixel coordinates and positioning coordinates are transformed by UT respectively, the positioning coordinates after UT transformation are used as the prior estimation value, and the pixel coordinates after UT transformation are used as the observation value, and the coordinate fusion is performed using the unscented Kalman filter algorithm.

5. The method for automatically identifying risk behaviors of tunnel workers according to claim 2, characterized in that: When the number of pixel coordinates in the pixel coordinate array is inconsistent with the number of positioning coordinates in the positioning coordinate array, the size of the monitoring range is continuously adjusted until the numbers of the two are consistent.

6. The method for automatically identifying risk behaviors of tunnel workers according to claim 3, characterized in that: The iterative loss is calculated based on the following formula: Among them, Loss represents the iterative loss, n represents the number of pixel coordinates in the pixel coordinate array, and p i Represents the i-th pixel coordinate in the pixel coordinate array, q′ i represents the jth fused coordinate in the fused coordinate array that is closest to the i-th pixel coordinate, |p i -q′ j 2 Represents the distance value between the i-th pixel coordinate and the j-th fusion coordinate. When Loss ≤ the preset threshold t, the iteration terminates.

7. The method for automatically identifying risk behaviors of tunnel workers according to claim 4, characterized in that: Coordinate fusion is performed based on the following formula: in, represents the new coordinates after fusion, represents the prior estimate, z represents the observed value, K represents the Kalman gain, S xy represents the covariance matrix between the prior estimates and the observed values, S y Represents the observation covariance matrix, and the superscript T represents the transpose.

8. An automatic identification system for risky behaviors of tunnel workers, characterized in that: include: A pixel coordinate array building module is used to obtain the pixel coordinates of each operator in the monitoring image with risky behavior and build a pixel coordinate array of the operators within the monitoring range; A positioning coordinate array construction module is used to obtain the positioning coordinates and personal information of the workers in the tunnel and associate them to construct a positioning coordinate array of the workers within the monitoring range; The array automatic matching module is used to automatically match based on the pixel coordinate array and the positioning coordinate array, and obtain the personal information of the risk behavior person according to the matching results.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the steps of the method according to any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium for storing a computer program for automatically identifying risky behaviors of tunnel workers, characterized in that: When the computer program is run on a computer, the steps of the method according to any one of claims 1 to 7 are executed.

Images