CN115205761A - Accident reason off-line intelligent diagnosis system - Google Patents
Accident reason off-line intelligent diagnosis system Download PDFInfo
- Publication number
- CN115205761A CN115205761A CN202210980387.6A CN202210980387A CN115205761A CN 115205761 A CN115205761 A CN 115205761A CN 202210980387 A CN202210980387 A CN 202210980387A CN 115205761 A CN115205761 A CN 115205761A
- Authority
- CN
- China
- Prior art keywords
- key feature
- recognition algorithm
- feature recognition
- accident
- algorithm model
- 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.)
- Pending
Links
Images
Classifications
-
- 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
- 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace 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
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an accident cause off-line intelligent diagnosis system, which relates to the technical field of dangerous chemical accident monitoring and comprises a video data reading layer, a central processing layer, a data storage layer and a key feature recognition algorithm model library; the key feature recognition algorithm model library stores a plurality of key feature recognition algorithm models, the key feature recognition algorithm models are deep learning neural network models and are trained on the corresponding chemical accident image data acquired from the starting data set; the video data reading layer is used for acquiring monitoring video image data; the central processing layer is used for selecting a key feature recognition algorithm model to diagnose the monitoring video image data and outputting a diagnosis result; the data storage layer is used for storing the diagnosis result. The invention utilizes the computer to assist or replace the manual work to carry out the identification and analysis on the off-line video, filters the useless information which is not concerned by the user, only extracts the key information of the specific event, and greatly improves the event diagnosis efficiency.
Description
Technical Field
The invention relates to the technical field of dangerous chemical production and operation safety management, in particular to an accident cause offline intelligent diagnosis system.
Background
With the gradual development of video monitoring technology, video monitoring is widely applied in various fields, the data volume of monitoring video is also increased in a large amount, and the video monitoring data is an important data object in a big data era.
At present, the intelligent video analysis technology is mostly based on the basic research of the intelligent video technology, but the intelligent video analysis key technology is applied to the accident cause diagnosis and analysis, and particularly the intelligent video analysis technology is rarely related to the safety production field of the industries such as dangerous chemical production and operation. For dangerous chemical production and operation enterprises, accident risks need to be prevented all the time, if accidents happen carelessly, an emergency plan should be taken immediately after the accidents happen, and then accident reasons should be investigated as soon as possible to avoid subsequent production and operation risks. When the accident reason is investigated, in the face of massive data information provided by a monitoring video, if an accident investigator does not have a method for effectively and scientifically extracting the massive data information, a large amount of human resources and time cost are consumed, and the problem of missed inspection and missed inspection is easy to occur.
Disclosure of Invention
The invention aims to provide an accident cause offline intelligent diagnosis system which can alleviate the problems.
In order to alleviate the above problems, the technical scheme adopted by the invention is as follows:
the invention provides an accident cause offline intelligent diagnosis system which comprises a video data reading layer, a central processing layer, a data storage layer and a key feature recognition algorithm model library, wherein the video data reading layer is used for reading video data;
the key feature recognition algorithm model base stores a plurality of key feature recognition algorithm models, the key feature recognition algorithm models are deep learning neural network models and are trained on the corresponding chemical accident image data acquired from the opening source data set;
the video data reading layer is used for acquiring monitoring video image data;
the central processing layer is used for selecting a key feature recognition algorithm model of a corresponding class from the key feature recognition algorithm model library as a current accident reason diagnosis model, inputting the monitoring video image data into the current accident reason diagnosis model, diagnosing the monitoring video image data by the current accident reason diagnosis model, if no key feature is recognized, not outputting a diagnosis result, if the key feature is recognized, outputting the probability that the key feature belongs to the accident reason of the class corresponding to the current accident reason diagnosis model, and taking a video segment and a screenshot corresponding to the key feature as the diagnosis result;
the data storage layer is used for storing the output diagnosis result.
In a preferred embodiment of the present invention, the key feature recognition algorithm model includes a convolutional neural network model and a plurality of classifiers, where the classification precision of the classifiers is adjustable, and the higher the precision is, the lower the recognition rate is, and otherwise, the higher the recognition rate is; for each key feature recognition algorithm model, the central processing layer can select classification operation with corresponding precision for diagnosing the monitoring video image data.
In a preferred embodiment of the present invention, the central processing layer includes a condition selection window, and the condition selection window is provided with an analysis precision selection window and a key feature recognition algorithm model selection window.
In a preferred embodiment of the present invention, the training method of the key feature recognition algorithm model includes convolutional neural network model training and classifier training.
In a preferred embodiment of the present invention, the training process of the key feature recognition algorithm model includes: training the convolutional neural network model by using the chemical accident image data acquired from the open source data set, so that the convolutional neural network model can identify key features in the chemical accident image data; performing image CNN feature extraction on chemical accident image data acquired from a development source data set by adopting a trained convolutional neural network model to construct a key feature training set; and training the classifier by adopting a key characteristic training set.
Compared with the prior art, the invention has the beneficial effects that:
the method applies the key characteristic recognition algorithm model base of the dangerous chemical production and operation enterprises established by the video analysis technology to the diagnosis and recognition of accident causes, utilizes a computer to assist or replace manpower to carry out recognition and analysis on the offline video, filters useless information which is not concerned by users, extracts only the key information of specific events, and greatly improves the event diagnosis efficiency;
the invention realizes the adjustable recognition rate of the intelligent recognition system through the selectable model recognition accuracy.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a block diagram of an off-line intelligent diagnosis system for accident causes according to the present invention;
FIG. 2 is a flow chart of accident diagnosis of the off-line intelligent diagnosis system for accident causes according to the present invention;
FIG. 3 is a block diagram of the structure of the key feature recognition algorithm model of the present invention and illustrates the diagnostic process thereof.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 to 3, the invention provides an accident cause offline intelligent diagnosis system, which includes a video data reading layer, a central processing layer, a data storage layer and a key feature recognition algorithm model library.
In the invention, a plurality of key feature recognition algorithm models are stored in a key feature recognition algorithm model library, the number of categories of the key feature recognition algorithm models can be expanded according to needs, for example, the key feature recognition algorithm models can comprise a personnel gathering key feature recognition algorithm model, an illegal break-in key feature recognition algorithm model, an off duty key feature recognition algorithm model, a fire fighting channel blocking key feature recognition algorithm model, an environmental firing key feature recognition algorithm model and an environmental smoking key feature recognition algorithm model, the models can be used for judging whether key features with higher similarity to the models appear in videos to be recognized, if the key features appear in pictures, the system automatically judges that abnormal events occur, extracts and stores video paragraphs in which the abnormal events occur, and generates a record table, so that workers can conveniently and quickly capture abnormal information to obtain accident reasons; if the key features in the model do not appear in the picture, a similar typeface such as the key features which are not detected is output.
In the invention, the key feature recognition algorithm model is a deep learning neural network model and is trained on the corresponding class of chemical accident image data acquired from an open source data set.
The method comprises the steps of starting a source data set, wherein the source data set can be image databases such as ImageNet and/or Pascal VOC, a large number of image sources are stored in the databases, required chemical accident image data can be selected from the image databases, all selected images are manually marked by using a LabelImg tool, namely information to be identified in the data set is marked (for example, the part on fire in the image is manually marked, and the mark label is 'on fire'), so that the marked image data and a label file corresponding to the image data are obtained.
In the invention, the key feature recognition algorithm model comprises a convolutional neural network model and a plurality of classifiers, the classification precision of the classifiers is adjustable, the higher the precision is, the lower the recognition rate is, and the higher the recognition rate is otherwise.
The structure of each classifier may be different, for example, softmax, SVM, random forest, etc., and the structure of each classifier may also be the same, but the training data sources used are different.
In the invention, the training method of the key feature recognition algorithm model comprises convolutional neural network model training and classifier training, and specifically comprises the following steps:
1) Training a convolutional neural network model by using the chemical accident image data acquired from the opening source data set, so that the convolutional neural network model can identify key features in the chemical accident image data;
2) Adopting a trained convolutional neural network model to perform image CNN feature extraction on chemical accident image data acquired from an opening source data set, and constructing a key feature training set;
3) And training the classifier by adopting a key characteristic training set.
In the invention, the central processing layer calls the key feature recognition algorithm model of the corresponding type from the key feature recognition algorithm model library as the current accident cause diagnosis model according to the accident type to be recognized, as shown in fig. 1.
The video image data to be diagnosed, which is acquired by the video data reading layer, is firstly input into a convolutional neural network model of a current accident cause diagnosis model, and required key features are identified.
In order to realize adjustable classification precision, the classifiers are combined for use, N classifiers are arranged, and the specific combination mode of obtaining diagnosis results with different precisions is as follows according to the sequence from low to high:
diagnosis result of first accuracy:
and only the identified key features are selected to be input into the first classifier, and the classification result obtained by the first classifier is used as a diagnosis result with first precision.
Diagnosis result of second accuracy:
and selecting and inputting the identified key features into a first classifier and a second classifier, and combining classification results respectively obtained by the first classifier and the second classifier to finally obtain a diagnosis result with second precision.
Diagnosis result of third precision:
and inputting the identified key features into a first classifier, a second classifier and a third classifier, and combining classification results respectively obtained by the first classifier, the second classifier and the third classifier to finally obtain a diagnosis result with third precision.
……
Diagnosis result of nth precision:
and selecting and inputting the identified key features into a first classifier, a second classifier, a third classifier, … … and an Nth classifier, and combining classification results respectively obtained by the first classifier, the second classifier, the third classifier, … … and the Nth classifier to finally obtain a diagnosis result with Nth precision.
Taking three classifiers as an example, the combination and data trend thereof are shown in fig. 3.
In the invention, the classification results of all classifiers can be combined by adopting a method for setting the weights of the classifiers, all the classifiers have respective corresponding weights, and the sum of the weights of all the classifiers is 1. An initial weight can be set initially, and then correction is carried out according to the performance state in practical application, and the sum of the weights of all classifiers still needs to be ensured to be 1 after correction.
The correction method comprises the following steps: when the classification result of one classifier is correct and the classification result of the other classifier is wrong, the weight of the classifier which detects the correct classifier is increased by 0.001, the weight of the classifier which detects the wrong classifier is decreased by 0.001, and the weight of the classifier which does not participate in the detection is not changed.
When key feature recognition is carried out, firstly, the weight of the currently selected classifier is adjusted according to the number of the currently selected classifier, so that the weight sum of the currently selected classifier is 1, and the relative proportion of the weights of the currently selected classifier is unchanged before and after adjustment;
and then multiplying the probabilities in the classification results of the currently selected classifiers by the respective adjusted weights, and adding the obtained results to obtain a new probability, namely the probability of the accident reason of the corresponding class of the current accident reason diagnosis model.
We found through multiple tests that the more classifiers selected, the higher the accuracy of the diagnosis, but the longer the time taken.
In the invention, the central processing layer comprises a condition selection window, and the window is provided with a key feature recognition algorithm model selection window and an analysis precision selection window.
The key feature recognition algorithm model selection window of the central processing layer can select the corresponding key feature recognition algorithm model as the current accident cause diagnosis model, and after the analysis precision selection window of the central processing layer can select the analysis precision of the current accident cause diagnosis model, the method is equivalent to the combination mode of the classifier of the current accident cause diagnosis model.
And inputting the monitoring video image data from the video data reading layer into the current accident reason diagnosis model, diagnosing the monitoring video image data by the current accident reason diagnosis model according to the selected combined operation mode corresponding to the analysis precision, if the key feature is not identified, not outputting the diagnosis result, if the key feature is identified, outputting the probability that the key feature belongs to the accident reason of the class corresponding to the current accident reason diagnosis model, and taking the video segment and screenshot corresponding to the key feature as the diagnosis result.
In an optional embodiment of the invention, a unidirectional video data transmission channel can be established between the video data reading layer and the video monitoring system of the target dangerous chemical production and operation enterprise, so that the real-time video image data in the target video monitoring system can be transmitted to the video data reading layer in real time, and then subsequent video analysis is performed. After the operation, once an accident occurs in a target dangerous chemical production and operation enterprise, a diagnosis result can be inquired in a data storage layer of the system at the first time, and the accident reason can be obtained at the first time.
In an alternative embodiment of the present invention, the system may be configured with an LED display screen that displays the results of the diagnostics in the data storage layer.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An accident cause offline intelligent diagnosis system is characterized by comprising a video data reading layer, a central processing layer, a data storage layer and a key feature recognition algorithm model library;
the key feature recognition algorithm model base stores a plurality of key feature recognition algorithm models, the key feature recognition algorithm models are deep learning neural network models and are trained on the corresponding chemical accident image data acquired from the opening source data set;
the video data reading layer is used for acquiring monitoring video image data;
the central processing layer is used for selecting a key feature recognition algorithm model of a corresponding class from the key feature recognition algorithm model library as a current accident reason diagnosis model, inputting the monitoring video image data into the current accident reason diagnosis model, diagnosing the monitoring video image data by the current accident reason diagnosis model, if no key feature is recognized, not outputting a diagnosis result, if the key feature is recognized, outputting the probability that the key feature belongs to the accident reason of the class corresponding to the current accident reason diagnosis model, and taking a video segment and a screenshot corresponding to the key feature as the diagnosis result;
the data storage layer is used for storing the output diagnosis result.
2. The offline intelligent diagnosis system for accident reasons according to claim 1, wherein said key feature recognition algorithm model base stores: the system comprises a personnel gathering key feature recognition algorithm model, an illegal break-in key feature recognition algorithm model, an off duty key feature recognition algorithm model in a duty room, a fire fighting channel blockage key feature recognition algorithm model, an environmental fire key feature recognition algorithm model and an environmental smoking key feature recognition algorithm model.
3. The offline intelligent diagnosis system for accident causes according to claim 1, wherein the opening data sets are ImageNet and/or Pascal VOC.
4. The accident cause offline intelligent diagnosis system according to claim 1, wherein the key feature recognition algorithm model comprises a convolutional neural network model and a plurality of classifiers, the classification precision of the classifiers is adjustable, the higher the precision of the classification operation, the lower the recognition rate, and vice versa; for each key feature recognition algorithm model, the central processing layer can select classification operation with corresponding precision for diagnosing the monitoring video image data.
5. The offline intelligent diagnosis system for accident causes according to claim 4, wherein the central processing layer comprises a condition selection window, and the window is provided with an analysis precision selection window and a key feature recognition algorithm model selection window.
6. The offline intelligent diagnosis system for accident causes according to claim 4, wherein the training method of the key feature recognition algorithm model comprises convolutional neural network model training and classifier training.
7. The offline intelligent diagnosis system for accident causes according to claim 6, wherein the training process of the key feature recognition algorithm model comprises: training the convolutional neural network model by using the chemical accident image data acquired from the open source data set, so that the convolutional neural network model can identify key features in the chemical accident image data; adopting a trained convolutional neural network model to perform image CNN feature extraction on chemical accident image data acquired from an opening source data set, and constructing a key feature training set; and training the classifier by adopting a key characteristic training set.
8. The offline intelligent diagnosis system for accident causes according to claim 7, wherein the chemical accident image data obtained from the source data set are all data labeled by using a LabelImg tool.
9. The offline intelligent diagnosis system for accident reasons according to claim 8, wherein the number of the chemical accident image data obtained from the opening source data set is more than 2000 sheets for each type.
10. The offline intelligent diagnosis system for accident cause according to claim 9, wherein the format of said surveillance video image data is MP4/M4V/FLV/MOV/WMV.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210980387.6A CN115205761A (en) | 2022-08-16 | 2022-08-16 | Accident reason off-line intelligent diagnosis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210980387.6A CN115205761A (en) | 2022-08-16 | 2022-08-16 | Accident reason off-line intelligent diagnosis system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115205761A true CN115205761A (en) | 2022-10-18 |
Family
ID=83585553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210980387.6A Pending CN115205761A (en) | 2022-08-16 | 2022-08-16 | Accident reason off-line intelligent diagnosis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115205761A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116311081A (en) * | 2023-05-12 | 2023-06-23 | 天津医科大学 | Medical laboratory monitoring image analysis method and system based on image recognition |
-
2022
- 2022-08-16 CN CN202210980387.6A patent/CN115205761A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116311081A (en) * | 2023-05-12 | 2023-06-23 | 天津医科大学 | Medical laboratory monitoring image analysis method and system based on image recognition |
| CN116311081B (en) * | 2023-05-12 | 2023-08-22 | 天津医科大学 | Medical laboratory monitoring image analysis method and system based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111475804B (en) | Alarm prediction method and system | |
| Cameron et al. | Emergency situation awareness from twitter for crisis management | |
| CN108228705B (en) | Automatic object and activity tracking device, method and medium in live video feedback | |
| EP3382643A1 (en) | Automated object tracking in a video feed using machine learning | |
| CN110659397A (en) | Behavior detection method and device, electronic equipment and storage medium | |
| KR102028930B1 (en) | method of providing categorized video processing for moving objects based on AI learning using moving information of objects | |
| CN111370113A (en) | Remote psychological counseling system and method based on Internet of things cloud | |
| US11776271B2 (en) | Systems and methods for creating a story board with forensic video analysis on a video repository | |
| US20230186634A1 (en) | Vision-based monitoring of site safety compliance based on worker re-identification and personal protective equipment classification | |
| CN115205761A (en) | Accident reason off-line intelligent diagnosis system | |
| CN113128412B (en) | Fire trend prediction method based on deep learning and fire monitoring video | |
| CN111611923B (en) | Emergency plan drilling evaluation method and device | |
| CN112686180A (en) | Method for calculating number of personnel in closed space | |
| CN115331346B (en) | Campus access control management method and device, electronic equipment and storage medium | |
| CN115457462A (en) | Epidemic prevention supervision method and device for realizing IA by combining RPA and AI and electronic equipment | |
| CN114882589A (en) | Granary safety operation early warning system and method based on intelligent video analysis | |
| CN112950438A (en) | Data processing method and device, computer equipment and storage medium | |
| CN113012006A (en) | Intelligent investigation and research method, system, computer equipment and storage medium | |
| CN112507972A (en) | Performance assessment system based on block chain | |
| US20230410516A1 (en) | Information acquisition support apparatus, information acquisition support method, and recording medium storing information acquisition support program | |
| CN114841656B (en) | Military aircraft fault detection method and system based on edge calculation | |
| CN116204670B (en) | Management method and system of vehicle target detection data and electronic equipment | |
| CN113095110B (en) | Method, device, medium and electronic equipment for dynamically warehousing face data | |
| CN115545969A (en) | Data summarization processing and safety management report system under power system | |
| Fan et al. | Design of Intelligent Human Resource Information Collection System Based on Decision Tree Algorithm |
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 |