CN116958903A - Intelligent factory safety supervision method under multi-linkage safety control mechanism - Google Patents
Intelligent factory safety supervision method under multi-linkage safety control mechanism Download PDFInfo
- Publication number
- CN116958903A CN116958903A CN202310957317.3A CN202310957317A CN116958903A CN 116958903 A CN116958903 A CN 116958903A CN 202310957317 A CN202310957317 A CN 202310957317A CN 116958903 A CN116958903 A CN 116958903A
- Authority
- CN
- China
- Prior art keywords
- image
- action
- visual
- distributed
- industrial
- 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.)
- Granted
Links
- 230000007246 mechanism Effects 0.000 title claims abstract description 39
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000009471 action Effects 0.000 claims abstract description 153
- 230000000007 visual effect Effects 0.000 claims abstract description 136
- 230000002159 abnormal effect Effects 0.000 claims abstract description 77
- 238000004519 manufacturing process Methods 0.000 claims abstract description 58
- 238000009776 industrial production Methods 0.000 claims description 65
- 238000012549 training Methods 0.000 claims description 31
- 230000000875 corresponding effect Effects 0.000 claims description 30
- 230000003213 activating effect Effects 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 11
- 230000015654 memory Effects 0.000 claims description 11
- 238000012795 verification Methods 0.000 claims description 10
- 238000013135 deep learning Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 abstract description 10
- 230000002265 prevention Effects 0.000 abstract description 8
- 238000012544 monitoring process Methods 0.000 description 30
- 230000033001 locomotion Effects 0.000 description 29
- 230000006399 behavior Effects 0.000 description 11
- 238000003860 storage Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 206010000117 Abnormal behaviour Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 239000000126 substance Substances 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012993 chemical processing Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 239000012770 industrial material Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000391 smoking effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Classifications
-
- 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
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- General Factory Administration (AREA)
- Alarm Systems (AREA)
Abstract
The application relates to a safety supervision method of an intelligent factory under a multi-linkage safety control mechanism, which can be used for carrying out action recognition on a distributed production image of the intelligent factory based on an AI image acquisition and action model, judging whether an abnormal action image exists or not, and if the abnormal action exists, sending an alarm through AI vision equipment in a corresponding area to give an alarm to an agent until the abnormal action is eliminated. Through AI visual safety production behavior analysis, the scientific management mode of dangerous pre-identification, analysis and control is carried out on the intelligent factory, the pre-control, prevention and prevention are realized, the gateway moves forward, and the aim of safety production is achieved.
Description
Technical Field
The disclosure relates to the technical field of industrial Internet of things, in particular to an intelligent factory safety supervision method, a supervision system and a control device under a multi-linkage safety management and control mechanism.
Background
The safety of the factory is an important ring in the current industrial production operation, and is the working center of gravity of each production enterprise. The industrial safety production not only relates to the standardized management and control of industrial production operators, but also relates to the standardized management of various application equipment and materials in industry, especially the standardized management of flammable and explosive chemical products. Production operation supervision for industrial production operators is a main safety supervision work in the existing enterprises, because the safety production behavior can relate to the safety of industrial materials.
The intelligent factory is a new stage of informatization development of modern factories. Most of the existing enterprises, especially chemical enterprises, conduct behavior supervision on industrial production operators of the enterprises through safety supervision staff, for example, the safety supervision technical scheme of most chemical processing enterprises is also artificial supervision and artificial safety supervision, and although the existing enterprises also have the functions of sensing and monitoring fire, for example, an infrared monitoring camera, the construction requirements of intelligent factories are not met, and the following technical defects still exist:
firstly, a safety supervision personnel is added in the manual supervision scheme, so that the labor cost of enterprises can be continuously increased;
secondly, because of subjectivity of manual supervision, security supervision defects of untight supervision and untimely supervision still exist, and risks still exist;
thirdly, the supervision scheme of the infrared monitoring camera for sensing and monitoring fire disaster only can monitor one party, is the internet of things monitoring equipment with single supervision property, lacks of AI visual monitoring and linkage management of production behaviors, and cannot achieve the efficacy of an intelligent factory;
fourth, the intelligent factories need to reduce manual intervention on the production line, and the security supervision adopted by most enterprises is contrary to the theme of the enterprises at present.
Disclosure of Invention
In order to solve the problems, the application provides an intelligent factory safety supervision method, an intelligent factory safety supervision device and a intelligent factory safety supervision control device under a multi-linkage safety control mechanism.
In one aspect of the present application, an intelligent factory safety supervision method under a multi-linkage safety control mechanism is provided, which includes the following steps:
setting working parameters of an action AI identification model through a background server of the intelligent factory;
collecting and reporting a distributed AI visual image of an intelligent factory, and storing the distributed AI visual image into a background image database;
identifying and classifying the distributed AI visual images, intelligently identifying the actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not;
if the abnormal action exists, the AI visual equipment where the image with the abnormal action exists is traced back, and an alarm is sent to the AI visual equipment through the background.
As an optional embodiment of the present application, optionally, the method for generating the action AI identification model includes:
extracting production action history data from an industrial production area of the intelligent factory from a background database of the intelligent factory;
taking the production action history data as training data, training and generating an initial training model of the production action based on a deep learning technology;
the production action initial training model is sent to a background server of an intelligent factory, and a background manager performs action recognition accuracy verification on the production action initial training model by utilizing production action real-time data of an industrial production area;
and obtaining the action AI identification model after verification, and deploying the action AI identification model on a background server of the intelligent plant.
As an optional embodiment of the present application, optionally, collecting and reporting a distributed AI visual image of the smart factory, and storing the distributed AI visual image in a background image database, including:
sending control instructions to each distributed AI visual camera deployed in an intelligent factory industrial production area, and starting to collect AI visual images reported by each AI visual camera to obtain the distributed AI visual images;
uploading the distributed AI visual image to an industrial gateway, and reporting the distributed AI visual image to a background server through the industrial gateway according to a preset timing message mechanism;
the background server receives and stores the distributed AI visual image into a background image database.
As an optional embodiment of the present application, optionally, when uploading the distributed AI visual image to an industrial gateway, further comprises:
dividing the distributed AI visual image into an industrial production image and an industrial demarcation region image according to an image division rule pre-configured on the industrial gateway;
sequentially reporting the industrial production image and the industrial delimited area image to a background server respectively according to an image identification priority pre-configured on the industrial gateway;
the timing message mechanism is a step-type time reporting mechanism, the message time of the first preferentially reported image is T1, and the message time of the second preferentially reported image is T2, wherein:
T2＝T1-k
k is a message time interval coefficient, and the value range is 20-40 s.
As an optional embodiment of the present application, optionally, the motion AI-recognition model is a binary model, including a first motion AI-recognition model for performing image recognition on the industrial production image and a second motion AI-recognition model for performing image recognition on the industrial delimited area image.
As an optional implementation manner of the present application, optionally, identifying and classifying the distributed AI visual image, and performing intelligent motion identification by using a corresponding motion AI identification model, to determine whether there is an abnormal motion, including:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial production image or not:
if the industrial production image is the industrial production image, activating the first action AI identification model, inputting the industrial production image into the first action AI identification model, extracting first action characteristics of industrial production actions in the industrial production image, carrying out action identification and judgment on the extracted first action characteristics, and judging whether preset abnormal actions exist in the first action characteristics or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
and if the image is not the industrial production image, activating the second action AI recognition model.
As an optional implementation manner of the present application, optionally, identifying and classifying the distributed AI visual image, and performing intelligent motion identification by using a corresponding motion AI identification model, to determine whether there is an abnormal motion, including:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial demarcation area image or not:
if the industrial demarcation region image is, activating the second action AI recognition model, inputting the industrial demarcation region image into the second action AI recognition model, extracting second action features of the person actions in the industrial demarcation region image, and carrying out action recognition and judgment on the extracted second action features to judge whether the second action features have preset abnormal actions or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
if not, the first action AI recognition model is activated.
As an optional implementation manner of the present application, optionally, if there is an abnormal action, an AI visual device where the image with the abnormal action is located is traced back, and an alarm is sent to the AI visual device through the background:
extracting a current frame image with abnormal actions, and tracing the image source of the current frame image according to the current frame image to obtain an AI visual camera with the image source;
acquiring the equipment ID of the AI visual camera, and sending alarm information corresponding to abnormal actions to the AI visual camera of the equipment ID through a background;
and judging whether the abnormal action is eliminated in the distributed AI visual image after the AI visual camera plays the alarm information.
In another aspect of the present application, a device for implementing the intelligent factory safety supervision method under the multi-linkage safety control mechanism is provided, including:
the intelligent factory intelligent gateway comprises an AI visual camera, an industrial gateway and an intelligent factory intelligent gateway, wherein the AI visual camera is distributed and deployed in an industrial production area of the intelligent factory and is used for collecting and reporting a distributed AI visual image of the intelligent factory and reporting the distributed AI visual image to the industrial gateway; performing voice alarm according to alarm information issued by a background;
the industrial gateway is used for reporting the distributed AI visual image to a background server according to a preset timing message mechanism;
the background server is used for registering the AI visual camera, identifying and classifying the distributed AI visual image through a configured action AI identification model, intelligently identifying actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not: if the abnormal action exists, the AI visual equipment where the image with the abnormal action exists is traced back, and an alarm is sent to the AI visual equipment through the background.
In another aspect of the present application, a control device is also provided, including:
a processor;
a memory for storing processor-executable instructions;
the processor is configured to implement the intelligent factory safety supervision method under the multi-linkage safety control mechanism when executing the executable instructions.
The application has the technical effects that:
the method comprises the steps of setting working parameters of an action AI identification model through a background server of an intelligent factory; collecting and reporting a distributed AI visual image of an intelligent factory, and storing the distributed AI visual image into a background image database; identifying and classifying the distributed AI visual images, intelligently identifying the actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not; if the abnormal action exists, the AI visual equipment where the image with the abnormal action exists is traced back, and an alarm is sent to the AI visual equipment through the background. The intelligent factory intelligent production system can perform action recognition on distributed production images of an intelligent factory based on AI image acquisition and an action model, judge whether images with abnormal actions exist, and send an alarm through AI vision equipment in a corresponding area if abnormal actions exist, and give an alarm to an agent until abnormal actions are eliminated. Through AI visual safety production behavior analysis, the scientific management mode of dangerous pre-identification, analysis and control is carried out on the intelligent factory, the pre-control, prevention and prevention are realized, the gateway moves forward, and the aim of safety production is achieved.
Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments, which proceeds with reference to the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments, features and aspects of the present disclosure and together with the description, serve to explain the principles of the disclosure.
FIG. 1 is a schematic diagram of an implementation flow of a smart factory safety supervision method under a multi-linkage safety control mechanism according to the present application;
FIG. 2 is a schematic view showing the monitoring of images of different areas of a monitor screen according to the present application;
FIG. 3 is a schematic diagram illustrating control among various application principals of the present application;
FIG. 4 is a schematic diagram of the working mechanism for the industrial gateway to report images preferentially according to the application;
fig. 5 shows a schematic diagram of an application of the control device of the present application.
Detailed Description
Various exemplary embodiments, features and aspects of the disclosure will be described in detail below with reference to the drawings. In the drawings, like reference numbers indicate identical or functionally similar elements. Although various aspects of the embodiments are illustrated in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The word "exemplary" is used herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments.
In addition, numerous specific details are set forth in the following detailed description in order to provide a better understanding of the present disclosure. It will be understood by those skilled in the art that the present disclosure may be practiced without some of these specific details. In some instances, well known means, elements, and circuits have not been described in detail so as not to obscure the present disclosure.
Example 1
As shown in fig. 1, in one aspect, the present application provides a method for intelligent factory safety supervision under a multi-linkage safety control mechanism, which includes the following steps:
s1, setting working parameters of an action AI identification model through a background server of an intelligent factory;
s2, acquiring and reporting a distributed AI visual image of the intelligent factory, and storing the distributed AI visual image into a background image database;
s3, identifying and classifying the distributed AI visual images, performing intelligent action identification by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not;
and S4, if abnormal actions exist, tracing the AI visual equipment where the images with the abnormal actions exist, and sending an alarm to the AI visual equipment through a background.
The intelligent factory industrial production area production action, namely the collected safe production action image, is subjected to AI action collection and mode detection, and an alarm can be triggered immediately by an edge AI intelligent detection analysis technology aiming at the identification of unsafe personnel actions (industrial production image) such as on/off duty, sleeping duty, dressing standardization, smoking, mobile phone playing and the like in the production process.
For area invasion, a forbidden zone which defines any polygon in a camera image picture is detected, and when a specified person or other target enters the forbidden zone (an industrial defined area image), an alarm is immediately generated.
The linked AI equipment gives an alarm, links the voice broadcast to carry out on-site reminding, and meanwhile, the message is pushed to relevant management personnel.
The method adopts an artificial intelligence technology as a means, adopts the artificial intelligence technologies such as machine vision, posture identification, abnormal behavior analysis and early warning and the like, links a broadcasting system, and realizes the purposes of pre-control, prevention and prevention of safety production due to forward gateway of the prior control in a scientific management mode of pre-identification, analysis and control of danger.
Steps S1 to S4 will be described in detail below.
As an optional embodiment of the present application, optionally, the method for generating the action AI identification model includes:
extracting production action history data from an industrial production area of the intelligent factory from a background database of the intelligent factory;
taking the production action history data as training data, training and generating an initial training model of the production action based on a deep learning technology;
the production action initial training model is sent to a background server of an intelligent factory, and a background manager performs action recognition accuracy verification on the production action initial training model by utilizing production action real-time data of an industrial production area;
and obtaining the action AI identification model after verification, and deploying the action AI identification model on a background server of the intelligent plant.
The application adopts AI action recognition technology and deep learning technology to train and generate action recognition model. And deploying the trained and generated action recognition model on a background server. The method mainly aims at identifying the image actions of personnel in an industrial production image and an industrial delimited area image, the industrial production image mainly aims at identifying and judging the safe production action of personnel in the intelligent factory generating operation, and whether unsafe personnel actions exist in the personnel in the production process is judged through a model. The industrial delimited area image is an area defined and designated in the captured image, and it is determined whether or not a person has invaded the area. If there is a person invading the area, an alarm is generated, see in particular the description of the two image areas above.
The monitoring image shown in fig. 2 is a monitoring image monitored by an AI vision camera and uploaded to a background database, and the schematic diagram includes four monitoring images (namely, industrial production images, specifically, according to the position of an area to be monitored), but on the image in the monitoring image 2, a certain area is divided into forbidden areas, the area divided by the circle is a designated forbidden area, and the generated image is a designated area monitoring image (an industrial demarcation area image). If personnel appear in the monitoring image of the appointed area, a corresponding behavior monitoring alarm can be sent out.
The industrial production image, namely, in the intelligent factory generating operation, the safety production behavior image of each person, such as each monitoring image of the person in the production area in fig. 2, is subjected to image recognition by the first action AI recognition model;
in the industrial delimited area image, that is, in the monitoring camera image, a certain designated area monitoring image, for example, in the monitoring image 2 in fig. 2 (the area is monitored by the monitoring camera 2), there is a designated area (part a storage area) which is defined by a circle and is used for independently monitoring the processing area of the part a, the designated area monitoring image is the industrial delimited area image, and the part a storage area exists in the area, so that personnel cannot enter the part a storage area at will, therefore, the part a storage area is independently monitored, and the designated area monitoring image (the monitoring image of the part a storage area) is subjected to image recognition by the second action AI recognition model.
Therefore, the images of the two areas are respectively identified and monitored by the two models, and the first action AI identification model mainly carries out image identification and analysis on the industrial production images of all the personnel in the monitoring image (industrial production image) to judge whether the actions of all the personnel reach standards. The second action AI recognition model mainly recognizes a monitoring image (a designated area monitoring image) of a designated area, judges whether a person exists in the designated area monitoring image, and if the person image appears in the designated area monitoring image, gives an alarm.
The action AI recognition model adopts action data of intelligent factory history to conduct model training, and production action history data from intelligent factory background database is used for conducting model training. Based on a deep learning technology, for example, a convolutional network nerve carries out model training on production action history data as training data to obtain an initial training model, the recognition accuracy of the initial training model is required to be checked by a background manager, the initial training model is sent to a background server, and the background manager carries out accuracy verification of the initial model action by utilizing the production action real-time data of a factory production area.
As for the selection of the deep learning technology, the user selects the deep learning technology by himself, the convolutional neural network is preferentially adopted for model training, the generated initial training model is input into the action initial training model by a background manager through action real-time image data of a certain frame, action recognition verification is carried out, whether action abnormality exists in the current frame image is judged, and whether the recognition accuracy of the action reaches an ideal value is judged and checked. The ideal value is set in a self-defining way, for example, accuracy verification is carried out by adopting an action similarity calculation, judgment algorithm or other algorithms, and the preset value is reached to indicate that the test training model is qualified, and the verification is passed. After passing the authentication, setting corresponding model parameters and the requirements of the intelligent factory on the action recognition precision of the initial training model, and disposing the model parameters and the requirements on the background server for the recognition of the action image of the later frame image and the judgment of abnormal actions/abnormal behaviors.
As an optional embodiment of the present application, optionally, collecting and reporting a distributed AI visual image of the smart factory, and storing the distributed AI visual image in a background image database, including:
sending control instructions to each distributed AI visual camera deployed in an intelligent factory industrial production area, and starting to collect AI visual images reported by each AI visual camera to obtain the distributed AI visual images;
uploading the distributed AI visual image to an industrial gateway, and reporting the distributed AI visual image to a background server through the industrial gateway according to a preset timing message mechanism;
the background server receives and stores the distributed AI visual image into a background image database.
As shown in fig. 3, a plurality of AI vision cameras are deployed in an industrial production area of the smart factory, the AI vision cameras are deployed in distributed nodes of each production industrial production area, each distributed node is set by a factory administrator, each forbidden area is included, and cameras are required to be deployed in places needing to be monitored. The AI vision camera has a voice broadcasting function, and can receive a background instruction and background alarm information to carry out voice broadcasting and alarm. In the embodiment, the industrial gateway is adopted to control all the AI cameras distributed and used for reporting the collected monitoring images reported by all the distributed and deployed cameras. The industrial gateway is provided with a timing message mechanism, images of different areas can be uploaded to the industrial gateway according to the timing message mechanism, then the images are uploaded to a background server by the industrial gateway, and after the images are received by the background, the received distributed AI visual images are stored in a background image database. The gateway can control a plurality of AI vision cameras in the production area, and can manage the monitoring images in a centralized way. The method can report the industrial production images and the industrial demarcation region images collected and reported in different regions in batches, and lighten the reported network transmission pressure.
And when the AI visual images are acquired, the background uniformly transmits control instructions to all the AI visual cameras. When each AI vision camera works, the images collected by each AI vision camera are reported to the industrial gateway, and the industrial gateway reports the images in batches according to a preset message mechanism.
As an optional embodiment of the present application, optionally, when uploading the distributed AI visual image to an industrial gateway, further comprises:
dividing the distributed AI visual image into an industrial production image and an industrial demarcation region image according to an image division rule pre-configured on the industrial gateway;
sequentially reporting the industrial production image and the industrial delimited area image to a background server respectively according to an image identification priority pre-configured on the industrial gateway;
the timing message mechanism is a step-type time reporting mechanism, the message time of the first preferentially reported image is T1, and the message time of the second preferentially reported image is T2, wherein:
T2＝T1-k
k is a message time interval coefficient, and the value range is 20-40 s.
As shown in fig. 4, the distributed AI visual image on the industrial gateway is reported, including the safe production image of the production work area and the behavioural action image in the forbidden zone. The AI visual image is divided before the industrial gateway reports to the background server. Dividing the images reported by the AI vision cameras into an industrial production image and an industrial demarcation region image, and reporting the two images back and forth in batches according to time intervals and the priority of image identification.
And after the industrial gateway receives the industrial production images and the industrial forbidden zone images reported by the AI vision cameras, the industrial production images and the industrial zone images are sequentially reported to the background server according to the identified sequence by identifying the priority of the images which are pre-configured on the industrial gateway. For example, if an image set as an industrial delimited area image, that is, an image of a forbidden area is identified as an image to be preferentially processed, the industrial delimited area image is preferentially reported to a background server through an industrial gateway.
The industrial production image and the industrial delimited area image need to be reported according to a certain time sequence, and the embodiment is provided with a step-type time reporting mechanism. The first image to be reported is reported again at intervals of half a minute after reporting, and the interval time is specifically set by a background manager, so that the interval time can be properly reduced. For a strict reporting time of the registration control, a second priority reporting time of 0.5s later than the first priority reporting time may be taken, and so on.
The image division rule configured on the gateway may, for example, divide according to the attribute of the AI camera device to which the current image belongs. And for the images in the forbidden zone, the corresponding attribute values can be distinguished by automatic configuration during uploading.
As an optional embodiment of the present application, optionally, the motion AI-recognition model is a binary model, including a first motion AI-recognition model for performing image recognition on the industrial production image and a second motion AI-recognition model for performing image recognition on the industrial delimited area image.
In the present embodiment, when training the motion AI-recognition model, two models, i.e., a double-sub model, including a first motion AI-recognition model for performing image recognition on an industrial production image and a second motion AI-recognition model for performing recognition on a forbidden region motion behavior image, need to be trained and generated. The two motion recognition models are obtained by performing deep training by adopting historical motion image data of two area images respectively, and the description of model training is specifically shown in the deep learning.
And when the background receives the image reported by the current industrial gateway, respectively calling and activating the image according to the attribute of the image. The corresponding action AI identifies the model.
As an optional implementation manner of the present application, optionally, identifying and classifying the distributed AI visual image, and performing intelligent motion identification by using a corresponding motion AI identification model, to determine whether there is an abnormal motion, including:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial production image or not:
if the industrial production image is the industrial production image, activating the first action AI identification model, inputting the industrial production image into the first action AI identification model, extracting first action characteristics of industrial production actions in the industrial production image, carrying out action identification and judgment on the extracted first action characteristics, and judging whether preset abnormal actions exist in the first action characteristics or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
and if the image is not the industrial production image, activating the second action AI recognition model.
For example, if the image reported by the gateway in current contact is an industrial production image, the first action AI recognition model is activated. And cutting the current industrial production image according to a frame image mode. And inputting each frame of image into an action AI recognition model, extracting action characteristics of production actions in the industrial production image by the model, and carrying out action recognition and judgment on the action characteristics. The action AI recognition model may predict the behavior of the extracted action feature, and determine whether a preset abnormal action exists in the first action feature, for example, an action that production staff has production job non-normative exists. For the manner of motion feature extraction and motion recognition judgment by the motion AI recognition model, see the technical principle of convolutional neural network specifically.
By adopting the action AI recognition model, action characteristics can be extracted, action standard judgment can be carried out on the action characteristics, whether the action characteristics are matched with the characteristics of earlier training and learning is judged, if so, abnormal actions exist in the production actions of workers in the current frame image, and an alarm is required.
The present embodiment will not be described with respect to the division of frame images and the input recognition of each frame image. And for the true image with abnormal actions, storing the current frame image in a background image database, so that the follow-up tracking of the original image to which the current frame image belongs is facilitated, the source equipment of the current frame image is tracked, and the AI action vision camera to which the current frame image belongs is judged. After the source equipment of the frame image is tracked, the equipment address of the AI visual camera is recorded and saved, and corresponding alarm information is sent to the AI visual camera of the equipment ID by the background. After the AI vision camera of the equipment address receives alarm information issued by the background, the alarm information is broadcasted, corresponding production staff are reminded of paying attention to behaviors, and correction is timely carried out. Information forwarding between actions of the period will be forwarded by the industrial gateway.
The background activates the corresponding AI recognition model according to the priority image reported by the industrial gateway. If the current image is not of priority, activating an action AI identification model of another priority, and waiting for the frame image of another image to be input into the activated model.
As an optional implementation manner of the present application, optionally, identifying and classifying the distributed AI visual image, and performing intelligent motion identification by using a corresponding motion AI identification model, to determine whether there is an abnormal motion, including:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial demarcation area image or not:
if the industrial demarcation region image is, activating the second action AI recognition model, inputting the industrial demarcation region image into the second action AI recognition model, extracting second action features of the person actions in the industrial demarcation region image, and carrying out action recognition and judgment on the extracted second action features to judge whether the second action features have preset abnormal actions or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
if not, the first action AI recognition model is activated.
The principle of the second motion AI identification model refers to the first motion AI identification model, and will not be described herein.
As an optional implementation manner of the present application, optionally, if there is an abnormal action, an AI visual device where the image with the abnormal action is located is traced back, and an alarm is sent to the AI visual device through the background:
extracting a current frame image with abnormal actions, and tracing the image source of the current frame image according to the current frame image to obtain an AI visual camera with the image source;
acquiring the equipment ID of the AI visual camera, and sending alarm information corresponding to abnormal actions to the AI visual camera of the equipment ID through a background;
and judging whether the abnormal action is eliminated in the distributed AI visual image after the AI visual camera plays the alarm information.
The background finds out a frame image with abnormal actions, then the source image of the frame image can be tracked, the image source of the current frame image is traced back, the equipment ID of the AI visual camera to which the frame image belongs is obtained, the background records the equipment ID, and alarm information corresponding to the abnormal actions is sent to the AI visual camera corresponding to the equipment ID. The alarm information can be composed of abnormal actions and preset alarm voices of the background, and is specifically set by a background manager.
And the AI vision camera broadcasts after receiving the alarm information sent by the background, and reminds corresponding production personnel to correct errors. After the AI vision camera plays the alarm information, the behavior of the staff in the area can be continuously monitored. The background can continuously recognize and judge the follow-up action of the staff according to the processing steps. If the action of the person is still not standard or not corrected in time, the alarm warning of the corresponding level is continuously sent until the abnormal action is eliminated.
Therefore, the intelligent factory distributed production image processing method and the intelligent factory distributed production image processing system can be used for carrying out action recognition on the intelligent factory distributed production image based on the AI image acquisition and the action model, judging whether an abnormal action image exists or not, and if abnormal action exists, sending an alarm through AI vision equipment in a corresponding area to give an alarm to an agent until the abnormal action is eliminated. Through AI visual safety production behavior analysis, the scientific management mode of dangerous pre-identification, analysis and control is carried out on the intelligent factory, the pre-control, prevention and prevention are realized, the gateway moves forward, and the aim of safety production is achieved.
It should be apparent to those skilled in the art that implementing all or part of the above-described embodiments may be accomplished by computer programs to instruct related hardware, and the programs may be stored in a computer readable storage medium, which when executed may include the processes of the embodiments of the controls described above. It will be appreciated by those skilled in the art that implementing all or part of the above-described embodiments may be accomplished by computer programs to instruct related hardware, and the programs may be stored in a computer readable storage medium, which when executed may include the processes of the embodiments of the controls described above. The storage medium may be a magnetic disk, an optical disc, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), a flash memory (flash memory), a hard disk (HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
Example 2
As shown in fig. 3, based on the implementation principle of embodiment 1, another aspect of the present application provides an apparatus for implementing the intelligent factory safety supervision method under the multi-linkage safety control mechanism, which includes:
the intelligent factory intelligent gateway comprises an AI visual camera, an industrial gateway and an intelligent factory intelligent gateway, wherein the AI visual camera is distributed and deployed in an industrial production area of the intelligent factory and is used for collecting and reporting a distributed AI visual image of the intelligent factory and reporting the distributed AI visual image to the industrial gateway; performing voice alarm according to alarm information issued by a background;
the industrial gateway is used for reporting the distributed AI visual image to a background server according to a preset timing message mechanism;
the background server is used for registering the AI visual camera, identifying and classifying the distributed AI visual image through a configured action AI identification model, intelligently identifying actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not: if the abnormal action exists, the AI visual equipment where the image with the abnormal action exists is traced back, and an alarm is sent to the AI visual equipment through the background.
The interaction between the respective application subjects described above is specifically referred to the description of embodiment 1.
The modules or steps of the application described above may be implemented in a general-purpose computing device, they may be centralized in a single computing device, or distributed across a network of computing devices, or they may alternatively be implemented in program code executable by a computing device, such that they may be stored in a memory device and executed by a computing device, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.
Example 3
Still further, another aspect of the present application provides a control device, including:
a processor;
a memory for storing processor-executable instructions;
the processor is configured to implement the intelligent factory safety supervision method under the multi-linkage safety control mechanism when executing the executable instructions.
Embodiments of the present disclosure control an apparatus that includes a processor and a memory for storing processor-executable instructions. The processor is configured to implement any one of the above-described intelligent factory safety supervision methods under a multi-linkage safety control mechanism when executing the executable instructions.
Here, it should be noted that the number of processors may be one or more. Meanwhile, in the control device of the embodiment of the present disclosure, an input device and an output device may be further included. The processor, the memory, the input device, and the output device may be connected by a bus, or may be connected by other means, which is not specifically limited herein.
The memory, as a computer-readable storage medium, may be used to store software programs, computer-executable programs, and various modules such as: a program or module corresponding to an intelligent factory safety supervision method under a multi-linkage safety control mechanism in an embodiment of the disclosure. The processor executes various functional applications and data processing of the control device by running software programs or modules stored in the memory.
The input device may be used to receive an input number or signal. Wherein the signal may be a signal related to user settings of the device/terminal/server and function control. The output means may comprise a display device such as a display screen.
The foregoing description of the embodiments of the present disclosure has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the various embodiments described. The terminology used herein was chosen in order to best explain the principles of the embodiments, the practical application, or the technical improvements in the marketplace, or to enable others of ordinary skill in the art to understand the embodiments disclosed herein.
Claims (8)
1. The intelligent factory safety supervision method under the multi-linkage safety control mechanism is characterized by comprising the following steps of:
setting working parameters of an action AI identification model through a background server of the intelligent factory;
collecting and reporting a distributed AI visual image of an intelligent factory, and storing the distributed AI visual image into a background image database;
identifying and classifying the distributed AI visual images, intelligently identifying the actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not;
if the abnormal action exists, tracing the AI visual equipment where the image with the abnormal action exists, and sending an alarm to the AI visual equipment through a background;
wherein:
collecting and reporting a distributed AI visual image of an intelligent factory, storing the distributed AI visual image into a background image database, and comprising the following steps:
sending control instructions to each distributed AI visual camera deployed in an intelligent factory industrial production area, and starting to collect AI visual images reported by each AI visual camera to obtain the distributed AI visual images;
uploading the distributed AI visual image to an industrial gateway, and reporting the distributed AI visual image to a background server through the industrial gateway according to a preset timing message mechanism;
the background server receives and stores the distributed AI visual image into a background image database;
upon uploading the distributed AI visual image to an industrial gateway, further comprising:
dividing the distributed AI visual image into an industrial production image and an industrial demarcation region image according to an image division rule pre-configured on the industrial gateway;
sequentially reporting the industrial production image and the industrial delimited area image to a background server respectively according to an image identification priority pre-configured on the industrial gateway;
the timing message mechanism is a step-type time reporting mechanism, the message time of the first preferentially reported image is T1, and the message time of the second preferentially reported image is T2, wherein:
T2＝T1-k
k is a message time interval coefficient, and the value range is 20-40 s.
2. The intelligent factory safety supervision method under the multi-linkage safety control mechanism according to claim 1, wherein the generating method of the action AI identification model comprises the following steps:
extracting production action history data from an industrial production area of the intelligent factory from a background database of the intelligent factory;
taking the production action history data as training data, training and generating an initial training model of the production action based on a deep learning technology;
the production action initial training model is sent to a background server of an intelligent factory, and a background manager performs action recognition accuracy verification on the production action initial training model by utilizing production action real-time data of an industrial production area;
and obtaining the action AI identification model after verification, and deploying the action AI identification model on a background server of the intelligent plant.
3. The intelligent factory safety supervision method under the multi-linkage safety control mechanism according to claim 2, wherein the action AI-recognition model is a binary model, and comprises a first action AI-recognition model for performing image recognition on an industrial production image and a second action AI-recognition model for performing image recognition on an industrial delimited area image.
4. The intelligent factory safety supervision method under the multi-linkage safety control mechanism according to claim 3, wherein the steps of identifying and classifying the distributed AI visual images, intelligently identifying the actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist include:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial production image or not:
if the industrial production image is the industrial production image, activating the first action AI identification model, inputting the industrial production image into the first action AI identification model, extracting first action characteristics of industrial production actions in the industrial production image, carrying out action identification and judgment on the extracted first action characteristics, and judging whether preset abnormal actions exist in the first action characteristics or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
and if the image is not the industrial production image, activating the second action AI recognition model.
5. The intelligent factory safety supervision method under the multi-linkage safety control mechanism according to claim 4, wherein the steps of identifying and classifying the distributed AI visual images, intelligently identifying actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist include:
the background receives the distributed AI visual image, carries out type identification on the currently received distributed AI visual image, and judges whether the current AI visual image is an industrial demarcation area image or not:
if the industrial demarcation region image is, activating the second action AI recognition model, inputting the industrial demarcation region image into the second action AI recognition model, extracting second action features of the person actions in the industrial demarcation region image, and carrying out action recognition and judgment on the extracted second action features to judge whether the second action features have preset abnormal actions or not:
if the abnormal action exists, recording a current frame image with the abnormal action and storing the current frame image into a background image database;
if not, discarding;
if not, the first action AI recognition model is activated.
6. The method for intelligent factory safety supervision under a multi-linkage safety control mechanism according to claim 4, wherein if there is an abnormal action, the AI vision equipment where the image with the abnormal action is located is traced back, and an alarm is sent to the AI vision equipment through a background:
extracting a current frame image with abnormal actions, and tracing the image source of the current frame image according to the current frame image to obtain an AI visual camera with the image source;
acquiring the equipment ID of the AI visual camera, and sending alarm information corresponding to abnormal actions to the AI visual camera of the equipment ID through a background;
and judging whether the abnormal action is eliminated in the distributed AI visual image after the AI visual camera plays the alarm information.
7. An apparatus for implementing the intelligent factory safety supervision method under the multi-linkage safety control mechanism according to any one of claims 1 to 6, comprising:
the intelligent factory intelligent gateway comprises an AI visual camera, an industrial gateway and an intelligent factory intelligent gateway, wherein the AI visual camera is distributed and deployed in an industrial production area of the intelligent factory and is used for collecting and reporting a distributed AI visual image of the intelligent factory and reporting the distributed AI visual image to the industrial gateway; performing voice alarm according to alarm information issued by a background;
the industrial gateway is used for reporting the distributed AI visual image to a background server according to a preset timing message mechanism;
the background server is used for registering the AI visual camera, identifying and classifying the distributed AI visual image through a configured action AI identification model, intelligently identifying actions by adopting a corresponding action AI identification model, and judging whether abnormal actions exist or not: if the abnormal action exists, the AI visual equipment where the image with the abnormal action exists is traced back, and an alarm is sent to the AI visual equipment through the background.
8. A control apparatus, characterized by comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to implement the intelligent plant safety supervision method under the multi-linkage safety control mechanism of any one of claims 1-6 when executing the executable instructions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310957317.3A CN116958903B (en) | 2023-08-01 | 2023-08-01 | Intelligent factory safety supervision method under multi-linkage safety control mechanism |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310957317.3A CN116958903B (en) | 2023-08-01 | 2023-08-01 | Intelligent factory safety supervision method under multi-linkage safety control mechanism |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116958903A true CN116958903A (en) | 2023-10-27 |
| CN116958903B CN116958903B (en) | 2024-06-18 |
Family
ID=88461717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310957317.3A Active CN116958903B (en) | 2023-08-01 | 2023-08-01 | Intelligent factory safety supervision method under multi-linkage safety control mechanism |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116958903B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110176117A (en) * | 2019-06-17 | 2019-08-27 | 广东翔翼科技信息有限公司 | A kind of monitoring device and monitoring method of Behavior-based control identification technology |
| CN111753724A (en) * | 2020-06-24 | 2020-10-09 | 上海依图网络科技有限公司 | Abnormal behavior identification method and device |
| CN112165601A (en) * | 2020-08-27 | 2021-01-01 | 南京市晨枭软件技术有限公司 | State monitoring platform of distributed industrial system |
| CN112839200A (en) * | 2020-12-25 | 2021-05-25 | 朗坤智慧科技股份有限公司 | Power plant violation behavior identification method and system based on 5G technology and network server |
| CN114333273A (en) * | 2021-12-21 | 2022-04-12 | 江苏核电有限公司 | Intelligent analysis system and method for nuclear power construction site |
| CN114446004A (en) * | 2022-01-29 | 2022-05-06 | 贵州中烟工业有限责任公司 | Security protection system |
| CN114488921A (en) * | 2022-04-14 | 2022-05-13 | 山东省农业机械科学研究院 | Internet of things management and control system for intelligent orchard |
| KR20220085192A (en) * | 2020-12-15 | 2022-06-22 | (주)유엠로직스 | Behavior abnormal detection intelligent sharing system and method in smart factory based on profiling |
| KR20220122923A (en) * | 2021-11-29 | 2022-09-05 | 주식회사 알엠에이 | System for monitoring and acquring data in smart factory |
| CN115471865A (en) * | 2022-08-19 | 2022-12-13 | 安徽继远软件有限公司 | Operation site digital safety control method, device, equipment and storage medium |
| CN116311036A (en) * | 2023-02-10 | 2023-06-23 | 中国工商银行股份有限公司 | Behavior recognition method, behavior recognition device, computer equipment and storage medium |
| CN116308955A (en) * | 2023-03-16 | 2023-06-23 | 中浙信科技咨询有限公司 | Intelligent community for health care management based on AI artificial intelligence |
-
2023
- 2023-08-01 CN CN202310957317.3A patent/CN116958903B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110176117A (en) * | 2019-06-17 | 2019-08-27 | 广东翔翼科技信息有限公司 | A kind of monitoring device and monitoring method of Behavior-based control identification technology |
| CN111753724A (en) * | 2020-06-24 | 2020-10-09 | 上海依图网络科技有限公司 | Abnormal behavior identification method and device |
| CN112165601A (en) * | 2020-08-27 | 2021-01-01 | 南京市晨枭软件技术有限公司 | State monitoring platform of distributed industrial system |
| KR20220085192A (en) * | 2020-12-15 | 2022-06-22 | (주)유엠로직스 | Behavior abnormal detection intelligent sharing system and method in smart factory based on profiling |
| CN112839200A (en) * | 2020-12-25 | 2021-05-25 | 朗坤智慧科技股份有限公司 | Power plant violation behavior identification method and system based on 5G technology and network server |
| KR20220122923A (en) * | 2021-11-29 | 2022-09-05 | 주식회사 알엠에이 | System for monitoring and acquring data in smart factory |
| CN114333273A (en) * | 2021-12-21 | 2022-04-12 | 江苏核电有限公司 | Intelligent analysis system and method for nuclear power construction site |
| CN114446004A (en) * | 2022-01-29 | 2022-05-06 | 贵州中烟工业有限责任公司 | Security protection system |
| CN114488921A (en) * | 2022-04-14 | 2022-05-13 | 山东省农业机械科学研究院 | Internet of things management and control system for intelligent orchard |
| CN115471865A (en) * | 2022-08-19 | 2022-12-13 | 安徽继远软件有限公司 | Operation site digital safety control method, device, equipment and storage medium |
| CN116311036A (en) * | 2023-02-10 | 2023-06-23 | 中国工商银行股份有限公司 | Behavior recognition method, behavior recognition device, computer equipment and storage medium |
| CN116308955A (en) * | 2023-03-16 | 2023-06-23 | 中浙信科技咨询有限公司 | Intelligent community for health care management based on AI artificial intelligence |
Non-Patent Citations (3)
| Title |
|---|
| CLEMENS POHLT等: "Human Work Activity Recognition for Working Cells in Industrial Production Contexts", 2019 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC), 28 November 2019 (2019-11-28), pages 4225 - 4230 * |
| 周亘儒: "煤矿安全智能视频识别系统研究", 内蒙古煤炭经济, 15 June 2022 (2022-06-15), pages 106 - 108 * |
| 谢鹏等: "基于机器学习的军工视频分析云平台技术研究", 新技术新工艺, 25 February 2023 (2023-02-25), pages 1 - 7 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116958903B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108009728B (en) | Regional security management method and system in park | |
| CN111080775A (en) | Server routing inspection method and system based on artificial intelligence | |
| CN110929648A (en) | Monitoring data processing method and device, computer equipment and storage medium | |
| US20170076576A1 (en) | Activity monitoring method and system | |
| US11631306B2 (en) | Methods and system for monitoring an environment | |
| CN113992893A (en) | Park inspection method and device, storage medium and electronic device | |
| CN110472749B (en) | Remote monitoring method and monitoring equipment for equipment | |
| CN114418388A (en) | Personnel management method and device based on image recognition and storage medium | |
| US10984650B2 (en) | Systems and methods for managing alert notifications from a secured area | |
| CN112187914A (en) | Remote control robot management method and system | |
| CN117808265A (en) | Dynamic display reminding method, device, equipment and medium for enterprise project production information | |
| CN115410112A (en) | Elevator abnormal stay recognition method and device and storage medium | |
| CN118115939A (en) | Alarm event processing method and device, nonvolatile storage medium and electronic equipment | |
| CN116958903B (en) | Intelligent factory safety supervision method under multi-linkage safety control mechanism | |
| CN113486799A (en) | Device linkage method, device, storage medium and program product | |
| CN116308955A (en) | Intelligent community for health care management based on AI artificial intelligence | |
| CN113837138B (en) | Dressing monitoring method, dressing monitoring system, dressing monitoring medium and electronic terminal | |
| CN114768158B (en) | Intelligent fire fighting system and automatic inspection method thereof | |
| CN113420739B (en) | Intelligent emergency monitoring method and system based on neural network and readable storage medium | |
| CN115342937A (en) | Temperature anomaly detection method and device | |
| CN110891039B (en) | LM neural network-based coal mine monitoring system safety early warning method | |
| CN114372500A (en) | Intelligent factory control system based on big data | |
| CN112162906A (en) | Server behavior monitoring method of probe management platform architecture | |
| Raja | An Intelligent Framework for Safeguarding and Surveillance of Women in Smart Cities | |
| CN118097543A (en) | AI-based early warning method and device, storage medium and electronic equipment |
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 | ||
| GR01 | Patent grant |