Intelligent management system and method for electronic safety tools

Abstract

The invention discloses an intelligent management system and method for an electronic safety tool, and the method comprises the steps: automatically photographing a tool image when the tool is put in and out of a warehouse, carrying out the multi-level feature analysis of the tool image through employing a computer vision technology based on deep learning, and capturing the boundary features and global spatial structure features of the tool, and multi-level tool state information is obtained through joint perception and dependency relationship modeling of the two, so that intelligent detection of the integrity of the tools and intelligent management of in and out of a warehouse of the tools are realized. Therefore, the tool management efficiency and accuracy can be remarkably improved, human errors are reduced, and operation safety is ensured.

Description

Technical Field

[0001] The present application relates to the field of intelligent management, and more specifically, to an intelligent management system and method for electronic safety tools. Background Art

[0002] In today's industrialized society, industries such as electricity, chemicals, and manufacturing form the cornerstone of the global economy. In these industries, whether it is a towering power plant or a busy production line, safety is always the top priority of business operations, and the management of safety tools is a key link in ensuring safe and efficient production operations, and its importance cannot be ignored.

[0003] Safety tools, including protective clothing, safety helmets, insulating tools, protective glasses, etc., are an important barrier for workers to prevent accidental injuries at work. The integrity of their condition directly affects the life safety of workers and the normal progress of production activities.

[0004] Traditionally, the management of tools and equipment mainly relies on manual records and regular inspections. However, this method has many shortcomings, such as low efficiency, error-proneness, and difficulty in traceability. First, manual records are prone to errors, and it is difficult to track the current status and location of tools and equipment in real time, resulting in low management efficiency. Secondly, regular inspections often rely on the experience and subjective judgment of operators, and it is difficult to ensure the comprehensiveness and accuracy of the inspections, especially when faced with a large number of tools and equipment of various types. This problem is particularly prominent. Especially for occasions that require a high degree of safety assurance, such as power system maintenance, chemical production environment, etc., traditional manual management methods often cannot meet the high safety standards and real-time monitoring of tool management needs.

[0005] Therefore, an intelligent electronic safety tool management method is expected. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides an intelligent management system and method for electronic safety tools, which automatically captures tool images when tools are in and out of the warehouse, and uses deep learning-based computer vision technology to perform multi-level feature analysis on tool images, respectively capturing the boundary features and global spatial structure features of the tools, and then through joint perception and dependency modeling of the two, to obtain multi-level tool status information, thereby realizing intelligent detection of tool integrity and intelligent management of tool in and out of the warehouse. In this way, the efficiency and accuracy of tool management can be significantly improved, human errors can be reduced, and operational safety can be ensured.

[0007] According to one aspect of the present application, there is provided an intelligent management system for electronic safety tools, which includes:

[0008] A tool record management module is used to input tool information and predict the inspection cycle based on the tool type and usage information;

[0009] An in-and-out management module, used to automatically take images of the tools when they are in and out of the warehouse, and to perform integrity checks based on the images of the tools and archive records;

[0010] An expected reminder module, used for automatically reminding the due inspection based on the detection cycle;

[0011] A statistical analysis module, used to count the usage frequency and usage time of the tools and generate statistical reports;

[0012] Wherein, the inbound and outbound management module includes:

[0013] A tool image acquisition unit, used to acquire an image of the tool;

[0014] An image multi-level feature extraction unit, used to extract boundary features and global context features of the image of the tool respectively to obtain a tool surface state boundary feature map and a tool surface state global feature map;

[0015] A feature autocorrelation enhancement unit, used for performing feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map;

[0016] A multi-level feature joint perception unit, used for jointly perceiving the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map;

[0017] The tool integrity analysis unit is used to generate a tool integrity detection result based on the multi-scale joint perception feature map of the tool surface state.

[0018] According to another aspect of the present application, there is provided an intelligent management method for electronic safety tools, which includes:

[0019] Acquiring an image of the tool;

[0020] Extracting boundary features and global context features of the image of the tool respectively to obtain a tool surface state boundary feature map and a tool surface state global feature map;

[0021] Performing feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map;

[0022] Performing joint perception on the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map;

[0023] Based on the multi-scale joint perception feature map of the tool surface state, an integrity detection result of the tool is generated.

[0024] Compared with the prior art, the electronic safety tool intelligent management system and method provided by the present application automatically captures tool images when tools are in and out of the warehouse, and uses computer vision technology based on deep learning to perform multi-level feature analysis on tool images, respectively capturing the boundary features and global spatial structure features of tools, and then through joint perception and dependency modeling of the two, to obtain multi-level tool status information, thereby realizing intelligent detection of tool integrity and intelligent management of tool in and out of the warehouse. In this way, the efficiency and accuracy of tool management can be significantly improved, human errors can be reduced, and operation safety can be ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0026] Figure 1 is a block diagram of an intelligent management system for electronic safety tools according to an embodiment of the present application;

[0027] Figure 2 A data flow diagram of an intelligent management system for electronic safety tools according to an embodiment of the present application;

[0028] Figure 3 A block diagram of an in-and-out warehouse management module in an intelligent management system for electronic safety tools according to an embodiment of the present application;

[0029] Figure 4 A block diagram of an image multi-level feature extraction unit in an intelligent management system for electronic safety tools according to an embodiment of the present application;

[0030] Figure 5 Flow chart of the intelligent management method of electronic safety tools according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0032] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0033] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0034] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0035] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0036] Traditionally, the management of tools and equipment mainly relies on manual records and regular inspections. However, this method has many shortcomings, such as low efficiency, error-proneness, and difficulty in traceability. First, manual records are prone to errors, and it is difficult to track the current status and location of tools and equipment in real time, resulting in low management efficiency. Secondly, regular inspections often rely on the experience and subjective judgment of operators, and it is difficult to ensure the comprehensiveness and accuracy of the inspection, especially when faced with a large number of tools and equipment of various types. This problem is particularly prominent. Especially for occasions that require a high degree of safety assurance, such as power system maintenance, chemical production environment, etc., traditional manual management methods often cannot meet the high safety standards and real-time monitoring of tool management needs. Therefore, an intelligent electronic safety tool management method is expected.

[0037] In the technical solution of the present application, an intelligent management system for electronic safety tools is proposed. Figure 14 is a block diagram of an intelligent management system for electronic safety tools according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the intelligent management system for electronic safety tools according to an embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the intelligent management system 300 for electronic safety tools includes: a tool inventory management module 310, which is used to input the information of tools and predict the detection cycle based on the type and usage information of the tools; an entry and exit management module 320, which is used to automatically take images of the tools when the tools are in and out of the warehouse, and perform integrity checks based on the images of the tools and archive the records; an expected reminder module 330, which is used to automatically remind due inspections based on the detection cycle; a statistical analysis module 340, which is used to count the usage frequency and usage time of the tools and generate statistical reports.

[0038] In particular, the tool inventory management module 310 is used to input tool information and predict the inspection cycle based on the type and usage information of the tool. It should be understood that different types of tools may have different inspection cycles. For example, high-risk tools may require more frequent inspections. In the technical solution of the present application, the inspection plan of tools is optimized to ensure that tools are regularly inspected and maintained. Specifically, a multi-level feature analysis of tool images is performed using deep learning-based computer vision technology to capture the boundary features and global spatial structure features of the tools, respectively, and then the joint perception and dependency modeling of the two are performed to obtain multi-level tool status information, thereby realizing intelligent detection of tool integrity and intelligent management of tool entry and exit, thereby improving the efficiency of tool management and safety of use.

[0039] In particular, the in-and-out management module 320 is used to automatically capture images of the tools when the tools are in and out of the warehouse, and to perform integrity checks based on the images of the tools and archive the records. In particular, in a specific example of the present application, Figure 3As shown, the in-and-out warehouse management module 320 includes: a tool image acquisition unit 321, used to acquire the image of the tool; an image multi-level feature extraction unit 322, used to extract the boundary features and global context features of the tool image respectively to obtain a tool surface state boundary feature map and a tool surface state global feature map; a feature autocorrelation enhancement unit 323, used to perform feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map; a multi-level feature joint perception unit 324, used to perform joint perception on the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map; a tool integrity analysis unit 325, used to generate a tool integrity detection result based on the tool surface state multi-scale joint perception feature map.

[0040] Specifically, the tool image acquisition unit 321 is used to acquire the image of the tool. It should be understood that by using image capture technology to collect images of tools, not only can the computer vision algorithm be combined to achieve automated integrity detection, but the image can also be saved as a historical record to facilitate subsequent data analysis and information tracing, and to be used to count the use frequency, damage status and other information of tools, so as to provide data support for the maintenance, replacement and management of tools.

[0041] Specifically, the image multi-level feature extraction unit 322 is used to extract the boundary features and global context features of the image of the tool to obtain a tool surface state boundary feature map and a tool surface state global feature map. In particular, in a specific example of the present application, Figure 4 As shown, the image multi-level feature extraction unit 322 includes: an image grayscale sub-unit 3221, used for performing grayscale processing on the image of the tool to obtain a grayscale image of the tool surface state; a multi-level feature extraction sub-unit 3222, used for inputting the tool surface state grayscale image into the MBCNet model including a boundary feature extraction branch and a backbone network to obtain a boundary feature map of the tool surface state and a global feature map of the tool surface state.

[0042] More specifically, the image grayscale subunit 3221 is used to grayscale the image of the tool to obtain a grayscale image of the tool surface state. Considering that during the image acquisition process, the image of the tool may be affected by changes in illumination, resulting in color deviation in the image, thereby affecting the accuracy of integrity detection. Therefore, the present application further grayscales the image of the tool to obtain a grayscale image of the tool surface state, thereby eliminating color differences, simplifying the complexity of subsequent feature extraction, and making the edge texture features of the tool in the image more prominent, so as to improve the accuracy of subsequent feature analysis.

[0043] More specifically, the multi-level feature extraction subunit 3222 is used to input the grayscale image of the surface state of the tool into the MBCNet model including the boundary feature extraction branch and the backbone network to obtain the boundary feature map of the surface state of the tool and the global feature map of the surface state of the tool. Considering that in the process of extracting features from the grayscale image of the surface state of the tool using the neural network model, as the number of network layers increases, the granularity of feature expression will gradually refine from local to global, resulting in the loss of image boundary information. Therefore, in order to simultaneously retain the boundary features and global structural features of the tool, the present application uses the MBCNet model including the boundary feature extraction branch and the backbone network to process the grayscale image of the surface state of the tool. Among them, the boundary feature extraction branch adopts a lightweight network structure with a small receptive field, focusing on capturing the edges and local details of the tool, and generating a boundary feature map of the surface state of the tool; while the backbone network adopts a deep large receptive field network structure to obtain the global context information of the tool and generate a global feature map of the surface state of the tool. This bifurcated network design can ensure that boundary information is not lost while extracting global features, thereby improving the accuracy of tool integrity detection.

[0044] It is worth mentioning that in other specific examples of the present application, the boundary features and global context features of the image of the tool can be extracted separately in other ways to obtain the boundary feature map of the tool surface state and the global feature map of the tool surface state, for example: input the image of the tool; use an edge detection algorithm (such as Canny edge detection) to extract the boundary from the tool image to obtain the boundary feature map of the tool surface state; wherein the boundary feature map is a binary image, in which the boundary pixels are 1 and the non-boundary pixels are 0. Use a global pooling layer to extract global context features from the tool image to obtain the global feature map of the tool surface state.

[0045] Specifically, the feature autocorrelation enhancement unit 323 is used to perform feature autocorrelation enhancement on the global feature map of the tool surface state to obtain an enhanced global feature map of the tool surface state. In a specific example of the present application, the global feature map of the tool surface state is input into a self-attention enhanced cross-channel feature space structure preservation module to obtain an enhanced global feature map of the tool surface state. It should be understood that in order to further enhance the global structural feature representation of the tool surface state, the present application introduces a self-attention enhanced cross-channel feature space structure preservation module to perform feature enhancement processing on the global feature map of the tool surface state. Specifically, the module first performs layer normalization processing on the global feature map of the tool surface state to eliminate scale differences between layers and ensure the stability of subsequent feature processing. Then, multi-layer convolution operations are used to capture the channel context association and spatial structure information of the global feature map of the tool surface state, and the channel context association information is used as the query, and the spatial structure information is used as the key and value. Based on the self-attention mechanism, the feature cross-channel global association interaction is performed. While retaining the spatial structure of the feature map, the channel association information between feature structures is fused in units of feature spatial structure, and the expressive power of the feature map is enhanced to generate an enhanced global feature map of the tool surface state, providing a more accurate feature representation for tool integrity assessment.

[0046] In an embodiment of the present application, feature autocorrelation enhancement is performed on the global feature map of the tool surface state to obtain an enhanced global feature map of the tool surface state, including: layer normalization is performed on the global feature map of the tool surface state to obtain a normalized global feature map of the chemical tool surface state; point convolution is performed on the normalized global feature map of the chemical tool surface state to obtain a tool surface state channel context association representation feature map; convolution encoding is performed on the tool surface state channel context association representation feature map to obtain a tool surface state space context association representation feature map; channel-space global interactive attention fusion is performed on the tool surface state channel context association representation feature map and the tool surface state space context association representation feature map to obtain the tool surface state enhanced global feature map.

[0047] Among them, the process of performing channel-space global interactive attention fusion on the tool surface state channel context association representation feature map and the tool surface state space context association representation feature map to obtain the tool surface state enhanced global feature map includes: copying the tool surface state space context association representation feature map to obtain a backup tool surface state space context association representation feature map; performing feature shape reshaping on the tool surface state channel context association representation feature map, the tool surface state space context association representation feature map and the backup tool surface state space context association representation feature map to obtain a tool surface state channel context association representation feature matrix, a tool surface state space context association representation feature matrix and A backup tool surface state space context association representation feature matrix; calculating the cross-channel cross covariance matrix between the tool surface state channel context association representation feature matrix and the tool surface state space context association representation feature matrix; using the Softmax function to activate the cross-channel cross covariance matrix to obtain a tool surface state feature global interactive attention matrix; calculating the product between the backup tool surface state space context association representation feature matrix and the tool surface state feature global interactive attention matrix to obtain an attention-enhanced tool surface state feature representation matrix; and reshaping the attention-enhanced tool surface state feature representation matrix to obtain the tool surface state enhanced global feature map.

[0048] In summary, in the above embodiment, the global feature map of the tool surface state is subjected to feature autocorrelation enhancement to obtain the enhanced global feature map of the tool surface state, including: processing the global feature map of the tool surface state with the following autocorrelation attention enhancement formula to obtain the enhanced global feature map of the tool surface state, wherein the autocorrelation attention enhancement formula is:

[0049] F ln =Layer Normalization(F i )

[0050] F Q =Conv 1×1 (F ln )

[0051] F K =Conv 3×3 (Conv 1×1 (F ln ))

[0052] F V =Copy(F K )

[0053] MQ =Reshape(F Q )

[0054] M K =Reshape(F K )

[0055] M V =Reshape(F V )

[0056]

[0057]

[0058] Among them, F i represents the global feature map of the tool surface state, Layer Normalization (·) represents the layer normalization operation, and F ln Represents the normalized global feature map of the surface state of the chemical tool, Conv 1×1 represents point convolution, F Q Represents the tool surface state channel context association representation feature map, Conv 3×3 represents the convolution processing based on 3×3 convolution kernel, F K represents the tool surface state space context association representation feature map, Copy(·) represents the copy operation, F V represents the backup tool surface state space context association representation feature map, reshape(·) represents the feature shape reshaping, M Q 、M K and M V represents the tool surface state channel context association representation feature matrix, the tool surface state space context association representation feature matrix and the backup tool surface state space context association representation feature matrix, respectively. a represents the cross-channel cross covariance matrix, θ is the scaling factor, softmax represents the normalized exponential function, Represents matrix multiplication operation, F a A global feature map showing the surface state of the tool is enhanced.

[0059] Specifically, the multi-level feature joint perception unit 324 is used to perform joint perception on the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain the tool surface state multi-scale joint perception feature map. In a specific example of the present application, the tool surface state boundary feature map and the tool surface state enhanced global feature map are input into the significant joint perception module based on the attention mechanism to obtain the tool surface state multi-scale joint perception feature map. It should be understood that the tool surface state boundary feature map mainly focuses on the edge, contour and other detail information of the tool surface, while the tool surface state enhanced global feature map emphasizes the global shape and structural features of the tool. In order to more comprehensively describe the integrity state of the tool, the present application introduces a significant joint perception module based on the attention mechanism to fuse the tool surface state boundary feature map and the tool surface state enhanced global feature map. Based on the dependency relationship between the boundary features of the tool surface and the global structural features, the deep fusion of multi-level features is realized, and the integrity of the tool is jointly perceived. Specifically, the module first reshapes the boundary feature map of the tool surface state and the enhanced global feature map of the tool surface state into feature matrices so that their dimensions and shapes are adapted to the input requirements of the attention mechanism. Then, the bidirectional dependency between the two sets of feature matrices is calculated through the feature channel-by-channel interaction perception module, the interaction between the features is quantified, and the boundary-global and global-boundary dependency matrices are generated, which serve as the guidance for feature attention optimization fusion. At the same time, in order to reduce the model's sensitivity to noise and prevent overfitting, the two sets of dependency matrices are further subjected to random inactivation processing, and the generalization ability of the model is enhanced by selectively ignoring or suppressing part of the feature interaction. Furthermore, based on the two sets of dependency matrices after random inactivation processing, the two sets of feature matrices after shape reshaping are weighted optimized, and then restored to the original feature map form by shape reshaping and weighted fusion, and the multi-scale joint perception feature map of the tool surface state is generated, thereby effectively improving the diversity and richness of feature representation, making the integrity assessment of the tool more accurate and comprehensive.

[0060] In an embodiment of the present application, the tool surface state boundary feature map and the tool surface state enhanced global feature map are jointly perceived to obtain a tool surface state multi-scale joint perception feature map, including: reshaping the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state boundary feature shape reshaping matrix and a tool surface state global feature shape reshaping matrix; inputting the tool surface state boundary feature shape reshaping matrix and the tool surface state global feature shape reshaping matrix into a feature channel-by-channel interactive perception module to obtain a boundary-global dependency matrix and a global-boundary dependency matrix; inputting the boundary-global dependency matrix and the global-boundary dependency matrix into a random dropout module to obtain a pruned boundary-global dependency matrix and a pruned global- A boundary dependency matrix; matrix multiplying the pruned boundary-global dependency matrix with the tool surface state global feature shape reshaping matrix to obtain a dependency-optimized tool surface state global feature matrix; matrix multiplying the pruned global-boundary dependency matrix with the tool surface state boundary feature shape reshaping matrix to obtain a dependency-optimized tool surface state boundary feature matrix; feature shape reshaping the dependency-optimized tool surface state boundary feature matrix and the dependency-optimized tool surface state global feature matrix to obtain an optimized tool surface state boundary feature map and an optimized tool surface state enhanced global feature map; calculating the weighted sum of the optimized tool surface state boundary feature map and the optimized tool surface state enhanced global feature map to obtain the tool surface state multi-scale joint perception feature map.

[0061] Among them, the process of inputting the tool surface state boundary feature shape reshaping matrix and the tool surface state global feature shape reshaping matrix into the feature channel-by-channel interactive perception module to obtain the boundary-global dependency matrix and the global-boundary dependency matrix includes: calculating the tool surface state boundary feature shape reshaping matrix multiplied by the transposed matrix of the tool surface state global feature shape reshaping matrix to obtain the tool surface state boundary-global association representation matrix; dividing the tool surface state boundary-global association representation matrix by the scale of the tool surface state global feature shape reshaping matrix and then inputting it into the softmax function to obtain the boundary-global dependency matrix; calculating the transposed matrix of the tool surface state boundary feature shape reshaping matrix multiplied by the tool surface state global feature shape reshaping matrix to obtain the tool surface state global-boundary association representation matrix; dividing the tool surface state boundary-global association representation matrix by the scale of the tool surface state boundary feature shape reshaping matrix and then inputting it into the softmax function to obtain the global-boundary dependency matrix.

[0062] In summary, in the above embodiment, the tool surface state boundary feature map and the tool surface state enhanced global feature map are jointly perceived to obtain a tool surface state multi-scale joint perception feature map, including: processing the tool surface state boundary feature map and the tool surface state enhanced global feature map using the following joint perception formula to obtain the tool surface state multi-scale joint perception feature map, wherein the joint perception formula is:

[0063] M1=Reshape(F1)

[0064] M2=Reshape(F2)

[0065]

[0066]

[0067] W 21 = Dropout(M 21 )

[0068] W 12 = Dropout(M 12 )

[0069]

[0070]

[0071] F1'=Reshape(M 1′ )

[0072] F2'=Reshape(M 2′ )

[0073] F c =αF1'+βF2'

[0074] Wherein, F1 represents the tool surface state boundary feature map, F2 represents the tool surface state enhanced global feature map, reshape(·) represents feature shape reshaping, M1 represents the tool surface state boundary feature shape reshaping matrix, M2 represents the tool surface state global feature shape reshaping matrix, represents matrix multiplication, (·) T represents the transpose of the matrix, S represents the scale of the tool surface state boundary feature shape reshaping matrix, that is, the width of the tool surface state boundary feature shape reshaping matrix multiplied by the height, and the tool surface state boundary feature shape reshaping matrix and the tool surface state global feature shape reshaping matrix have the same scale, Softmax is a normalized exponential function, M 21represents the boundary-global dependency matrix, M 12 represents the global-boundary dependency matrix, Dropout(·) represents random dropout processing, W 21 represents the pruned boundary-global dependency matrix, W 12 represents the pruned global-boundary dependency matrix, M 1′ Represents the dependency relationship optimization tool surface state boundary feature matrix, M 2′ represents the global feature matrix of the tool surface state optimized by dependency relationship, F1' represents the boundary feature map of the tool surface state optimized, F2' represents the enhanced global feature map of the tool surface state optimized, α and β represent different weight coefficients respectively, F c A multi-scale joint perception feature map representing the surface state of the tool.

[0075] Specifically, the tool integrity analysis unit 325 is used to generate a tool integrity detection result based on the tool surface state multi-scale joint perception feature map. In a specific example of the present application, the tool surface state multi-scale joint perception feature map is input into a classifier-based integrity detection result generation module to obtain the integrity detection result, which is used to indicate whether the tool is complete.

[0076] In a preferred example, inputting the multi-scale joint perception feature map of the tool surface state into a classifier-based integrity detection result generation module to obtain the integrity detection result comprises the following steps:

[0077] Clustering all eigenvalues ​​of the tool surface state multi-scale joint perception feature map based on the L2 distance between the eigenvalues, and arranging the cluster features into a tool surface state multi-scale joint perception cluster vector;

[0078] Determining a clustering ratio value of the number of eigenvalues ​​of the tool surface state multi-scale joint perception clustering vector and the number of eigenvalues ​​of the tool surface state multi-scale joint perception feature map;

[0079] Dividing the binary norm of the tool surface state multi-scale joint perception clustering vector by the binary norm of the tool surface state multi-scale joint perception feature vector obtained by expanding the tool surface state multi-scale joint perception feature map to obtain a tool surface state multi-scale joint perception conflict representation value;

[0080] Divide a first power value of a norm of the tool surface state multi-scale joint perception cluster vector with the cluster ratio value as an exponent by a second power value of a norm of the tool surface state multi-scale joint perception feature vector with the cluster ratio value as an exponent to obtain a tool surface state multi-scale joint perception confrontation representation value;

[0081] For each eigenvalue of the tool surface state multi-scale joint perception clustering vector, multiply it by the inverse of the difference between the tool surface state multi-scale joint perception conflict representation value and the tool surface state multi-scale joint perception confrontation representation value to obtain an optimized eigenvalue of the tool surface state multi-scale joint perception clustering vector;

[0082] For each feature value outside the cluster in the multi-scale joint perception feature map of the tool surface state, multiply it by the reciprocal of the sum of the multi-scale joint perception conflict representation value of the tool surface state and the multi-scale joint perception antagonism representation value of the tool surface state to obtain an optimized out-of-class feature value of the multi-scale joint perception feature map of the tool surface state;

[0083] Combining the optimized eigenvalues ​​of the tool surface state multi-scale joint perception clustering vector and the optimized out-of-class eigenvalues ​​of the tool surface state multi-scale joint perception feature map into an optimized tool surface state multi-scale joint perception feature map; and

[0084] The optimized multi-scale joint perception feature map of the tool surface state is input into the classifier-based integrity detection result generation module to obtain the integrity detection result.

[0085] It is expressed as:

[0086]

[0087] Among them, V n is the multi-scale joint perception feature vector of the tool surface state, n is the number of eigenvalues ​​of the multi-scale joint perception feature vector of the tool surface state, is the multi-scale joint perception clustering vector of the tool surface state, k is the number of eigenvalues ​​of the multi-scale joint perception clustering vector of the tool surface state, represents the clustering feature set corresponding to the multi-scale joint perception clustering vector of the tool surface state, ‖·‖2 and They represent the λ-power of the second norm and the first norm of the vector respectively.

[0088] That is, considering that the tool surface state boundary feature map and the tool surface state enhanced global feature map respectively represent the image shallow edge enhancement features and image global features of the tool surface state grayscale image, when further performing significant joint perception based on the attention mechanism, the difference in shallow-deep fusion attention weights of the image will cause image semantic significant joint alignment conflicts, thereby causing the loss of aggregated key information and affecting the expression effect of the multi-scale joint perception feature map of the tool surface state.

[0089] Therefore, in order to avoid the loss of key suffix semantic information relative to the original feature set as a whole due to aggregation conflict caused by the multi-scale joint perception feature map of the tool surface state being based on aggregated features, the clustering ratio of the number of eigenvalues ​​of the multi-scale joint perception clustering vector of the tool surface state relative to the number of eigenvalues ​​of the multi-scale joint perception feature map of the tool surface state is used as a decision function to perform an adversarial decision on the set absolute representation of the multi-scale joint perception clustering vector of the tool surface state and the norm of the multi-scale joint perception feature vector of the tool surface state, and respectively compare it with the multi-scale joint perception clustering vector of the tool surface state and the tool surface state. The clustering intrinsic conflict representation of the two-norm of the multi-scale joint perception feature vector of the tool surface state is interacted positively and negatively to construct a solid alignment guardrail with the original feature set as a whole based on the aggregated features of the optimized multi-scale joint perception feature map of the tool surface state, thereby achieving the mitigation of the intention of harmful information loss based on the aggregated risk transferability of the optimized multi-scale joint perception feature map of the tool surface state, improving the expression effect of the optimized multi-scale joint perception feature map of the tool surface state, and thus improving the accuracy of the integrity detection result obtained by the integrity detection result generation module based on the classifier input of the multi-scale joint perception feature map of the tool surface state.

[0090] In particular, the expected reminder module 330 and the statistical analysis module 340 are used to automatically remind the expiration inspection based on the detection cycle; and to count the use frequency and use time of the tools and generate statistical reports. It should be understood that tools that have not been inspected and maintained may fail, thereby posing a safety hazard, and automatic reminders for expiration inspections can help ensure that tools are inspected and maintained within the recommended time, so as to simplify the tool management process and improve efficiency. Furthermore, by counting the use frequency and use time of tools, it can help track the use of tools. The management of tools can be further optimized by analyzing the use of tools.

[0091] As described above, the intelligent management system 300 for electronic safety tools according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with an intelligent management algorithm for electronic safety tools. In a possible implementation, the intelligent management system 300 for electronic safety tools according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the intelligent management system 300 for electronic safety tools can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the intelligent management system 300 for electronic safety tools can also be one of the many hardware modules of the wireless terminal.

[0092] Alternatively, in another example, the electronic safety tool intelligent management system 300 and the wireless terminal may also be separate devices, and the electronic safety tool intelligent management system 300 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0093] Furthermore, a method for intelligent management of electronic safety tools is also provided.

[0094] Figure 5 FIG. 1 is a flow chart of an intelligent management method for electronic safety tools according to an embodiment of the present application. Figure 5 As shown, according to the embodiment of the present application, the intelligent management method of electronic safety tools includes the following steps: S1, acquiring an image of the tool; S2, respectively extracting boundary features and global context features of the image of the tool to obtain a tool surface state boundary feature map and a tool surface state global feature map; S3, performing feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map; S4, performing joint perception on the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map; S5, generating a tool integrity detection result based on the tool surface state multi-scale joint perception feature map.

[0095] In summary, the electronic safety tool intelligent management method according to the embodiment of the present application is explained, which automatically takes tool images when tools are in and out of the warehouse, and uses deep learning-based computer vision technology to perform multi-level feature analysis on tool images, respectively capturing the boundary features and global spatial structure features of tools, and then through joint perception and dependency modeling of the two, to obtain multi-level tool status information, thereby realizing intelligent detection of tool integrity and intelligent management of tool in and out of the warehouse. In this way, the efficiency and accuracy of tool management can be significantly improved, human errors can be reduced, and work safety can be ensured.

[0096] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. An intelligent management system for electronic safety tools, characterized in that: include: A tool record management module is used to input tool information and predict the inspection cycle based on the tool type and usage information; An in-and-out management module, used to automatically take images of the tools when they are in and out of the warehouse, and to perform integrity checks based on the images of the tools and archive records; An expected reminder module, used for automatically reminding the due inspection based on the detection cycle; A statistical analysis module, used to count the usage frequency and usage time of the tools and generate statistical reports; Wherein, the inbound and outbound management module includes: A tool image acquisition unit, used to acquire an image of the tool; An image multi-level feature extraction unit, used to extract boundary features and global context features of the image of the tool respectively to obtain a tool surface state boundary feature map and a tool surface state global feature map; A feature autocorrelation enhancement unit, used for performing feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map; A multi-level feature joint perception unit, used for jointly perceiving the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map; The tool integrity analysis unit is used to generate a tool integrity detection result based on the multi-scale joint perception feature map of the tool surface state.

2. The intelligent management system for electronic safety tools according to claim 1 is characterized in that: The image multi-level feature extraction unit comprises: An image grayscale subunit, used for grayscale processing the image of the tool to obtain a grayscale image of the tool surface state; The multi-level feature extraction subunit is used to input the grayscale image of the tool surface state into the MBCNet model including the boundary feature extraction branch and the backbone network to obtain the tool surface state boundary feature map and the tool surface state global feature map.

3. The intelligent management system for electronic safety tools according to claim 2 is characterized in that: The feature autocorrelation enhancement unit is used to: The tool surface state global feature map is input into a self-attention enhanced cross-channel feature space structure preservation module to obtain the tool surface state enhanced global feature map.

4. The intelligent management system for electronic safety tools according to claim 3 is characterized in that: The characteristic autocorrelation strengthening unit comprises: A layer normalization subunit, used for performing layer normalization on the tool surface state global feature map to obtain a normalized tool surface state global feature map; A channel context-related feature extraction subunit is used to perform point convolution processing on the normalized chemical tool surface state global feature map to obtain a tool surface state channel context-related representation feature map; A spatial context-related feature extraction subunit, used for performing convolution encoding on the tool surface state channel context-related representation feature map to obtain a tool surface state spatial context-related representation feature map; The global interactive attention fusion subunit is used to perform channel-space global interactive attention fusion on the tool surface state channel context association representation feature map and the tool surface state space context association representation feature map to obtain the tool surface state enhanced global feature map.

5. The intelligent management system for electronic safety tools according to claim 4 is characterized in that: The global interactive attention fusion subunit is used to: Copying the tool surface state space context association representation feature map to obtain a backup tool surface state space context association representation feature map; Reshaping the tool surface state channel context association representation feature map, the tool surface state space context association representation feature map and the backup tool surface state space context association representation feature map to obtain a tool surface state channel context association representation feature matrix, a tool surface state space context association representation feature matrix and a backup tool surface state space context association representation feature matrix; Calculating a cross-channel cross-covariance matrix between the tool surface state channel context association representation feature matrix and the tool surface state space context association representation feature matrix; Using a Softmax function to activate the cross-channel cross-covariance matrix to obtain a global interactive attention matrix of tool surface state features; Calculate the product of the backup tool surface state space context association representation feature matrix and the tool surface state feature global interactive attention matrix to obtain an attention-enhanced tool surface state feature representation matrix; The feature shape of the surface state feature representation matrix of the attention-enhanced tool is reshaped to obtain the tool surface state enhanced global feature map.

6. The intelligent management system for electronic safety tools according to claim 5 is characterized in that: The multi-level feature joint perception unit is used to: The tool surface state boundary feature map and the tool surface state enhanced global feature map are input into a salient joint perception module based on an attention mechanism to obtain a multi-scale joint perception feature map of the tool surface state.

7. The intelligent management system for electronic safety tools according to claim 6 is characterized in that: The multi-level feature joint perception unit includes: A first characteristic shape reshaping subunit is used to perform characteristic shape reshaping on the tool surface state boundary characteristic map and the tool surface state enhanced global characteristic map to obtain a tool surface state boundary characteristic shape reshaping matrix and a tool surface state global characteristic shape reshaping matrix; A bidirectional dependency calculation subunit, used for inputting the tool surface state boundary feature shape reshaping matrix and the tool surface state global feature shape reshaping matrix into a feature channel-by-channel interactive perception module to obtain a boundary-global dependency matrix and a global-boundary dependency matrix; A random dropout subunit, used for inputting the boundary-global dependency matrix and the global-boundary dependency matrix into a random dropout module to obtain a pruned boundary-global dependency matrix and a pruned global-boundary dependency matrix; A semantic feature dependency optimization subunit, used for performing matrix multiplication of the pruned boundary-global dependency matrix and the tool surface state global feature shape reshaping matrix to obtain a dependency optimized tool surface state global feature matrix; A shallow feature dependency optimization subunit, used for performing matrix multiplication of the pruned global-boundary dependency matrix and the tool surface state boundary feature shape reshaping matrix to obtain a dependency optimized tool surface state boundary feature matrix; The second feature shape reshaping subunit is used to perform feature shape reshaping on the dependency relationship optimized tool surface state boundary feature matrix and the dependency relationship optimized tool surface state global feature matrix to obtain an optimized tool surface state boundary feature map and an optimized tool surface state enhanced global feature map; The multi-scale joint subunit is used to calculate the weighted sum of the boundary feature map of the optimized tool surface state and the enhanced global feature map of the optimized tool surface state to obtain the multi-scale joint perception feature map of the tool surface state.

8. The intelligent management system for electronic safety tools according to claim 7 is characterized in that: The bidirectional dependency calculation subunit is used to: Calculate the tool surface state boundary characteristic shape reshaping matrix multiplied by the tool surface state global characteristic shape reshaping matrix to obtain a tool surface state boundary-global correlation representation matrix; The tool surface state boundary-global correlation representation matrix is ​​divided by the scale of the tool surface state global feature shape reshaping matrix and then input into the softmax function to obtain the boundary-global dependency matrix; Calculate the transposed matrix of the tool surface state boundary feature shape reshaping matrix and multiply it by the tool surface state global feature shape reshaping matrix to obtain a tool surface state global-boundary association representation matrix; The tool surface state boundary-global correlation representation matrix is ​​divided by the scale of the tool surface state boundary feature shape reshaping matrix and then input into the softmax function to obtain the global-boundary dependency matrix.

9. The intelligent management system for electronic safety tools according to claim 8 is characterized in that: The tool integrity analysis unit is used to: The multi-scale joint perception feature map of the tool surface state is input into a classifier-based integrity detection result generation module to obtain the integrity detection result, and the integrity detection result is used to indicate whether the tool is complete.

10. An intelligent management method for electronic safety tools, characterized in that: include: Acquiring an image of the tool; Extracting boundary features and global context features of the image of the tool respectively to obtain a tool surface state boundary feature map and a tool surface state global feature map; Performing feature autocorrelation enhancement on the tool surface state global feature map to obtain a tool surface state enhanced global feature map; Performing joint perception on the tool surface state boundary feature map and the tool surface state enhanced global feature map to obtain a tool surface state multi-scale joint perception feature map; Based on the multi-scale joint perception feature map of the tool surface state, an integrity detection result of the tool is generated.

Images