CN117911949B - Risk intelligent assessment method and system - Google Patents
Risk intelligent assessment method and system Download PDFInfo
- Publication number
- CN117911949B CN117911949B CN202410079510.6A CN202410079510A CN117911949B CN 117911949 B CN117911949 B CN 117911949B CN 202410079510 A CN202410079510 A CN 202410079510A CN 117911949 B CN117911949 B CN 117911949B
- Authority
- CN
- China
- Prior art keywords
- distribution
- personnel density
- semantic
- personnel
- density
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000013598 vector Substances 0.000 claims abstract description 109
- 239000011159 matrix material Substances 0.000 claims abstract description 72
- 238000012544 monitoring process Methods 0.000 claims abstract description 45
- 238000012549 training Methods 0.000 claims description 65
- 238000005259 measurement Methods 0.000 claims description 33
- 238000013527 convolutional neural network Methods 0.000 claims description 24
- 238000003062 neural network model Methods 0.000 claims description 12
- 230000000903 blocking effect Effects 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 4
- 238000012502 risk assessment Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 15
- 238000011156 evaluation Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000005457 optimization Methods 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000004044 response 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
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- 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
- G06N3/042—Knowledge-based neural networks; Logical representations of neural networks
-
- 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
- G06N3/045—Combinations of networks
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
- G06V10/443—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by matching or filtering
- G06V10/449—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters
- G06V10/451—Biologically inspired filters, e.g. difference of Gaussians [DoG] or Gabor filters with interaction between the filter responses, e.g. cortical complex cells
- G06V10/454—Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- 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
- G06V10/774—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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Resources & Organizations (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Educational Administration (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Biodiversity & Conservation Biology (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a risk intelligent assessment method and a risk intelligent assessment system, which relate to the technical field of intelligent assessment and are used for acquiring a personnel density monitoring image of a preset area acquired by a camera; extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors; constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas; and determining crowd distribution density early warning grades based on graph structural global association features between the sequence of the personnel density local distribution semantic feature vectors and the topological association feature matrix among the personnel density distribution areas. Therefore, the personnel distribution and density conditions can be known in time, and possible risks can be early warned in advance so as to make relevant safety countermeasures in time.
Description
Technical Field
The application relates to the technical field of intelligent assessment, in particular to an intelligent risk assessment method and system.
Background
Along with the continuous promotion of the urban process, the cities often gather a large number of people, and especially, the mobile people are relatively dense in public places such as markets, supermarkets, subways and the like. In order to prevent various security risks caused by crowds, a monitoring system is generally arranged in a public place similar to the above, so as to obtain the crowd distribution situation in time. Crowd distribution density is one of the important indicators for assessing security risk. Currently, some existing methods for evaluating crowd distribution density have some problems. For example, after the flow condition of personnel is obtained through the monitoring system, the level of the security risk is generally judged according to human experience, and the judgment mode has the problems of subjectivity and unilateral property, so that misjudgment or untimely response measures are easy to cause.
Therefore, an optimized risk intelligent assessment method and system are desired.
Disclosure of Invention
The present application has been made to solve the above-mentioned technical problems. The application provides a risk intelligent assessment method and a risk intelligent assessment system, which are used for acquiring a personnel density monitoring image of a preset area acquired by a camera; extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors; constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas; and determining crowd distribution density early warning grades based on graph structural global association features between the sequence of the personnel density local distribution semantic feature vectors and the topological association feature matrix among the personnel density distribution areas. Therefore, the safety risk situation can be known in time, and potential safety problems can be pre-warned and dealt with in advance.
In a first aspect, a risk intelligent assessment method is provided, which includes:
Acquiring a personnel density monitoring image of a preset area acquired by a camera;
extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors;
Constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas;
And determining crowd distribution density early warning grades based on graph structural global association features between the sequence of the personnel density local distribution semantic feature vectors and the topological association feature matrix among the personnel density distribution areas.
In a second aspect, there is provided a risk intelligent assessment system, comprising:
The monitoring image acquisition module is used for acquiring a personnel density monitoring image of a preset area acquired by the camera;
the local image semantic feature extraction module is used for extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors;
The local image semantic difference topological feature construction module is used for constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors so as to obtain a topological association feature matrix among personnel density distribution areas;
The crowd distribution density early warning level determining module is used for determining crowd distribution density early warning levels based on the sequence of the local distribution semantic feature vectors of the crowd distribution and the graph structural global association features between the topological association feature matrices among the crowd distribution areas.
Compared with the prior art, the risk intelligent assessment method and system provided by the application acquire the personnel density monitoring image of the preset area acquired by the camera; extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors; constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas; and determining crowd distribution density early warning grades based on graph structural global association features between the sequence of the personnel density local distribution semantic feature vectors and the topological association feature matrix among the personnel density distribution areas. Therefore, the risk situation of the personnel gathering place can be timely, and the possible safety risk can be early warned and dealt with in advance, so that relevant safety countermeasures can be timely taken.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a risk intelligent assessment method according to an embodiment of the present application.
Fig. 2 is a schematic diagram of a risk intelligent assessment method according to an embodiment of the application.
FIG. 3 is a block diagram of a risk intelligent assessment system according to an embodiment of the present application.
Fig. 4 is a schematic view of a scenario of a risk intelligent assessment method according to an embodiment of the present application.
Detailed Description
The following description of the technical solutions according to the embodiments of the present application will be given with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
The population distribution density is one of important indexes for evaluating safety risk, and is the ratio of the population quantity in a specific area to the total area of the area, and reflects the aggregation degree of population and the intensity of social activities. The crowd distribution density is closely related to the safety risk, and in general, the higher the crowd distribution density is, the larger the safety risk is, and meanwhile, the safety measures to be dealt with are also considered.
Therefore, the population distribution density of different areas is analyzed and evaluated to know the population aggregation degree and the activity intensity, so that the possible safety problems and risks are predicted and evaluated, corresponding decision support and risk management measures are made, and the safety of all aspects is ensured.
However, the conventional risk assessment method has some problems in assessing crowd distribution density. The traditional evaluation method often depends on opinion and experience judgment of experts, the evaluation result is easily influenced by subjective factors, different experts may have different opinions and judgment, subjectivity and one-sided performance of the evaluation result are caused, and inaccurate or deviated evaluation of crowd distribution density may be caused.
The traditional evaluation method lacks objective data support for crowd distribution density, the crowd distribution density is usually acquired through means of census, satellite remote sensing and the like, but the data are not easy to obtain or update, and inaccurate evaluation results may be caused by lack of accurate data support. Traditional evaluation methods often lack consideration of dynamic changes and timeliness of crowd distribution density, which may change with time and environmental changes, but traditional evaluation methods often do not update and reflect these changes in time, which may result in the evaluation results not reflecting the current security risk in time.
To address these issues, new evaluation methods and techniques are being introduced. For example, more accurate and real-time crowd distribution density data can be obtained by using a remote sensing technology and a geographic information system, and meanwhile, more objective and comprehensive assessment results can be provided by applying data mining and machine learning technologies. In addition, multidisciplinary research and cross-department cooperation are also key to solving the problem of crowd distribution density evaluation, and the influence of crowd distribution density on safety and stability can be more comprehensively evaluated by integrating expertise and data resources in different fields.
Some existing methods for evaluating crowd distribution density have problems such as subjectivity and unilaterality, lack of objective data support and lack of dynamic property and timeliness. To solve these problems, new assessment methods and techniques need to be introduced, and multidisciplinary research and cross-sector cooperation is enhanced, which will help to improve accuracy and timeliness of crowd distribution density assessment, thereby better assessing safety and stability risks.
Fig. 1 is a flowchart of a risk intelligent assessment method according to an embodiment of the present application. Fig. 2 is a schematic diagram of a risk intelligent assessment method according to an embodiment of the application. As shown in fig. 1 and 2, the risk intelligent assessment method includes: 110, acquiring a personnel density monitoring image of a predetermined area acquired by a camera; 120, extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors; 130, constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas; 140, determining crowd distribution density early warning levels based on graph structural global association features between the sequence of the personnel density local distribution semantic feature vectors and the topological association feature matrix between the personnel density distribution areas.
Aiming at the technical problems, the technical conception of the application is to introduce an artificial intelligence technology based on deep learning, process and analyze the image of the personnel density monitoring image, mine the crowd density distribution semantic information contained in the image, and intelligently judge the crowd density early warning level based on the crowd density distribution semantic information. Therefore, the safety risk situation can be known in time, and potential safety problems can be pre-warned and dealt with in advance.
Based on the above, in the technical scheme of the application, firstly, a personnel density monitoring image of a preset area acquired by a camera is acquired; and performing image blocking processing on the personnel density monitoring image to obtain a sequence of personnel density local distribution image blocks. The personnel density monitoring image is subjected to image blocking, so that fine granularity analysis can be performed on the personnel density monitoring image. That is, by dividing the personnel density monitoring image into small blocks, the personnel density distribution conditions of different local areas can be more accurately captured, and the characteristic expression that the difference between the personnel density distribution characteristics of each local area affects the overall personnel density is avoided.
In a specific embodiment of the present application, extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors includes: performing image blocking processing on the personnel density monitoring image to obtain a sequence of personnel density local distribution image blocks; and the sequence of the personnel density local distribution image blocks passes through a personnel density distribution semantic feature extractor based on a convolutional neural network model to obtain the sequence of the personnel density local distribution semantic feature vectors.
And then, passing the sequence of the personnel density local distribution image blocks through a personnel density distribution semantic feature extractor based on a convolutional neural network model to obtain a sequence of personnel density local distribution semantic feature vectors. That is, the semantic feature extractor for the distribution of the personnel density based on the convolutional neural network model extracts the semantic information of the distribution of the crowd, such as the shape of the crowd gathering, the spatial relationship among the personnel, and the like, contained in each of the local distribution image blocks of the personnel density from the sequence of the local distribution image blocks of the personnel density.
Then, calculating the personnel density distribution difference semantic measurement coefficients between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas consisting of a plurality of personnel density distribution difference semantic measurement coefficients; and the topological correlation matrix among the personnel density distribution areas passes through a topological feature extractor based on a convolutional neural network model to obtain the topological correlation feature matrix among the personnel density distribution areas. That is, in the technical scheme of the application, the spatial relevance and the degree of difference of the personnel density distribution in each local area are quantified and measured by calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors. Thus, the crowd density distribution variation among the local areas can be described and depicted. And constructing the topological feature extractor by utilizing the convolutional neural network model to capture implicit topological association relations contained in the topological association matrix among the personnel density distribution areas.
In a specific embodiment of the present application, constructing the local image semantic difference topological feature of the sequence of the personnel density local distribution semantic feature vector to obtain a topological association feature matrix between personnel density distribution areas, including: calculating the personnel density distribution difference semantic measurement coefficients between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas, wherein the personnel density distribution difference semantic measurement coefficients consist of a plurality of personnel density distribution difference semantic measurement coefficients; and the topological correlation matrix among the personnel density distribution areas passes through a topological feature extractor based on a convolutional neural network model to obtain the topological correlation feature matrix among the personnel density distribution areas.
More specifically, calculating a density distribution difference semantic metric coefficient between any two density local distribution semantic feature vectors in the sequence of density local distribution semantic feature vectors to obtain a density distribution inter-region topological correlation matrix composed of a plurality of density distribution difference semantic metric coefficients, including: calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors according to the following difference semantic measurement formula; wherein, the difference semantic measurement formula is: Wherein/> For the previous said people density local distribution semantic feature vector,For the latter of said people density local distribution semantic feature vectors,/>For each of said personnel density local distribution semantic feature vector dimensions,/>Semantic measurement coefficients for the personnel density distribution difference,/>A logarithmic function operation with a base of 2 is represented. And further, the sequence of the personnel density local distribution semantic feature vector and the topological association feature matrix among the personnel density distribution areas are subjected to a graph neural network model to obtain a personnel density distribution global feature matrix containing the topological information among the personnel density distribution areas. The characteristic representation of the nodes is the local distribution semantic characteristic vector of each personnel density, and the characteristic representation of the edges between the nodes is the topological association characteristic matrix between the personnel density distribution areas, so that the crowd distribution density characteristic information and the global density topological association information of each local area are fused, and the personnel density distribution global characteristic matrix containing the topological information between the personnel density distribution areas is obtained. Specifically, the graph neural network model can interact and update the local features and the topology association features in a manner of transmitting information, so that a more global feature representation is learned. The personnel density distribution characteristics of the predetermined area can be more comprehensively evaluated and judged, and the personnel density distribution characteristics comprise the whole structure of personnel density distribution and the spatial association relation between the local areas.
In a specific embodiment of the present application, determining a crowd distribution density early warning level based on a graph structural global correlation feature between the sequence of the crowd density local distribution semantic feature vectors and the topological correlation feature matrix between the crowd density distribution areas includes: the sequence of the personnel density local distribution semantic feature vector and the topological association feature matrix among the personnel density distribution areas are processed through a graph neural network model to obtain a personnel density distribution global feature matrix containing topological information among the personnel density distribution areas; and the personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas passes through a classifier to obtain a classification result, wherein the classification result is used for representing crowd distribution density early warning grade labels.
And then, the personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas passes through a classifier to obtain a classification result, wherein the classification result is used for representing the crowd distribution density early warning grade label.
In a specific embodiment of the present application, the global feature matrix of the people density distribution including topology information between people density distribution areas is passed through a classifier to obtain a classification result, where the classification result is used to represent a crowd distribution density early warning level label, and the method includes: expanding the personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas into classification feature vectors according to row vectors or column vectors; performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors; and passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
In one embodiment of the present application, the risk intelligent assessment method further includes a training step: training the personnel density distribution semantic feature extractor based on the convolutional neural network model, the topological feature extractor based on the convolutional neural network model, the graph neural network model and the classifier. The training step comprises the following steps: acquiring training data, wherein the training data comprises training personnel density monitoring images of a preset area acquired by a camera and the true value of crowd distribution density early warning grade labels; performing image blocking processing on the training personnel density monitoring image to obtain a sequence of training personnel density local distribution image blocks; passing the sequence of the training personnel density local distribution image blocks through the personnel density distribution semantic feature extractor based on the convolutional neural network model to obtain a sequence of training personnel density local distribution semantic feature vectors; calculating the personnel density distribution difference semantic measurement coefficients between any two training personnel density local distribution semantic feature vectors in the sequence of the training personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among training personnel density distribution areas consisting of a plurality of personnel density distribution difference semantic measurement coefficients; the topological correlation matrix among the density distribution areas of the training staff passes through the topological feature extractor based on the convolutional neural network model to obtain the topological correlation feature matrix among the density distribution areas of the training staff; the training personnel density local distribution semantic feature vector sequence and the training personnel density distribution inter-region topological association feature matrix are passed through the graph neural network model to obtain a training personnel density distribution global feature matrix containing the personnel density distribution inter-region topological information; the training personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas passes through a classifier to obtain a classification loss function value; calculating a specific loss function value of the training personnel density distribution global feature matrix containing topology information among personnel density distribution areas and the sequence of the personnel density local distribution semantic feature vectors; training the personnel density distribution semantic feature extractor based on the convolutional neural network model, the topological feature extractor based on the convolutional neural network model, the graph neural network model and the classifier by taking the weighted sum of the classified loss function value and the specific loss function value as the loss function value.
In the technical scheme of the application, each training personnel density local distribution semantic feature vector in the training personnel density local distribution semantic feature vector sequence expresses the image semantic feature of the training personnel density monitoring image under the local image semantic space domain determined by image segmentation in the global image semantic space domain, and after the training personnel density local distribution semantic feature vector sequence and the training personnel density distribution regional topological association feature matrix pass through a graph neural network model, the image semantic feature under the local image semantic space domain can be subjected to topological association based on the local image semantic space domain semantic difference measurement topology, but the training personnel density distribution global feature matrix containing the personnel density distribution regional topological information has different feature group density representation relative to the training personnel density local distribution semantic feature vector sequence, so that iteration imbalance exists between the image semantic feature extraction of the convolutional neural network model and the topological association of the graph neural network model during model integral training, and the integral training efficiency of the model is affected.
Therefore, the consistency of the feature group density representation of the training personnel density distribution global feature matrix containing the topological information among the personnel density distribution areas relative to the sequence of the training personnel density local distribution semantic feature vectors is considered to be improved, so that a specific loss function aiming at the training personnel density distribution global feature matrix containing the topological information among the personnel density distribution areas and the sequence of the training personnel density local distribution semantic feature vectors is further introduced, and the specific loss function is expressed as: calculating a specific loss function value of the training personnel density distribution global feature matrix containing topology information among personnel density distribution areas and the sequence of the personnel density local distribution semantic feature vector according to the following optimization formula; wherein, the optimization formula is: Wherein/> Is the first eigenvector of the training personnel density distribution global eigenvector which contains the topological information among personnel density distribution areas after being unfoldedIs the second feature vector after the sequence cascade of the training personnel density local distribution semantic feature vector,/>Is the length of the feature vector, and/>Representing the square of the two norms of the vector,/>Is the eigenvalue of the first eigenvector,/>Is the eigenvalue of the second eigenvector,/>Is the specific loss function value,/>Representing the calculation of a value of a natural exponential function raised to a power by a value,/>Representing per-position subtraction. Here, the specific loss function performs group count attention based on feature group density by performing adaptive attention of different density expression patterns between the training person density distribution global feature matrix containing topology information between person density distribution areas and the sequence of training person density local distribution semantic feature vectors by recursively mapping group count as output feature group density. By taking the model as a loss function to train the model, the model can avoid overestimation and underestimation aiming at different density modes under the characteristic distribution of the training personnel density distribution global characteristic matrix containing the topological information among personnel density distribution areas and the training personnel density local distribution semantic characteristic vector sequence, and learn the corresponding relation between the characteristic value distribution and the group density distribution, thereby realizing the consistency optimization of the characteristic group density representation among the training personnel density distribution global characteristic matrix containing the topological information among personnel density distribution areas and the training personnel density local distribution semantic characteristic vector sequence with different characteristic densities, and improving the overall training efficiency of the model.
In summary, the risk intelligent assessment method based on the embodiment of the application is clarified, an artificial intelligent technology based on deep learning is introduced, image processing and analysis are carried out on the personnel density monitoring image, crowd density distribution semantic information contained in the personnel density monitoring image is mined, and the crowd density early warning level is intelligently judged based on the crowd density semantic information. Therefore, the risk situation of the area with high personnel density can be known in time, early warning is carried out in advance, and potential safety problems can be solved.
In one embodiment of the application, FIG. 3 is a block diagram of a risk intelligent assessment system according to an embodiment of the application. As shown in fig. 3, a risk intelligent assessment system 200 according to an embodiment of the present application includes: a monitoring image acquisition module 210 for acquiring a person density monitoring image of a predetermined area acquired by the camera; the local image semantic feature extraction module 220 is configured to extract local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors; the local image semantic difference topological feature construction module 230 is configured to construct local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix between personnel density distribution areas; the crowd distribution density early warning level determining module 240 is configured to determine a crowd distribution density early warning level based on a graph structural global association feature between the sequence of the local crowd distribution semantic feature vectors and the topological association feature matrix between the crowd distribution areas.
In the risk intelligent assessment system, the local image semantic feature extraction module includes: the image blocking unit is used for carrying out image blocking processing on the personnel density monitoring image to obtain a sequence of personnel density local distribution image blocks; and the personnel density distribution semantic feature extraction unit is used for enabling the sequence of the personnel density local distribution image blocks to pass through a personnel density distribution semantic feature extractor based on a convolutional neural network model so as to obtain the sequence of the personnel density local distribution semantic feature vectors.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described risk intelligent assessment system have been described in detail in the above description of the risk intelligent assessment method with reference to fig. 1 to 2, and thus, repetitive descriptions thereof will be omitted.
As described above, the risk intelligent assessment system 200 according to the embodiment of the present application may be implemented in various terminal devices, such as a server for risk intelligent assessment, and the like. In one example, the risk intelligent assessment system 200 according to an embodiment of the present application may be integrated into the terminal device as a software module and/or hardware module. For example, the risk intelligent assessment system 200 can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the risk intelligent assessment system 200 can equally be one of a number of hardware modules of the terminal device.
Alternatively, in another example, the risk intelligent assessment system 200 and the terminal device may be separate devices, and the risk intelligent assessment system 200 may be connected to the terminal device via a wired and/or wireless network and transmit the interaction information in a agreed data format.
Fig. 4 is a schematic view of a scenario of a risk intelligent assessment method according to an embodiment of the present application. As shown in fig. 4, in the application scene, first, a person density monitoring image of a predetermined area acquired by a camera is acquired (e.g., C as illustrated in fig. 4); the acquired people density monitoring image is then input into a server (e.g., S as illustrated in fig. 4) deployed with a risk intelligent assessment algorithm, wherein the server is capable of processing the people density monitoring image based on the risk intelligent assessment algorithm to determine a crowd distribution density pre-warning level.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (6)
1. The intelligent risk assessment method is characterized by comprising the following steps of:
Acquiring a personnel density monitoring image of a preset area acquired by a camera;
extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors;
Constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors to obtain a topological association feature matrix among personnel density distribution areas;
Determining crowd distribution density early warning levels based on the sequence of the personnel density local distribution semantic feature vectors and graph structural global association features between the topological association feature matrices among the personnel density distribution areas;
the step of extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors comprises the following steps:
Performing image blocking processing on the personnel density monitoring image to obtain a sequence of personnel density local distribution image blocks;
The sequence of the personnel density local distribution image blocks passes through a personnel density distribution semantic feature extractor based on a convolutional neural network model to obtain a sequence of the personnel density local distribution semantic feature vectors;
the step of constructing the local image semantic difference topological feature of the sequence of the personnel density local distribution semantic feature vector to obtain a topological association feature matrix among personnel density distribution areas comprises the following steps:
Calculating the personnel density distribution difference semantic measurement coefficients between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas, wherein the personnel density distribution difference semantic measurement coefficients consist of a plurality of personnel density distribution difference semantic measurement coefficients;
the topological correlation matrix among the personnel density distribution areas passes through a topological feature extractor based on a convolutional neural network model to obtain the topological correlation feature matrix among the personnel density distribution areas;
The step of calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas consisting of a plurality of personnel density distribution difference semantic measurement coefficients comprises the following steps:
calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors according to the following difference semantic measurement formula; wherein, the difference semantic measurement formula is: ; wherein/> For the former said people density local distribution semantic feature vector,/>For the latter of said people density local distribution semantic feature vectors,/>For each of said personnel density local distribution semantic feature vector dimensions,/>Semantic measurement coefficients for the personnel density distribution difference,/>A logarithmic function operation with a base of 2 is represented.
2. The risk intelligent assessment method according to claim 1, wherein determining crowd distribution density pre-warning levels based on graph-structured global correlation features between the sequence of people density local distribution semantic feature vectors and the inter-people density distribution region topological correlation feature matrix comprises:
The sequence of the personnel density local distribution semantic feature vector and the topological association feature matrix among the personnel density distribution areas are processed through a graph neural network model to obtain a personnel density distribution global feature matrix containing topological information among the personnel density distribution areas;
And the personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas passes through a classifier to obtain a classification result, wherein the classification result is used for representing crowd distribution density early warning grade labels.
3. The risk intelligent assessment method according to claim 2, wherein the global feature matrix of the people density distribution including the topology information among the people density distribution areas is passed through a classifier to obtain a classification result, and the classification result is used for representing a crowd distribution density early warning grade label, and the method comprises the following steps:
expanding the personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas into classification feature vectors according to row vectors or column vectors;
performing full-connection coding on the classification feature vectors by using a plurality of full-connection layers of the classifier to obtain coded classification feature vectors;
And passing the coding classification feature vector through a Softmax classification function of the classifier to obtain the classification result.
4. The risk intelligent assessment method according to claim 3, further comprising a training step of: training the personnel density distribution semantic feature extractor based on the convolutional neural network model, the topological feature extractor based on the convolutional neural network model, the graph neural network model and the classifier.
5. The method of claim 4, wherein the training step comprises:
acquiring training data, wherein the training data comprises training personnel density monitoring images of a preset area acquired by a camera and the true value of crowd distribution density early warning grade labels;
Performing image blocking processing on the training personnel density monitoring image to obtain a sequence of training personnel density local distribution image blocks;
Passing the sequence of the training personnel density local distribution image blocks through the personnel density distribution semantic feature extractor based on the convolutional neural network model to obtain a sequence of training personnel density local distribution semantic feature vectors;
calculating the personnel density distribution difference semantic measurement coefficients between any two training personnel density local distribution semantic feature vectors in the sequence of the training personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among training personnel density distribution areas consisting of a plurality of personnel density distribution difference semantic measurement coefficients;
The topological correlation matrix among the density distribution areas of the training staff passes through the topological feature extractor based on the convolutional neural network model to obtain the topological correlation feature matrix among the density distribution areas of the training staff;
The training personnel density local distribution semantic feature vector sequence and the training personnel density distribution inter-region topological association feature matrix are passed through the graph neural network model to obtain a training personnel density distribution global feature matrix containing the personnel density distribution inter-region topological information;
The training personnel density distribution global feature matrix containing the topology information among the personnel density distribution areas passes through a classifier to obtain a classification loss function value;
Calculating a specific loss function value of the training personnel density distribution global feature matrix containing topology information among personnel density distribution areas and the sequence of the personnel density local distribution semantic feature vectors;
Training the personnel density distribution semantic feature extractor based on the convolutional neural network model, the topological feature extractor based on the convolutional neural network model, the graph neural network model and the classifier by taking the weighted sum of the classified loss function value and the specific loss function value as the loss function value.
6. A risk intelligent assessment system, comprising:
The monitoring image acquisition module is used for acquiring a personnel density monitoring image of a preset area acquired by the camera;
the local image semantic feature extraction module is used for extracting the local image semantic features of the personnel density monitoring image to obtain a sequence of personnel density local distribution semantic feature vectors;
The local image semantic difference topological feature construction module is used for constructing local image semantic difference topological features of the sequence of the personnel density local distribution semantic feature vectors so as to obtain a topological association feature matrix among personnel density distribution areas;
The crowd distribution density early warning level determining module is used for determining crowd distribution density early warning levels based on the sequence of the local distribution semantic feature vectors of the crowd density and the graph structural global association features between the topological association feature matrices of the crowd density distribution areas;
The local image semantic feature extraction module comprises:
the image blocking unit is used for carrying out image blocking processing on the personnel density monitoring image to obtain a sequence of personnel density local distribution image blocks;
the personnel density distribution semantic feature extraction unit is used for enabling the sequence of the personnel density local distribution image blocks to pass through a personnel density distribution semantic feature extractor based on a convolutional neural network model so as to obtain a sequence of the personnel density local distribution semantic feature vectors;
The local image semantic difference topological feature construction module comprises:
Calculating the personnel density distribution difference semantic measurement coefficients between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas, wherein the personnel density distribution difference semantic measurement coefficients consist of a plurality of personnel density distribution difference semantic measurement coefficients;
the topological correlation matrix among the personnel density distribution areas passes through a topological feature extractor based on a convolutional neural network model to obtain the topological correlation feature matrix among the personnel density distribution areas;
The step of calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors to obtain a topological correlation matrix among personnel density distribution areas consisting of a plurality of personnel density distribution difference semantic measurement coefficients comprises the following steps:
calculating the personnel density distribution difference semantic measurement coefficient between any two personnel density local distribution semantic feature vectors in the sequence of the personnel density local distribution semantic feature vectors according to the following difference semantic measurement formula; wherein, the difference semantic measurement formula is: ; wherein/> For the former said people density local distribution semantic feature vector,/>For the latter of said people density local distribution semantic feature vectors,/>For each of said personnel density local distribution semantic feature vector dimensions,/>Semantic measurement coefficients for the personnel density distribution difference,/>A logarithmic function operation with a base of 2 is represented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410079510.6A CN117911949B (en) | 2024-01-19 | 2024-01-19 | Risk intelligent assessment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410079510.6A CN117911949B (en) | 2024-01-19 | 2024-01-19 | Risk intelligent assessment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117911949A CN117911949A (en) | 2024-04-19 |
| CN117911949B true CN117911949B (en) | 2024-06-11 |
Family
ID=90694190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410079510.6A Active CN117911949B (en) | 2024-01-19 | 2024-01-19 | Risk intelligent assessment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117911949B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118940930B (en) * | 2024-10-14 | 2025-02-14 | 泰安市园林绿化管理服务中心 | A smart garden management method, device, system and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101739569A (en) * | 2009-12-17 | 2010-06-16 | 北京中星微电子有限公司 | Crowd density estimation method, device and monitoring system |
| CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
| CN101976353A (en) * | 2010-10-28 | 2011-02-16 | 北京智安邦科技有限公司 | Statistical method and device of low density crowd |
| CN103745230A (en) * | 2014-01-14 | 2014-04-23 | 四川大学 | Adaptive abnormal crowd behavior analysis method |
| CN108171223A (en) * | 2018-02-27 | 2018-06-15 | 北京中晟信达科技有限公司 | A kind of face identification method and system based on multi-model multichannel |
| CN110651310A (en) * | 2017-04-05 | 2020-01-03 | 卡内基梅隆大学 | Deep learning method for estimating object density and/or flow and related method and software |
| CN210072642U (en) * | 2019-06-20 | 2020-02-14 | 马国祥 | Crowd abnormal behavior detection system based on video monitoring |
| CN111144377A (en) * | 2019-12-31 | 2020-05-12 | 北京理工大学 | Dense area early warning method based on crowd counting algorithm |
| CN117270611A (en) * | 2023-11-22 | 2023-12-22 | 浙江威星电子系统软件股份有限公司 | Intelligent environmental control system and method for intelligent sports stadium |
| WO2024015632A1 (en) * | 2022-07-15 | 2024-01-18 | Nvidia Corporation | Surround scene perception using multiple sensors for autonomous systems and applications |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11386655B2 (en) * | 2019-02-15 | 2022-07-12 | Rutgers, The State University Of New Jersey | Image processing neural network systems and methods with scene understanding |
-
2024
- 2024-01-19 CN CN202410079510.6A patent/CN117911949B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101753992A (en) * | 2008-12-17 | 2010-06-23 | 深圳市先进智能技术研究所 | Multi-mode intelligent monitoring system and method |
| CN101739569A (en) * | 2009-12-17 | 2010-06-16 | 北京中星微电子有限公司 | Crowd density estimation method, device and monitoring system |
| CN101976353A (en) * | 2010-10-28 | 2011-02-16 | 北京智安邦科技有限公司 | Statistical method and device of low density crowd |
| CN103745230A (en) * | 2014-01-14 | 2014-04-23 | 四川大学 | Adaptive abnormal crowd behavior analysis method |
| CN110651310A (en) * | 2017-04-05 | 2020-01-03 | 卡内基梅隆大学 | Deep learning method for estimating object density and/or flow and related method and software |
| CN108171223A (en) * | 2018-02-27 | 2018-06-15 | 北京中晟信达科技有限公司 | A kind of face identification method and system based on multi-model multichannel |
| CN210072642U (en) * | 2019-06-20 | 2020-02-14 | 马国祥 | Crowd abnormal behavior detection system based on video monitoring |
| CN111144377A (en) * | 2019-12-31 | 2020-05-12 | 北京理工大学 | Dense area early warning method based on crowd counting algorithm |
| WO2024015632A1 (en) * | 2022-07-15 | 2024-01-18 | Nvidia Corporation | Surround scene perception using multiple sensors for autonomous systems and applications |
| CN117270611A (en) * | 2023-11-22 | 2023-12-22 | 浙江威星电子系统软件股份有限公司 | Intelligent environmental control system and method for intelligent sports stadium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117911949A (en) | 2024-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110458195B (en) | Abnormal data discrimination method based on multi-criterion fusion | |
| CN111008337B (en) | A deep attention rumor identification method and device based on ternary features | |
| CN109508360A (en) | A kind of polynary flow data space-time autocorrelation analysis method of geography based on cellular automata | |
| CN109493119B (en) | POI data-based urban business center identification method and system | |
| CN118470550B (en) | A natural resource asset data collection method and platform | |
| CN109657697A (en) | Classified optimization method based on semi-supervised learning and fine granularity feature learning | |
| CN112700056A (en) | Complex network link prediction method, complex network link prediction device, electronic equipment and medium | |
| CN109117883A (en) | SAR image sea ice classification method and system based on long memory network in short-term | |
| CN115034600A (en) | Early warning method and system for geological disaster monitoring | |
| CN116610816A (en) | Personnel portrait knowledge graph analysis method and system based on graph convolution neural network | |
| CN118114827B (en) | Flood disaster forecasting and early warning method and system for plain river network area based on hydrologic and hydrokinetic model | |
| CN117911949B (en) | Risk intelligent assessment method and system | |
| CN116363601A (en) | Data acquisition and analysis method and system for pollution monitoring equipment | |
| CN110414715B (en) | A passenger flow early warning method based on community detection | |
| CN107515842A (en) | A method and system for dynamic prediction of urban population density | |
| CN111353051A (en) | K-means and Apriori-based algorithm maritime big data association analysis method | |
| Uslu et al. | Image-based 3D reconstruction and recognition for enhanced highway condition assessment | |
| Sun et al. | Automatic building age prediction from street view images | |
| CN113158084B (en) | Method, device, computer equipment and storage medium for processing movement track data | |
| KR20230167856A (en) | Visibility Prediction Method using Tree-based Machine Learning Algorithm and Meteorological Forecasting Data | |
| CN117827863B (en) | Atmospheric environment monitoring and analysis method and system based on CLDAS database | |
| CN115761519B (en) | Index prediction method, device, equipment, storage medium and program product | |
| CN118196475A (en) | Target object risk detection method and device and electronic equipment | |
| CN114925994A (en) | Urban village risk assessment and risk factor positioning method based on deep learning | |
| CN117874498B (en) | Intelligent forestry big data system, method, equipment and medium based on data lake |
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 |