CN117893477B - Intelligent security check system based on AI image recognition - Google Patents
Intelligent security check system based on AI image recognition Download PDFInfo
- Publication number
- CN117893477B CN117893477B CN202311728402.9A CN202311728402A CN117893477B CN 117893477 B CN117893477 B CN 117893477B CN 202311728402 A CN202311728402 A CN 202311728402A CN 117893477 B CN117893477 B CN 117893477B
- Authority
- CN
- China
- Prior art keywords
- accident
- image
- model
- equipment
- establishing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007689 inspection Methods 0.000 claims abstract description 50
- 238000010586 diagram Methods 0.000 claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims description 39
- 230000004927 fusion Effects 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 15
- 238000013507 mapping Methods 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 11
- 238000002372 labelling Methods 0.000 claims description 6
- 238000012216 screening Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 239000000203 mixture Substances 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 abstract description 5
- 230000000007 visual effect Effects 0.000 abstract description 5
- 238000011179 visual inspection Methods 0.000 abstract description 5
- 230000009286 beneficial effect Effects 0.000 description 10
- 238000003723 Smelting Methods 0.000 description 5
- 239000000428 dust Substances 0.000 description 5
- 238000010801 machine learning Methods 0.000 description 4
- 238000003058 natural language processing Methods 0.000 description 4
- 229910000831 Steel Inorganic materials 0.000 description 3
- 230000000740 bleeding effect Effects 0.000 description 3
- 239000010959 steel Substances 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 238000005266 casting Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/005—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles during manufacturing process
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30232—Surveillance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/06—Recognition of objects for industrial automation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Manufacturing & Machinery (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Alarm Systems (AREA)
Abstract
The invention provides an intelligent security check system based on AI image recognition, comprising: the method comprises the steps of establishing an accident conceptual diagram corresponding to each historical safety accident, establishing an accident image sample corresponding to each accident device to generate an accident image library, counting a plurality of existing devices contained in a preset factory, establishing a three-dimensional factory model of the preset factory, operating the three-dimensional factory model to obtain a current image corresponding to each existing device, carrying out AI identification on the current image by utilizing the accident image library to obtain a safety grade corresponding to each existing device, marking the safety grade corresponding to each existing device in the three-dimensional factory model, establishing safety supervision information of the preset factory, transmitting to an information display module to display, realizing the establishment of a three-dimensional visual model of the heavy point inspection device by utilizing three-dimensional simulation, supporting the assistance of equipment visual inspection at a mobile phone end, and greatly improving the expertise, convenience and standardization of inspection personnel.
Description
Technical Field
The invention relates to the technical field of security inspection systems, in particular to an intelligent security inspection system based on AI image recognition.
Background
The existing security inspection system can only realize on-line management of inspection flow generally, and inspection personnel can plan, output results and store data from the system and call out inspection contents and inspection items on site. However, in the safety inspection scene, the field process equipment is complex and various, the inspector does not know the 'important inspection equipment' and the 'problem hidden danger' described in the inspection table at all because of the limitation of the professional property, the problem can not be naturally found out, the field problem hidden danger can not be rectified, the problem hidden danger is stored for a long time, and the accident occurrence probability is greatly increased. An intelligent checking tool is urgently needed, and can be used for looking at checking key points, judging problem hidden dangers and associating checking basis and standard like an expert, so that the checking professionals are improved, and the checking effect is ensured.
Therefore, the invention provides an intelligent security check system based on AI image recognition.
Disclosure of Invention
According to the intelligent security inspection system based on AI image recognition, the three-dimensional visual model establishment of the heavy point inspection equipment is realized by utilizing three-dimensional simulation, the equipment visual inspection assistance at the mobile phone end is supported, and the expertise, convenience and standardization of inspection personnel are greatly improved.
The invention provides an intelligent security check system based on AI image recognition, comprising:
the map library establishing module is used for establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
The model building module is used for counting a plurality of existing devices contained in a preset factory and building a three-dimensional factory model of the preset factory according to the image information of each existing device and combining the plan of the preset factory;
the checking execution module is used for operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
The accident analysis module is used for marking the safety level corresponding to each existing device in the three-dimensional factory model, establishing the safety supervision information of the preset factory, and transmitting the safety supervision information to the information display module for display.
In one embodiment of the present invention, in one possible implementation,
Further comprises:
the image acquisition module is used for acquiring accident data corresponding to the historical safety accidents, drawing accident conceptual diagrams corresponding to each historical safety accident according to the accident data by using an AI composition technology, and storing the accident conceptual diagrams in a classified mode according to the accident grade corresponding to each historical safety accident.
In one embodiment of the present invention, in one possible implementation,
Further comprises:
The information calling module is used for determining the time range searched by the user according to a calling instruction issued by the user, acquiring target safety supervision information in the time range of the preset factory, and transmitting the safety supervision information to the information display module for display.
In one embodiment of the present invention, in one possible implementation,
The information display module is used for displaying the safety supervision information;
the information display module is also used for displaying the target safety supervision information.
In one embodiment of the present invention, in one possible implementation,
The gallery creation module comprises:
The concept presentation unit is used for respectively carrying out equipment screening and equipment labeling on each accident concept graph to obtain a plurality of concept equipment contained in the accident concept graph, acquiring presentation images corresponding to the same concept equipment in different accident concept graphs, and establishing a presentation image set corresponding to each concept equipment;
The identification execution unit is used for acquiring equipment parameters corresponding to each concept equipment, establishing an equipment normal diagram corresponding to each concept equipment according to the equipment parameters, and establishing an identification sample corresponding to each concept equipment based on the equipment normal diagram;
the grade identification unit is used for acquiring the accident grade corresponding to each accident conceptual diagram, and carrying out key point identification on the corresponding presentation image set by utilizing the identification sample to obtain the accident key points corresponding to the conceptual equipment under different accident grades;
And the drawing library establishing unit is used for counting different accident key points of each concept device, respectively marking each accident key point on the normal drawing of the device to obtain corresponding accident image samples of the concept device under different accident levels, and establishing an accident image library according to the accident image samples.
In one embodiment of the present invention, in one possible implementation,
The model building module comprises:
The device modeling unit is used for acquiring image information corresponding to each existing device in the preset factory, and respectively inputting each image information into preset 3DSMAX software to perform device modeling to obtain a device model corresponding to each existing device;
The hierarchical fusion unit is used for constructing a scene model according to the plan view of the preset factory, splitting the scene model to obtain a plurality of model layers, and respectively inputting each equipment model into different model layers to perform model mapping processing to obtain the fusion degree between each equipment model and different model layers;
The mapping analysis unit is used for determining mapping positions of each equipment model in the scene model according to the fusion degree, and pasting the equipment models in corresponding model layers in the scene model to generate a live-action model;
The model generation unit is used for dividing the real model into a plurality of real areas, carrying out model baking treatment on each real area to obtain light and shadow information corresponding to each real area, projecting the light and shadow information into the real model, and generating a three-dimensional factory model.
In one embodiment of the present invention, in one possible implementation,
The inspection execution module includes:
the image construction unit is used for operating the three-dimensional factory model, respectively carrying out 3D digital processing on the three-dimensional factory model in the operation process, and obtaining corresponding current images of each existing device in different preset time periods according to the processing results;
The abnormal point determining unit is used for counting a plurality of current images corresponding to the same existing device, sequencing the current images according to the sequence of a preset time period corresponding to each current image, establishing an image sequence corresponding to each existing device, and respectively carrying out image superposition comparison analysis on the current images in each image sequence to obtain image abnormal points between two adjacent current images in each image sequence;
The identification analysis unit is used for drawing an abnormal image corresponding to the existing equipment according to the image outlier corresponding to the same image sequence, carrying out AI identification on the abnormal image by utilizing the accident image library to obtain the coincidence degree between the abnormal image and each accident image sample, and judging whether the maximum coincidence degree between the abnormal image and each accident image sample is larger than a preset and coincidence degree threshold value;
And the grade determining unit is used for acquiring a plurality of selected accident samples larger than the preset coincidence degree threshold and the accident grade corresponding to each selected accident sample when the maximum coincidence degree between the abnormal image and each accident image sample is larger than the preset coincidence degree threshold, establishing a selected weight for the corresponding selected accident sample according to the selected coincidence degree corresponding to each selected accident sample, and adjusting the accident grade of the corresponding selected accident sample based on the selected weight to generate the safety grade of the existing equipment.
In one embodiment of the present invention, in one possible implementation,
The grade determining unit is further configured to:
and when the maximum coincidence degree between the abnormal image and each accident image is smaller than a preset coincidence degree threshold value, establishing a normal security level for the existing equipment corresponding to the abnormal image.
In one embodiment of the present invention, in one possible implementation,
The accident analysis module comprises:
The first analysis unit is used for marking the security level corresponding to each existing device in the three-dimensional factory model, dividing the three-dimensional factory model into a plurality of three-dimensional factory areas, and establishing the area security level corresponding to each three-dimensional factory area according to marking results;
The second analysis unit is used for establishing regional supervision information for the corresponding three-dimensional factory region according to the regional security level and establishing equipment supervision information for the corresponding existing equipment according to the security level;
And the third analysis unit is used for counting the regional supervision information and the equipment supervision information contained in the three-dimensional factory model, establishing the safety supervision information of the preset factory according to the regional supervision information and the equipment supervision information, and transmitting the safety supervision information to the display model for display.
The invention relates to an intelligent security inspection method based on AI image recognition, which comprises the following steps:
Step 1: establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
Step 2: counting a plurality of existing devices contained in a preset factory, and establishing a three-dimensional factory model of the preset factory according to the image information of each existing device and combining with a plan view of the preset factory;
Step 3: operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
Step 4: and marking the safety level corresponding to each existing device in the three-dimensional factory model, establishing the safety supervision information of the preset factory, and transmitting the safety supervision information to an information display module for display.
The invention has the beneficial effects that: in order to carry out safety inspection on equipment in a preset factory, an accident conceptual diagram is built according to historical safety accidents before the safety inspection is formally entered, then an accident image sample is built according to the accident conceptual diagram to generate an accident image library, a three-dimensional factory model of the preset factory is further built, AI identification is carried out on the three-dimensional factory model by utilizing the accident image library, further the safety grade of each existing equipment pair in the preset factory is determined, the safety grade of each existing equipment is further marked in the three-dimensional factory model, safety supervision information is built, a user obtains the actual situation of the preset factory by checking the safety supervision information, the system can realize intelligent identification on a reaction kettle, a storage tank, a vacuum pump, a bipyramid, a gas cylinder, a centrifuge, a major linkage source, a condenser, a reboiler, a pipeline pipe fitting, dust removing equipment, a steel wire rope lifting device, a deep well pouring system, a flameless bleeding port, a fixed smelting furnace and other important inspection equipment and a solution with obvious hidden danger of appearance characteristic problems, through laboratory and field verification, natural language processing, machine learning model prediction, word vector and other technologies are reasonably adopted, two branches are mutually supported, accuracy is improved, a single-word recognition accuracy is realized, and a special purpose recognition technology vector is not realized; the three-dimensional visual model of the heavy point inspection equipment is built through three-dimensional simulation, equipment visual inspection assistance is supported at the mobile phone end, and the expertise, convenience and standardization of inspection staff are greatly improved.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of an intelligent security inspection system based on AI image recognition in an embodiment of the invention;
fig. 2 is a schematic diagram illustrating the composition of an analysis module of an intelligent security inspection system based on AI image recognition in an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Example 1
The present embodiment provides an intelligent security inspection system based on AI image recognition, as shown in fig. 1, including:
the map library establishing module is used for establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
The model building module is used for counting a plurality of existing devices contained in a preset factory and building a three-dimensional factory model of the preset factory according to the image information of each existing device and combining the plan of the preset factory;
the checking execution module is used for operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
The accident analysis module is used for marking the safety level corresponding to each existing device in the three-dimensional factory model, establishing the safety supervision information of the preset factory, and transmitting the safety supervision information to the information display module for display.
In this example, the historical security incidents represent security incidents that have occurred;
in this example, the accident device represents a device that has failed or has been in question when a security incident occurs;
in this example, the incident image samples represent samples built from the appearance that the incident device presents when different faults or different problems occur;
In this example, the existing equipment represents equipment currently in a preset factory, and the existing equipment may be a reaction vessel, a storage tank, a vacuum pump, a bipyramid, a gas cylinder, a centrifuge, a heavy chain source, a condenser, a reboiler, a pipe fitting, dust removal equipment, a wire rope lifting device, a deep well casting system, a flameless vent, a fixed smelting furnace, or the like;
in this example, the three-dimensional plant model represents a model of a preset plant drawn in a three-dimensional space;
in this example, the accident concept graph represents an image used to describe a historical security accident scene;
in this example, the current image represents an image presented by the existing device at the current time;
In this example, the security level represents a level established according to the probability that the existing device is malfunctioning or problematic in the current state;
In this example, the safety supervision information includes a safety level corresponding to each area in the preset factory and a safety level corresponding to each existing device;
In this example, the information display terminal may be a mobile phone terminal of the user.
The working principle of the technical scheme has the beneficial effects that: in order to carry out safety inspection on equipment in a preset factory, an accident conceptual diagram is built according to historical safety accidents before the safety inspection is formally entered, then an accident image sample is built according to the accident conceptual diagram to generate an accident image library, a three-dimensional factory model of the preset factory is further built, AI identification is carried out on the three-dimensional factory model by utilizing the accident image library, further the safety grade of each existing equipment pair in the preset factory is determined, the safety grade of each existing equipment is further marked in the three-dimensional factory model, safety supervision information is built, a user obtains the actual situation of the preset factory by checking the safety supervision information, the system can realize intelligent identification on a reaction kettle, a storage tank, a vacuum pump, a bipyramid, a gas cylinder, a centrifuge, a major linkage source, a condenser, a reboiler, a pipeline pipe fitting, dust removing equipment, a steel wire rope lifting device, a deep well pouring system, a flameless bleeding port, a fixed smelting furnace and other important inspection equipment and a solution with obvious hidden danger of appearance characteristic problems, through laboratory and field verification, natural language processing, machine learning model prediction, word vector and other technologies are reasonably adopted, two branches are mutually supported, accuracy is improved, a single-word recognition accuracy is realized, and a special purpose recognition technology vector is not realized; the three-dimensional visual model of the heavy point inspection equipment is built through three-dimensional simulation, equipment visual inspection assistance is supported at the mobile phone end, and the expertise, convenience and standardization of inspection staff are greatly improved.
Example 2
On the basis of embodiment 1, the intelligent security inspection system based on AI image recognition further comprises:
the image acquisition module is used for acquiring accident data corresponding to the historical safety accidents, drawing accident conceptual diagrams corresponding to each historical safety accident according to the accident data by using an AI composition technology, and storing the accident conceptual diagrams in a classified mode according to the accident grade corresponding to each historical safety accident.
In this example, the accident data may be text data, voice data, image data, video data, etc.;
in the example, accident data is processed by utilizing natural language processing and machine learning model prediction technology;
In this example, the role of the taxonomy store is to categorize incident conceptual diagrams with consistent incident levels into one category.
The working principle of the technical scheme has the beneficial effects that: in order to establish an accident image library with high practicability and wide coverage content, an accident conceptual diagram of the historical safety accident is drawn according to accident data corresponding to the historical safety accident, and the conceptual diagram is classified and stored, so that the subsequent searching work is facilitated.
Example 3
On the basis of embodiment 1, the intelligent security inspection system based on AI image recognition further comprises:
The information calling module is used for determining the time range searched by the user according to a calling instruction issued by the user, acquiring target safety supervision information in the time range of the preset factory, and transmitting the safety supervision information to the information display module for display.
The working principle of the technical scheme has the beneficial effects that: in order to facilitate users to view safety supervision information in different time ranges at any time, an information calling module is arranged for users to search and view at any time.
Example 4
Based on embodiment 3, the intelligent security check system based on AI image recognition:
The information display module is used for displaying the safety supervision information;
the information display module is also used for displaying the target safety supervision information.
Example 5
Based on embodiment 1, the intelligent security inspection system based on AI image recognition, the gallery creation module includes:
The concept presentation unit is used for respectively carrying out equipment screening and equipment labeling on each accident concept graph to obtain a plurality of concept equipment contained in the accident concept graph, acquiring presentation images corresponding to the same concept equipment in different accident concept graphs, and establishing a presentation image set corresponding to each concept equipment;
The identification execution unit is used for acquiring equipment parameters corresponding to each concept equipment, establishing an equipment normal diagram corresponding to each concept equipment according to the equipment parameters, and establishing an identification sample corresponding to each concept equipment based on the equipment normal diagram;
the grade identification unit is used for acquiring the accident grade corresponding to each accident conceptual diagram, and carrying out key point identification on the corresponding presentation image set by utilizing the identification sample to obtain the accident key points corresponding to the conceptual equipment under different accident grades;
And the drawing library establishing unit is used for counting different accident key points of each concept device, respectively marking each accident key point on the normal drawing of the device to obtain corresponding accident image samples of the concept device under different accident levels, and establishing an accident image library according to the accident image samples.
In this example, device screening means that devices included in the incident conceptual diagram are screened out;
In this example, the device labels represent the labeling out of the devices contained in the incident conceptual diagram;
In this example, the concept devices represent devices contained in the accident concept image;
in this example, the presented image represents the appearance that the concept device presented in the incident concept graph;
in the example, the presented image set represents an image set obtained by counting presented images corresponding to the same conceptual equipment in different accident conceptual diagrams;
In this example, the device parameters represent basic parameters of the conceptual device;
in this example, the device normal chart represents a conceptual chart drawn according to the normal operation state of the conceptual device;
in this example, the identification sample represents a sample used to identify different concept devices;
in this example, one accident concept map corresponds to one accident level;
In this example, the key point recognition represents a process of judging the recognition sample corresponding to the presentation image by analyzing the key point between the recognition sample and the presentation image;
in this example, the incident keypoints represent keypoints that the concept device presents at one incident level.
The working principle of the technical scheme has the beneficial effects that: in order to establish an effective and real-time updated accident image library, firstly, device screening and device labeling are carried out in an accident concept graph, a plurality of concept devices contained in the accident concept graph are determined, further, presentation images corresponding to different concept devices in different accident concept graphs are determined, so that a presentation image set is established, then, a device normal graph of the concept devices is established according to device parameters corresponding to the concept devices, an identification sample is established for the concept devices according to the device normal graph, the accident level corresponding to each accident concept graph is combined, key point identification is carried out on the presentation image set by utilizing the identification sample, so that accident key points corresponding to the concept devices under different accident levels are obtained, accident image samples of each concept device are established according to the accident key points and the accident level one by one, finally, an effective accident image library is generated, the accident image set is established by analyzing the accident key points corresponding to each concept device under different accident levels, not only can be generated, but also the accuracy of the image set is improved, and the contrast quality is improved.
Example 6
On the basis of embodiment 1, the intelligent security inspection system based on AI image recognition, the model building module includes:
The device modeling unit is used for acquiring image information corresponding to each existing device in the preset factory, and respectively inputting each image information into preset 3DSMAX software to perform device modeling to obtain a device model corresponding to each existing device;
The hierarchical fusion unit is used for constructing a scene model according to the plan view of the preset factory, splitting the scene model to obtain a plurality of model layers, and respectively inputting each equipment model into different model layers to perform model mapping processing to obtain the fusion degree between each equipment model and different model layers;
The mapping analysis unit is used for determining mapping positions of each equipment model in the scene model according to the fusion degree, and pasting the equipment models in corresponding model layers in the scene model to generate a live-action model;
The model generation unit is used for dividing the real model into a plurality of real areas, carrying out model baking treatment on each real area to obtain light and shadow information corresponding to each real area, projecting the light and shadow information into the real model, and generating a three-dimensional factory model.
In this example, the image information represents information corresponding to the case of representing the existing device in an image manner;
In this example, the preset 3D max software represents software for building a 3D model;
In this example, the device model represents the existing device expressed by way of a model;
In this example, the splitting process represents the process of splitting a scene model into several model layers that can be used to place a device model;
In this example, the mapping process represents a process of pasting the device model in the model layer;
In this example, the fusion degree represents the fusion degree between the equipment model and the model layer, and the fusion degree between different equipment models and different model layers may be the same or different;
in this example, the live-action region represents one region in the live-action model;
In this example, the model baking process represents a process of adding a light shadow in each live-action area separately;
In this example, the shadow information represents light shading information at different locations in the field area.
The working principle of the technical scheme has the beneficial effects that: modeling the image information of each existing device in a preset factory by using preset 3DSMAX software to generate a device model corresponding to the existing device, then constructing a scene model, splitting the scene model to obtain a plurality of model layers, inputting the device model into different model layers, determining the mapping position of each device model in the scene model by analyzing the fusion degree in the different model layers, generating a real model, finally performing light and shadow processing on the real model, rendering the real model, and finally establishing a three-dimensional factory model, wherein the model which is fit with the reality can be established through the modeling, so that the follow-up fault inspection and problem analysis are facilitated.
Example 7
On the basis of embodiment 1, the intelligent security inspection system based on AI image recognition, the inspection execution module includes:
the image construction unit is used for operating the three-dimensional factory model, respectively carrying out 3D digital processing on the three-dimensional factory model in the operation process, and obtaining corresponding current images of each existing device in different preset time periods according to the processing results;
The abnormal point determining unit is used for counting a plurality of current images corresponding to the same existing device, sequencing the current images according to the sequence of a preset time period corresponding to each current image, establishing an image sequence corresponding to each existing device, and respectively carrying out image superposition comparison analysis on the current images in each image sequence to obtain image abnormal points between two adjacent current images in each image sequence;
The identification analysis unit is used for drawing an abnormal image corresponding to the existing equipment according to the image outlier corresponding to the same image sequence, carrying out AI identification on the abnormal image by utilizing the accident image library to obtain the coincidence degree between the abnormal image and each accident image sample, and judging whether the maximum coincidence degree between the abnormal image and each accident image sample is larger than a preset and coincidence degree threshold value;
And the grade determining unit is used for acquiring a plurality of selected accident samples larger than the preset coincidence degree threshold and the accident grade corresponding to each selected accident sample when the maximum coincidence degree between the abnormal image and each accident image sample is larger than the preset coincidence degree threshold, establishing a selected weight for the corresponding selected accident sample according to the selected coincidence degree corresponding to each selected accident sample, and adjusting the accident grade of the corresponding selected accident sample based on the selected weight to generate the safety grade of the existing equipment.
In this example, the 3D digital processing represents the process of converting corresponding factory images at different time points in a three-dimensional factory model into digital images using a three-dimensional tool;
In this example, the current image represents an image corresponding to an existing device in a different time period;
In this example, the preset time period may be 20 minutes;
In this example, the image sequence represents a sequence generated by sorting the current images corresponding to the existing device in time order;
in this example, the image superimposition comparison analysis indicates an analysis process of superimposing a current image in one image sequence, and analyzing a different point between two current images after the superimposition;
In this example, the image outlier represents a different point between two adjacent current images;
In this example, the abnormal image represents an abnormal image generated after drawing a plurality of image outliers in an image sequence;
in this example, the coincidence degree represents the coincidence degree between the abnormal image and the accident image;
In this example, the preset overlap threshold represents 80%.
The working principle of the technical scheme has the beneficial effects that: in order to further inspect a preset factory, a three-dimensional factory model is firstly operated in the inspection process, a 3D digital processing technology is utilized to process the three-dimensional factory model in the operation process, so that current images of existing equipment in different preset time periods are obtained, then current images corresponding to the same existing equipment are sequenced according to time sequence to generate an image sequence, at the moment, the produced image sequence is actually used for integrating and sequencing the current images of the existing equipment, image loss is effectively avoided, then two adjacent current images in the same image sequence are subjected to image superposition comparison analysis, image outliers between the two adjacent current images are obtained, further, abnormal images of the existing equipment can be established according to the image outliers, AI identification is conducted on the abnormal images through an accident image library, the accident level of the existing equipment is determined through analyzing the coincidence degree between the abnormal images and an accident image sample, and therefore the safety level of the existing equipment is determined, and the all-dimensional multi-angle inspection work is realized.
Example 8
On the basis of embodiment 7, the intelligent security inspection system based on AI image recognition, the grade determining unit is further configured to:
and when the maximum coincidence degree between the abnormal image and each accident image is smaller than a preset coincidence degree threshold value, establishing a normal security level for the existing equipment corresponding to the abnormal image.
The working principle of the technical scheme has the beneficial effects that: when the coincidence ratio between the abnormal image and the accident image is too low, the existing equipment is indicated to have a short fault and does not need to be processed, so that the productivity is at a normal safety level.
Example 9
On the basis of embodiment 1, the intelligent security inspection system based on AI image recognition, the accident analysis module includes:
The first analysis unit is used for marking the security level corresponding to each existing device in the three-dimensional factory model, dividing the three-dimensional factory model into a plurality of three-dimensional factory areas, and establishing the area security level corresponding to each three-dimensional factory area according to marking results;
The second analysis unit is used for establishing regional supervision information for the corresponding three-dimensional factory region according to the regional security level and establishing equipment supervision information for the corresponding existing equipment according to the security level;
And the third analysis unit is used for counting the regional supervision information and the equipment supervision information contained in the three-dimensional factory model, establishing the safety supervision information of the preset factory according to the regional supervision information and the equipment supervision information, and transmitting the safety supervision information to the display model for display.
In this example, the three-dimensional factory area represents the result of dividing the three-dimensional factory model into a plurality of areas of prescribed specification size;
In this example, the regional security level represents a security level of the region generated from security levels of existing equipment contained in the three-dimensional factory region;
In this example, the area supervision information represents information describing various security levels of the area;
in this example, the device supervision information represents information describing the security level of the existing device.
The working principle of the technical scheme has the beneficial effects that: the safety levels of the areas of different factories and the safety levels of the existing equipment are analyzed to establish equipment supervision information and area supervision information by marking the safety levels in the three-dimensional working model, and finally the safety supervision information of the preset factories is established according to the equipment supervision information and the area supervision information, so that the purpose of supervision is realized.
Example 10
The embodiment provides an intelligent security inspection method based on AI image recognition, which is characterized by comprising the following steps:
Step 1: establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
Step 2: counting a plurality of existing devices contained in a preset factory, and establishing a three-dimensional factory model of the preset factory according to the image information of each existing device and combining with a plan view of the preset factory;
Step 3: operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
Step 4: and marking the safety level corresponding to each existing device in the three-dimensional factory model, establishing the safety supervision information of the preset factory, and transmitting the safety supervision information to an information display module for display.
In this example, the historical security incidents represent security incidents that have occurred;
in this example, the accident device represents a device that has failed or has been in question when a security incident occurs;
in this example, the incident image samples represent samples built from the appearance that the incident device presents when different faults or different problems occur;
In this example, the existing equipment represents equipment currently in a preset factory, and the existing equipment may be a reaction vessel, a storage tank, a vacuum pump, a bipyramid, a gas cylinder, a centrifuge, a heavy chain source, a condenser, a reboiler, a pipe fitting, dust removal equipment, a wire rope lifting device, a deep well casting system, a flameless vent, a fixed smelting furnace, or the like;
in this example, the three-dimensional plant model represents a model of a preset plant drawn in a three-dimensional space;
in this example, the accident concept graph represents an image used to describe a historical security accident scene;
in this example, the current image represents an image presented by the existing device at the current time;
In this example, the security level represents a level established according to the probability that the existing device is malfunctioning or problematic in the current state;
In this example, the safety supervision information includes a safety level corresponding to each area in the preset factory and a safety level corresponding to each existing device;
In this example, the information display terminal may be a mobile phone terminal of the user.
The working principle of the technical scheme has the beneficial effects that: in order to carry out safety inspection on equipment in a preset factory, an accident conceptual diagram is built according to historical safety accidents before the safety inspection is formally entered, then an accident image sample is built according to the accident conceptual diagram to generate an accident image library, a three-dimensional factory model of the preset factory is further built, AI identification is carried out on the three-dimensional factory model by utilizing the accident image library, further the safety grade of each existing equipment pair in the preset factory is determined, the safety grade of each existing equipment is further marked in the three-dimensional factory model, safety supervision information is built, a user obtains the actual situation of the preset factory by checking the safety supervision information, the system can realize intelligent identification on a reaction kettle, a storage tank, a vacuum pump, a bipyramid, a gas cylinder, a centrifuge, a major linkage source, a condenser, a reboiler, a pipeline pipe fitting, dust removing equipment, a steel wire rope lifting device, a deep well pouring system, a flameless bleeding port, a fixed smelting furnace and other important inspection equipment and a solution with obvious hidden danger of appearance characteristic problems, through laboratory and field verification, natural language processing, machine learning model prediction, word vector and other technologies are reasonably adopted, two branches are mutually supported, accuracy is improved, a single-word recognition accuracy is realized, and a special purpose recognition technology vector is not realized; the three-dimensional visual model of the heavy point inspection equipment is built through three-dimensional simulation, equipment visual inspection assistance is supported at the mobile phone end, and the expertise, convenience and standardization of inspection staff are greatly improved.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (7)
1. An intelligent security check system based on AI image recognition, comprising:
the map library establishing module is used for establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
The model building module is used for counting a plurality of existing devices contained in a preset factory and building a three-dimensional factory model of the preset factory according to the image information of each existing device and combining the plan of the preset factory;
the checking execution module is used for operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
The accident analysis module is used for marking the safety level corresponding to each existing device in the three-dimensional factory model, establishing the safety supervision information of the preset factory, and transmitting the safety supervision information to the information display module for display;
The gallery creation module comprises:
The concept presentation unit is used for respectively carrying out equipment screening and equipment labeling on each accident concept graph to obtain a plurality of concept equipment contained in the accident concept graph, acquiring presentation images corresponding to the same concept equipment in different accident concept graphs, and establishing a presentation image set corresponding to each concept equipment;
The identification execution unit is used for acquiring equipment parameters corresponding to each concept equipment, establishing an equipment normal diagram corresponding to each concept equipment according to the equipment parameters, and establishing an identification sample corresponding to each concept equipment based on the equipment normal diagram;
the grade identification unit is used for acquiring the accident grade corresponding to each accident conceptual diagram, and carrying out key point identification on the corresponding presentation image set by utilizing the identification sample to obtain the accident key points corresponding to the conceptual equipment under different accident grades;
The drawing library establishing unit is used for counting different accident key points of each concept device, respectively marking each accident key point on the normal drawing of the device to obtain corresponding accident image samples of the concept device under different accident levels, and establishing an accident image library according to the accident image samples;
the model building module comprises:
The device modeling unit is used for acquiring image information corresponding to each existing device in the preset factory, and respectively inputting each image information into preset 3DSMAX software to perform device modeling to obtain a device model corresponding to each existing device;
The hierarchical fusion unit is used for constructing a scene model according to the plan view of the preset factory, splitting the scene model to obtain a plurality of model layers, and respectively inputting each equipment model into different model layers to perform model mapping processing to obtain the fusion degree between each equipment model and different model layers;
The mapping analysis unit is used for determining mapping positions of each equipment model in the scene model according to the fusion degree, and pasting the equipment models in corresponding model layers in the scene model to generate a live-action model;
The model generation unit is used for dividing the real model into a plurality of real areas, carrying out model baking treatment on each real area to obtain light and shadow information corresponding to each real area, projecting the light and shadow information into the real model, and generating a three-dimensional factory model;
The inspection execution module includes:
the image construction unit is used for operating the three-dimensional factory model, respectively carrying out 3D digital processing on the three-dimensional factory model in the operation process, and obtaining corresponding current images of each existing device in different preset time periods according to the processing results;
The abnormal point determining unit is used for counting a plurality of current images corresponding to the same existing device, sequencing the current images according to the sequence of a preset time period corresponding to each current image, establishing an image sequence corresponding to each existing device, and respectively carrying out image superposition comparison analysis on the current images in each image sequence to obtain image abnormal points between two adjacent current images in each image sequence;
The identification analysis unit is used for drawing an abnormal image corresponding to the existing equipment according to the image outlier corresponding to the same image sequence, carrying out AI identification on the abnormal image by utilizing the accident image library to obtain the coincidence degree between the abnormal image and each accident image sample, and judging whether the maximum coincidence degree between the abnormal image and each accident image sample is larger than a preset and coincidence degree threshold value;
And the grade determining unit is used for acquiring a plurality of selected accident samples larger than the preset coincidence degree threshold and the accident grade corresponding to each selected accident sample when the maximum coincidence degree between the abnormal image and each accident image sample is larger than the preset coincidence degree threshold, establishing a selected weight for the corresponding selected accident sample according to the selected coincidence degree corresponding to each selected accident sample, and adjusting the accident grade of the corresponding selected accident sample based on the selected weight to generate the safety grade of the existing equipment.
2. An AI-image-recognition-based intelligent security check system as recited in claim 1, further comprising:
the image acquisition module is used for acquiring accident data corresponding to the historical safety accidents, drawing accident conceptual diagrams corresponding to each historical safety accident according to the accident data by using an AI composition technology, and storing the accident conceptual diagrams in a classified mode according to the accident grade corresponding to each historical safety accident.
3. An AI-image-recognition-based intelligent security check system as recited in claim 1, further comprising:
The information calling module is used for determining the time range searched by the user according to a calling instruction issued by the user, acquiring target safety supervision information in the time range of the preset factory, and transmitting the safety supervision information to the information display module for display.
4. An AI-image-recognition-based intelligent security check system as recited in claim 3, wherein:
The information display module is used for displaying the safety supervision information;
the information display module is also used for displaying the target safety supervision information.
5. The AI-image-recognition-based intelligent security check system of claim 1, wherein the rank determination unit is further configured to:
and when the maximum coincidence degree between the abnormal image and each accident image is smaller than a preset coincidence degree threshold value, establishing a normal security level for the existing equipment corresponding to the abnormal image.
6. The intelligent security inspection system based on AI image recognition of claim 1, wherein the incident analysis module comprises:
The first analysis unit is used for marking the security level corresponding to each existing device in the three-dimensional factory model, dividing the three-dimensional factory model into a plurality of three-dimensional factory areas, and establishing the area security level corresponding to each three-dimensional factory area according to marking results;
The second analysis unit is used for establishing regional supervision information for the corresponding three-dimensional factory region according to the regional security level and establishing equipment supervision information for the corresponding existing equipment according to the security level;
And the third analysis unit is used for counting the regional supervision information and the equipment supervision information contained in the three-dimensional factory model, establishing the safety supervision information of the preset factory according to the regional supervision information and the equipment supervision information, and transmitting the safety supervision information to the information display model for display.
7. An intelligent security inspection method based on AI image recognition is characterized by comprising the following steps:
Step 1: establishing an accident conceptual diagram corresponding to each historical security accident, establishing an accident image sample corresponding to each accident device according to the accident conceptual diagram, and establishing an accident image library according to the accident image samples;
Step 2: counting a plurality of existing devices contained in a preset factory, and establishing a three-dimensional factory model of the preset factory according to the image information of each existing device and combining with a plan view of the preset factory;
Step 3: operating the three-dimensional factory model to obtain a current image corresponding to each existing device, and carrying out AI identification on the current image by utilizing the accident image library to obtain a security level corresponding to each existing device;
Step 4: marking the corresponding security level of each existing device in the three-dimensional factory model, establishing the security supervision information of the preset factory, and transmitting the security supervision information to an information display module for display;
The step 1 comprises the following steps:
Respectively carrying out equipment screening and equipment labeling on each accident concept graph to obtain a plurality of concept equipment contained in the accident concept graph, obtaining presentation images corresponding to the same concept equipment in different accident concept graphs, and establishing a presentation image set corresponding to each concept equipment;
Acquiring equipment parameters corresponding to each concept equipment, establishing an equipment normal diagram corresponding to each concept equipment according to the equipment parameters, and establishing an identification sample corresponding to each concept equipment based on the equipment normal diagram;
Obtaining the accident grade corresponding to each accident concept graph, and carrying out key point identification on the corresponding presentation image set by utilizing the identification sample to obtain the accident key points corresponding to the concept equipment under different accident grades;
Counting different accident key points of each concept device, respectively marking each accident key point on the device normal graph to obtain corresponding accident image samples of the concept device under different accident levels, and establishing an accident image library according to the accident image samples;
the step 2 includes:
Acquiring image information corresponding to each existing device in the preset factory, and respectively inputting each image information into preset 3DSMAX software to perform device modeling to obtain a device model corresponding to each existing device;
constructing a scene model according to a plan view of the preset factory, splitting the scene model to obtain a plurality of model layers, and respectively inputting each equipment model into different model layers to perform model mapping processing to obtain the fusion degree between each equipment model and different model layers;
Determining the mapping position of each equipment model in the scene model according to the fusion degree, and attaching the equipment model to a corresponding model layer in the scene model to generate a live-action model;
Dividing the live-action model into a plurality of live-action areas, carrying out model baking treatment on each live-action area to obtain light and shadow information corresponding to each live-action area, projecting the light and shadow information into the live-action model, and generating a three-dimensional factory model;
the step 3 includes:
Operating the three-dimensional factory model, respectively performing 3D digital processing on the three-dimensional factory model in the operation process, and obtaining corresponding current images of each existing device in different preset time periods according to processing results;
Counting a plurality of current images corresponding to the same existing equipment, sequencing the current images according to the sequence of a preset time period corresponding to each current image, establishing an image sequence corresponding to each existing equipment, and respectively carrying out image superposition comparison analysis on the current images in each image sequence to obtain image outliers between two adjacent current images in each image sequence;
Drawing an abnormal image corresponding to the existing equipment according to the image outlier corresponding to the same image sequence, carrying out AI identification on the abnormal image by utilizing the accident image library to obtain the coincidence degree between the abnormal image and each accident image sample, and judging whether the maximum coincidence degree between the abnormal image and each accident image sample is larger than a preset and coincidence degree threshold value;
When the maximum coincidence degree between the abnormal image and each accident image sample is larger than a preset coincidence degree threshold value, a plurality of selected accident samples larger than the preset coincidence degree threshold value and accident grades corresponding to each selected accident sample are obtained, a selected weight is established for the corresponding selected accident sample according to the selected coincidence degree corresponding to each selected accident sample, the accident grade of the corresponding selected accident sample is adjusted based on the selected weight, and the safety grade of the existing equipment is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311728402.9A CN117893477B (en) | 2023-12-15 | 2023-12-15 | Intelligent security check system based on AI image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311728402.9A CN117893477B (en) | 2023-12-15 | 2023-12-15 | Intelligent security check system based on AI image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117893477A CN117893477A (en) | 2024-04-16 |
| CN117893477B true CN117893477B (en) | 2024-07-02 |
Family
ID=90651587
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311728402.9A Active CN117893477B (en) | 2023-12-15 | 2023-12-15 | Intelligent security check system based on AI image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117893477B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258432A (en) * | 2013-04-19 | 2013-08-21 | 西安交通大学 | Traffic accident automatic identification processing method and system based on videos |
| CN111091609A (en) * | 2019-12-11 | 2020-05-01 | 云南电网有限责任公司保山供电局 | Transformer substation field operation management and control system and method based on three-dimensional dynamic modeling |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509486A (en) * | 2018-02-08 | 2018-09-07 | 浙江大学 | A kind of safe big data structural management method of intelligent plant multi-source |
| KR20190115384A (en) * | 2018-04-02 | 2019-10-11 | (주)싱크워터 | Integrated registration and display system based on underground space and ground characteristics on electronic map based on intelligent monitoring of safety incident response |
| JP2020095617A (en) * | 2018-12-14 | 2020-06-18 | コニカミノルタ株式会社 | Safety management support system and control program |
| GB2629117B (en) * | 2019-11-11 | 2025-03-05 | Mobileye Vision Technologies Ltd | Systems and methods for determining road safety |
| CN113762183A (en) * | 2021-09-13 | 2021-12-07 | 墙管家建筑科技(上海)有限公司 | Intelligent checking and analyzing system for existing building safety and operation method |
| CN116630872A (en) * | 2023-03-31 | 2023-08-22 | 山东高速建设管理集团有限公司 | Building site safety monitoring method based on three-dimensional modeling |
-
2023
- 2023-12-15 CN CN202311728402.9A patent/CN117893477B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103258432A (en) * | 2013-04-19 | 2013-08-21 | 西安交通大学 | Traffic accident automatic identification processing method and system based on videos |
| CN111091609A (en) * | 2019-12-11 | 2020-05-01 | 云南电网有限责任公司保山供电局 | Transformer substation field operation management and control system and method based on three-dimensional dynamic modeling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117893477A (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108830837B (en) | Method and device for detecting steel ladle corrosion defect | |
| CN112070135B (en) | Power equipment image detection method and device, power equipment and storage medium | |
| CN112115927B (en) | Intelligent machine room equipment identification method and system based on deep learning | |
| CN111401419A (en) | Improved RetinaNet-based employee dressing specification detection method | |
| CN113554540A (en) | Emergency handling method and system for marine dangerous chemical substance sudden accident | |
| CN118171920B (en) | LLM model-based park safety emergency response method, device and medium | |
| CN113191274A (en) | Oil field video intelligent safety event detection method and system based on neural network | |
| CN112633661A (en) | BIM-based emergency dispatching command method, system, computer equipment and readable medium | |
| CN118116100B (en) | Intelligent inspection system and method based on digital twinning | |
| CN118279664A (en) | Identification method and system for mountain fire of corridor of power transmission line | |
| CN115761300A (en) | Method, system and detection device for dividing safety exit abnormity | |
| CN118887490A (en) | A method, device, equipment and storage medium for identifying hidden dangers of distribution box | |
| KR20230056006A (en) | Apparatus for artificial intelligence based safety diagnosis through 3d model and method thereof | |
| CN116277001A (en) | Continuous casting robot management method and system based on digital twin | |
| CN115035328B (en) | Converter image incremental automatic machine learning system and its establishment and training method | |
| CN118711118A (en) | An intelligent identification and early warning method and system for external damage of power cable channel corridor | |
| CN117893477B (en) | Intelligent security check system based on AI image recognition | |
| CN117474457B (en) | Intelligent auxiliary system for dangerous chemical and industrial and trade equipment emergency management law enforcement inspection | |
| CN114881665A (en) | Method and system for identifying electricity stealing suspected user based on target identification algorithm | |
| CN111062827B (en) | Engineering supervision method based on artificial intelligence mode | |
| CN117114420B (en) | Image recognition-based industrial and trade safety accident risk management and control system and method | |
| CN117036670B (en) | Training method, device, equipment, medium and program product of quality detection model | |
| CN110827264A (en) | Evaluation system for apparent defects of concrete member | |
| CN113222947B (en) | Intelligent detection method and system for welding defects of non-metallic materials | |
| CN115439674A (en) | Intelligent image labeling method and device based on power image knowledge graph |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |