Intelligent monitoring data analysis system and method based on cloud computing

Through an intelligent monitoring data analysis system based on cloud computing, deep learning technology is used to extract and correlate the feature of urban security monitoring video and basic information, and the problem of rapid positioning and low resource utilization in the existing technology is solved, and the monitoring system in the existing technology is achieved. More efficient security management and abnormal event detection.

CN118968691APending Publication Date: 2024-11-15SUZHOU NENGJIA INFORMATION TECHNOLOGY CO LTD
View PDF 0 Cites 5 Cited by

Patent Information

Application Number
CN202410982806.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The existing urban monitoring system has challenges in rapid positioning, finding and identifying personnel, lacks proactive prevention and intervention capabilities, has low resource utilization, is difficult to manage video failures, loss and massive video data, and traditional security means cannot meet the needs of rapid retrieval.

Method used

An intelligent monitoring data analysis system based on cloud computing uses video collected by cameras and basic information collected by cloud computing platforms, and uses deep learning technology to extract feature vectors and perform correlation analysis to determine whether an abnormal event alert is issued to achieve real-time monitoring and security management.

Benefits of technology

It improves the overall security of the city, reduces criminal incidents and other security risks, enhances the sensitivity and timeliness of abnormal situations, and improves the ability to allocate resources and pre-events.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118968691A_ABST
    Figure CN118968691A_ABST
Patent Text Reader

Abstract

The invention relates to the field of monitoring data analysis, and particularly discloses an intelligent monitoring data analysis system and method based on cloud computing, and the method comprises the steps: firstly obtaining an urban security monitoring video collected by a camera and urban security basic information collected by a cloud computing platform, and then carrying out the deep learning technology, and feature extraction and correlation analysis are carried out on the two types of data, and finally a classification result is obtained through a classifier to judge whether an abnormal event alarm needs to be given out, so that an abnormal condition is found in time and corresponding measures are taken, thereby being beneficial to improving the overall safety of a city and reducing criminal events and other safety risks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of monitoring data analysis, and more specifically, to an intelligent monitoring data analysis system and method based on cloud computing. Background Art

[0002] Cities represent the progress of human civilization and the achievements of economic development. Urbanization is an inevitable trend in the development of human civilization and an important part of the modernization process. In the current global economic integration environment, urbanization is profoundly changing the world and people's lifestyles. However, with the acceleration of urbanization, the problems and challenges faced by cities in terms of environment, resources and management have become increasingly severe and prominent. The demand for cities is growing, while the supply of resources in cities is limited, and this contradiction is also intensifying. The sustainable development of cities requires finding and exploring new development paths.

[0003] With the increase in urban population mobility, security personnel are facing greater challenges in personnel monitoring. Urban security equipment and technical means can no longer meet the core needs of quickly locating, finding and identifying people. The current monitoring system tends to respond after the fact, lacks the ability to take the initiative to prevent and intervene, and the resource utilization rate of the monitoring system is also low. In addition, video failures, video loss, unclear video quality and the management of massive video data have brought great difficulties to the investigation work. With the accumulation of video data and the continuous expansion of the system, traditional security means and manual identification can no longer meet the needs of rapid retrieval.

[0004] Therefore, an intelligent monitoring data analysis system and method based on cloud computing is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a cloud computing-based intelligent monitoring data analysis system and method, which first obtains the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform, and then uses deep learning technology to extract features and perform correlation analysis on the two, and finally obtains the classification result through the classifier to determine whether it is necessary to issue an abnormal event alarm, thereby timely discovering abnormal situations and taking corresponding measures, thereby helping to improve the overall security of the city and reduce criminal incidents and other security risks.

[0006] According to one aspect of the present application, there is provided an intelligent monitoring data analysis system based on cloud computing, which includes:

[0007] The city security data acquisition module is used to obtain the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform;

[0008] A city security data extraction module, used to extract city security monitoring interest attention feature vectors and city security basic text semantic association feature vectors from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform;

[0009] The security abnormal event alarm module is used to determine whether it is necessary to issue an abnormal event alarm based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector.

[0010] According to another aspect of the present application, there is provided a cloud computing-based intelligent monitoring data analysis method, which includes:

[0011] Obtain urban security surveillance videos collected by cameras and basic urban security information collected by cloud computing platforms;

[0012] Extracting the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform;

[0013] Based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector, it is determined whether it is necessary to issue an abnormal event alarm.

[0014] Compared with the prior art, the present application provides a cloud computing-based intelligent monitoring data analysis system and method, which first obtains the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the classification result through the classifier to determine whether it is necessary to issue an abnormal event alarm, and then promptly discover the abnormal situation and take corresponding measures, which helps to improve the overall security of the city and reduce criminal incidents and other security risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0016] Figure 1 It is a block diagram of an intelligent monitoring data analysis system based on cloud computing according to an embodiment of the present application.

[0017] Figure 2It is a block diagram of the urban security data extraction module in the cloud computing-based intelligent monitoring data analysis system according to an embodiment of the present application.

[0018] Figure 3 It is a block diagram of a security abnormal event alarm module in a cloud computing-based intelligent monitoring data analysis system according to an embodiment of the present application.

[0019] Figure 4 The present invention is a flowchart of an intelligent monitoring data analysis method based on cloud computing according to an embodiment of the present application.

[0020] Figure 5 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0022] Figure 1 Schematic diagram of a cloud computing-based intelligent monitoring data analysis system according to an embodiment of the present application. Figure 1 As shown, according to the cloud computing-based intelligent monitoring data analysis system 100 of the embodiment of the present application, it includes: a city security data acquisition module 110, which is used to obtain the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform; a city security data extraction module 120, which is used to extract the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform; a security abnormal event alarm module 130, which is used to determine whether it is necessary to issue an abnormal event alarm based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector.

[0023] In the above-mentioned intelligent monitoring data analysis system 100 based on cloud computing, the urban security data acquisition module 110 is used to obtain the urban security monitoring video collected by the camera and the basic urban security information collected by the cloud computing platform. It should be understood that the city symbolizes the progress of human civilization and the achievement of economic prosperity. Urbanization is an inevitable trend in the development of human civilization and a key component of the modernization process. Under the current background of global economic integration, urbanization is profoundly changing the world and people's lifestyles. However, with the acceleration of urbanization, the challenges faced by cities in terms of environment, resources and management have become increasingly severe. Urban demand continues to grow, while resource supply appears to be limited, and this contradiction is constantly intensifying. The sustainable development of cities requires finding new development paths. With the increase in urban population mobility, security personnel face greater challenges in personnel monitoring. Urban security equipment and technical means can no longer meet the core needs of quickly locating, finding and identifying personnel. The current monitoring system focuses more on post-event response, lacks active prevention and intervention capabilities, and has low resource utilization. Video failure, loss, unclear quality and massive video data management bring great difficulties to investigation work. With the accumulation of video data and the expansion of the system, traditional security means and manual identification can no longer meet the needs of rapid retrieval. Therefore, in the technical solution of this application, by obtaining the urban security monitoring video collected by the camera and the basic urban security information collected by the cloud computing platform, and combining with deep learning technology, it is judged whether it is necessary to issue an abnormal event alarm. By timely discovering abnormal situations and taking corresponding measures, it is helpful to enhance the overall security of the city and reduce the occurrence of criminal incidents and other security risks.

[0024] Specifically, the urban security surveillance video collected by the camera is a real-time captured visual information that can provide real-time conditions in various areas of the city. These video data can help monitoring personnel to promptly detect abnormal situations, such as traffic accidents, crowds or criminal acts. By analyzing the video, the system can achieve real-time monitoring, behavior recognition and event detection, thereby improving urban security. The basic urban security information collected by the cloud computing platform includes data such as various urban facilities, traffic conditions, and personnel flow. This information is crucial for urban security management and emergency response. By collecting and analyzing this data, the system can achieve effective allocation of urban resources, event prediction and intelligent decision-making. Combining the video collected by the camera and the basic information collected by the cloud computing platform, the intelligent monitoring system can achieve multi-dimensional and all-round urban security monitoring and management. By comprehensively utilizing this data, the intelligent monitoring system can better respond to various security challenges and protect the lives and property of urban residents.

[0025] In the above-mentioned cloud computing-based intelligent monitoring data analysis system 100, the urban security data extraction module 120 is used to extract the urban security monitoring interest attention feature vector and the urban security basic text semantic association feature vector from the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform. It should be understood that by combining the features of visual information and text information, the system can more comprehensively analyze the urban security situation, discover potential risks and take timely measures. Therefore, extracting the urban security monitoring interest attention feature vector and the urban security basic text semantic association feature vector is a key data processing step in the intelligent monitoring system, which provides important support for the intelligent and efficient operation of the system.

[0026] Figure 2 FIG. 1 is a block diagram of a city security data extraction module in a cloud computing-based intelligent monitoring data analysis system according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the city security data extraction module 120 includes: a city security monitoring data preprocessing unit 121, which is used to perform data preprocessing on the city security monitoring video collected by the camera to obtain a plurality of noise-reduced city security monitoring key frame images; a key frame image feature encoding unit 122, which is used to perform feature encoding on the plurality of noise-reduced city security monitoring key frame images to obtain the city security monitoring interest attention feature vector; a city security basic information semantic encoding unit 123, which is used to perform semantic encoding on the city security basic information collected by the cloud computing platform to obtain the city security basic text semantic association feature vector.

[0027] It should be understood that urban security surveillance videos may be affected by many factors, such as lighting conditions, weather conditions, camera quality, etc., resulting in noise and blur in the image. Data preprocessing can help remove noise and interference in the image and improve image quality and clarity. By extracting key frame images from urban security surveillance videos, the amount of data can be reduced and attention can be focused on important information. Urban security surveillance videos are usually continuous streaming data containing a lot of redundant information. Extracting key frame images can extract the most representative and informative frames in the video sequence, reducing the burden of data storage and processing while retaining important monitoring content. This helps the system to perform subsequent image analysis and event recognition more effectively and improve the system's operating efficiency.

[0028] Furthermore, feature encoding of multiple noise-reduced urban security monitoring key frame images can help the system focus on key areas and events more accurately, thereby improving the monitoring system's ability to identify and respond to urban security situations. Among them, feature encoding can convert the rich information in the image into a high-dimensional feature vector and extract the key features of the image. Considering that urban security monitoring key frame images contain a large amount of visual information, such as target objects, action features, scene content, etc. Through the feature encoding algorithm, this information can be extracted and represented as a feature vector with semantic meaning, realizing the effective extraction and representation of the image content. This helps the system to understand the monitoring scene more comprehensively and capture important monitoring objects and events.

[0029] Furthermore, semantic coding of the basic urban security information collected by the cloud computing platform can convert text information into vector representations with semantic associations, thereby achieving a deeper understanding and analysis of the city's security situation. Semantic coding can convert the text content in the basic urban security information into semantically related feature vectors to achieve semantic representation of information. Basic urban security information may include various text data, such as alarm information, event descriptions, and area tags. Through semantic coding technology, these text information can be converted into high-dimensional vector representations, so that the text information has richer semantic information and associations. This helps the system better understand and analyze urban security events, thereby improving the perception and handling capabilities of security issues.

[0030] In a specific embodiment of the present application, the city security monitoring data preprocessing unit 121 includes: extracting multiple city security monitoring key frames from the city security monitoring video collected by the camera; passing the multiple city security monitoring key frames through an urban security monitoring key frame denoising module based on an automatic encoder to obtain the multiple denoised city security monitoring key frame images.

[0031] It should be understood that urban security surveillance videos usually contain a lot of redundant information, such as continuous scene changes, repeated background content, etc. Extracting multiple urban security surveillance key frames can help the system reduce the amount of data and improve processing efficiency. By extracting key frames, the system can select representative frame images from the video sequence, reduce unnecessary repeated information, reduce the complexity of data processing, and improve the operating efficiency of the system. Here, key frame extraction can help the system capture important information and events in the surveillance video. In urban security surveillance videos, there are often some key moments or important events, such as abnormal behavior, emergencies, etc. By extracting key frames, the system can capture these important scenes and events, helping monitoring personnel to locate and identify problems more quickly, and take corresponding measures in time to ensure urban safety.

[0032] Furthermore, in urban security monitoring, due to environmental factors and camera quality, the monitoring image is often disturbed by various noises, such as blur, particles, etc. Through the processing of the noise reduction module, the system can remove these interferences, making the key frame image clearer and easier to analyze, which helps the system to accurately capture key information and events. The de-noised urban security monitoring key frame image can improve the system's understanding and analysis capabilities of the monitoring scene. Clear and accurate images help the system better identify and analyze target objects, actions and events in the monitoring scene, thereby improving the intelligence level of the monitoring system. Specifically, the multiple urban security monitoring key frames are input into the encoder of the urban security monitoring key frame noise reduction module based on the autoencoder, wherein the encoder uses a convolution layer to perform explicit spatial encoding on the multiple urban security monitoring key frames to obtain image features; and the image features are input into the decoder of the urban security monitoring key frame noise reduction module based on the autoencoder, wherein the decoder uses a deconvolution layer to perform deconvolution processing on the image features to obtain the multiple de-noised urban security monitoring key frame images.

[0033] In a specific embodiment of the present application, the key frame image feature encoding unit 122 includes: passing the multiple denoised urban security monitoring key frame images through an urban security monitoring target detection model to obtain multiple urban security monitoring regions of interest; passing the multiple urban security monitoring regions of interest through an urban security monitoring feature of interest encoder based on a temporal attention mechanism to obtain the urban security monitoring attention feature vector of interest.

[0034] It should be understood that by processing the denoised key frame images through the urban security monitoring target detection model, the system's recognition accuracy of target objects in the monitoring scene can be improved. The target detection model can automatically identify different targets in the image and mark their locations and categories, helping the system to quickly and accurately locate key targets in the monitoring scene, such as suspicious persons, vehicles, etc., thereby achieving effective monitoring and early warning of potential risks. The urban security monitoring target detection model can help the system achieve refined analysis and monitoring of areas of interest. By identifying the target objects in the monitoring scene, the system can monitor and analyze these areas of interest in real time, understand the behavioral dynamics of the target objects, detect abnormal situations and take timely countermeasures, and improve the efficiency and response speed of urban security management. Specifically, the urban security monitoring target detection model is used to process the multiple denoised urban security monitoring key frame images with the following target detection formula to obtain the multiple urban security monitoring areas of interest;

[0035] Among them, the target detection formula is:

[0036] ROI=(cls(ψ det,B),Regr(ψ det ,B))

[0037] Among them, ψ det are multiple denoised urban security monitoring key frame images, B is the anchor frame, ROI is the region of interest for multiple urban security monitoring, cls(ψ det ,B) indicates classification, Regr(ψ det ,B) indicates regression.

[0038] Furthermore, processing the urban security monitoring region of interest through the urban security monitoring feature of interest encoder based on the temporal attention mechanism can help the system better capture the key features of the region of interest, improve the understanding and analysis capabilities of the monitoring scene, and thus achieve more accurate and efficient security monitoring and event recognition. Among them, the urban security monitoring feature of interest encoder based on the temporal attention mechanism can dynamically adjust the weight distribution of feature extraction according to the temporal changes and evolution of the region of interest, highlighting important time periods and key features. This mechanism enables the system to pay more attention to the areas and events that have changed in the monitoring scene, improve sensitivity to potential risks and abnormal situations, and help the system capture key information more accurately. Specifically, an adjacent first monitoring region of interest and a second monitoring region of interest are extracted from the multiple urban security monitoring regions of interest; the first monitoring region of interest and the second monitoring region of interest are respectively passed through the first convolution layer and the second convolution layer of the urban security monitoring feature of interest encoder based on the temporal attention mechanism to obtain a first convolution feature map and a second convolution feature map; the position point multiplication between the first convolution feature map and the second convolution feature map is calculated to obtain a temporal attention map; the temporal attention map is input into a Softmax activation function to obtain a temporal attention feature map; the second monitoring key frame is passed through the third convolution layer of the urban security monitoring feature of interest encoder based on the temporal attention mechanism to obtain a third convolution feature map; and, the position point multiplication between the third convolution feature map and the temporal attention feature map is calculated to obtain the urban security monitoring feature of interest attention vector.

[0039] In a specific embodiment of the present application, the city security basic information semantic encoding unit 123 includes: passing the city security basic information collected by the cloud computing platform through a city security basic text semantic context encoder to obtain a plurality of city security basic text semantic feature vectors; and cascading the plurality of city security basic text semantic feature vectors to obtain the city security basic text semantic association feature vector.

[0040] It should be understood that the city security basic text semantic context encoder can convert the text content in the city security basic information into a semantic feature vector, and capture the semantic relationship and context information in the text information. In the technical solution of the present application, the city security basic text semantic context encoder can help the system to mine the hidden information and rules in the city security basic information, and improve the abstract representation ability of text information. Through the obtained text semantic feature vector, the system can better identify the important features and keywords in the text information, realize the automatic processing and analysis of the city security related information, and provide faster and more accurate information support for urban security management. Specifically, the city security basic information collected by the cloud computing platform is segmented to obtain a word sequence; the embedding layer of the city security basic text semantic context encoder is used to map each word in the word sequence to a word embedding vector to obtain a sequence of word embedding vectors; the converter-based Bert model of the city security basic text semantic context encoder is used to perform global context semantic encoding on the sequence of word embedding vectors to obtain multiple city security basic text semantic feature vectors.

[0041] Furthermore, cascading multiple semantic feature vectors of basic urban security texts can integrate the semantic features of different text information, thereby more comprehensively and accurately expressing the associations and connections between basic urban security information, and providing more comprehensive information support for urban safety management and decision-making. Cascading multiple semantic feature vectors of basic urban security texts can help the system comprehensively consider the semantic relationships and mutual influences between different text information, and realize information integration and cross-analysis. By cascading these feature vectors together, the system can more comprehensively capture the association features between basic urban security information, reveal the hidden connections and laws between text information, and provide the system with deeper information understanding and analysis capabilities.

[0042] In the above-mentioned cloud computing-based intelligent monitoring data analysis system 100, the security abnormal event alarm module 130 is used to determine whether it is necessary to issue an abnormal event alarm based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector. It should be understood that through the city security monitoring interest attention feature vector, the system can monitor the video data of various areas of the city in real time and capture key visual features and abnormal situations. These feature vectors can help the system identify areas or events that may have security risks and improve the sensitivity and timeliness of abnormal situations. Combined with the city security basic text semantic association feature vector, the system can obtain more background and context information from the text information to assist in a deeper understanding and analysis of events occurring in the monitoring video. This comprehensive consideration of video and text information can improve the accuracy of the system's judgment of urban security events and reduce false alarms and missed reports. Taking into account the interest attention feature vector and the text semantic association feature vector, the system can establish a more comprehensive urban security event judgment model. Once the system detects an abnormal situation, it can issue an alarm in time to notify the relevant departments to handle it, thereby improving the efficiency and response speed of urban security management.

[0043] Figure 3 FIG. 1 is a block diagram of a security abnormal event alarm module in a cloud computing-based intelligent monitoring data analysis system according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the security abnormal event alarm module 130 includes: a security data feature fusion unit 131, which is used to fuse the urban security monitoring interest attention feature vector and the urban security basic text semantic association feature vector to obtain an abnormal event alarm judgment classification feature vector; a security data feature optimization unit 132, which is used to optimize the abnormal event alarm judgment classification feature vector to obtain an optimized abnormal event alarm judgment classification feature vector; an abnormal event alarm judgment generation unit 133, which is used to pass the optimized abnormal event alarm judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether an abnormal event alarm needs to be issued.

[0044] It should be understood that the fusion of these two feature vectors can comprehensively utilize the advantages of video surveillance and text information to achieve cross-validation and complementarity of multimodal information. The fused abnormal event alarm judgment classification feature vector can more comprehensively characterize the city's security status, improve the accuracy and prediction ability of abnormal events, reduce false alarms and missed alarms, and provide more effective support for urban security management. The generation of this comprehensive feature vector can not only help the system better understand the complexity and diversity of urban security information, but also provide more comprehensive and intelligent information support for the classification and prediction of abnormal events.

[0045] Furthermore, in a specific embodiment of the present application, the security data feature optimization unit 132 includes: performing a significance improvement of the category probability description of the abnormal event alarm judgment classification feature vector based on the feature attribute layout to obtain an optimization factor; and, using the optimization factor as a weight, weighting the abnormal event alarm judgment classification feature vector to obtain an optimized abnormal event alarm judgment classification feature vector.

[0046] Among them, the abnormal event alarm judgment classification feature vector is significantly improved based on the category probability description of the feature attribute layout to obtain the optimization factor, including: multiplying the abnormal event alarm judgment classification feature vector with the classification weight matrix of the classifier to obtain the abnormal event alarm judgment classification weight feature vector; cascading the abnormal event alarm judgment classification feature vector and the abnormal event alarm judgment classification weight feature vector to obtain the abnormal event alarm judgment classification feature-weight information joint vector; passing the abnormal event alarm judgment classification feature-weight information joint vector through the first fully connected layer and then through the sigmoid function to obtain the first activation output value; adding the abnormal event alarm judgment classification feature vector and the abnormal event alarm judgment classification weight feature vector by position to obtain the abnormal event alarm judgment classification feature-weight information sum vector; the abnormal event alarm judgment classification feature-weight information sum vector is passed through After the second fully connected layer, the sigmoid function is passed to obtain the second activation output value; the mean of the first activation output value and the second activation output value is calculated, and 1 is subtracted from the mean to obtain the first weighting coefficient; the mean of the first activation output value and the second activation output value is calculated to obtain the second weighting coefficient; the natural exponential function value is calculated by raising the eigenvalue of the abnormal event alarm judgment classification weight feature vector to a power to obtain the first exponential abnormal event alarm judgment classification weight feature vector; the natural exponential function value is calculated by raising the eigenvalue of the abnormal event alarm judgment classification feature vector to a power to obtain the second exponential abnormal event alarm judgment classification feature vector; based on the first weighting coefficient and the second weighting coefficient, the weighted sum of the first exponential abnormal event alarm judgment classification weight feature vector and the second exponential abnormal event alarm judgment classification feature vector is calculated to obtain the weighted sum feature vector, and the bi-norm of the weighted sum feature vector is calculated to obtain the optimization factor.

[0047] In particular, in the technical solution of the present application, the urban security monitoring video and the urban security basic information come from different data sources and have different data characteristics and structures. Video data is high-dimensional time series data, while basic information may be structured data. This difference may lead to inconsistent distribution of feature vectors in high-dimensional space. The monitoring key frame is denoised by an automatic encoder, and then features are extracted by a target detection model; while the basic information is extracted by a text semantic context encoder. Different extraction methods may lead to inconsistent distribution of feature vectors in high-dimensional space. Feature encoding based on the time attention mechanism may emphasize certain time series features and ignore other features, which may lead to inconsistent distribution of feature vectors at different time points. When fusing the monitoring interest attention feature vector and the basic text semantic association feature vector, the weights and interaction effects of the two need to be considered. If the fusion method is not refined enough, it may lead to inconsistent distribution of feature vectors in high-dimensional space. If the abnormal event alarm judgment classification feature vector is inconsistently distributed in high-dimensional space, it may be difficult for the classifier to capture the pattern of all features during training, resulting in inaccurate correspondence between the class label learned by the model and the true label. The inconsistency of the abnormal event alarm judgment classification feature vector may cause the classification result to have high uncertainty in some dimensions, which will directly affect the accuracy and reliability of the alarm system. That is, the manifold geometric consistency of the overall feature distribution of the abnormal event alarm judgment classification feature vector is poor, resulting in poor regression standardization of the class label of the classifier when it is classified and regressed through the classifier, affecting the accuracy of the classification result. Based on this, the category probability description of the abnormal event alarm judgment classification feature vector based on the feature attribute layout is significantly improved.

[0048] The abnormal event alarm judgment classification feature vector is improved based on the category probability description of the feature attribute layout to obtain an optimization factor, including: the abnormal event alarm judgment classification feature vector is improved based on the category probability description of the feature attribute layout according to the following optimization formula, wherein the optimization formula is:

[0049]

[0050] Among them, v c represents the abnormal event alarm judgment classification feature vector, M represents the classification weight matrix of the classifier, represents matrix multiplication, Indicates positional addition, concat indicates cascade function, W 1 represents the first weight matrix of the first fully connected layer, b 1 represents the first bias vector of the first fully connected layer, t 1 represents the first activation output value, W 2represents the second weight matrix of the second fully connected layer, b 2 represents the second bias vector of the second fully connected layer, t 2 represents the second activation output value, sigmoid represents the S-shaped function, ||·|| 2 represents the second norm of the vector, and α represents the optimization factor.

[0051] That is, in the technical solution of the present application, the geometric consistency of the manifold of the overall feature distribution of the abnormal event alarm judgment classification feature vector is poor, resulting in poor standardization of the class label regression of the classifier when it is classified and regressed through the classifier, affecting the accuracy of the classification result. Therefore, in the technical solution of the present application, the abnormal event alarm judgment classification feature vector is improved based on the class attribute layout of the category probability description significance, which is performed by using the classification weight matrix of the classifier to perform auxiliary analysis of the feature attribute layout of the abnormal event alarm judgment classification feature vector, and the statistical norm of the differential features between the auxiliary analysis result and the original abnormal event alarm judgment classification feature vector in the class high-dimensional space is used to perform backward correlation prediction, so as to improve the improved abnormal event alarm judgment classification feature vector The description significance of the category probability of the predetermined category of the classifier. In this way, maintaining the geometric consistency of the abnormal event alarm judgment classification feature vector during the migration transformation to the class label regression domain of the classifier not only helps to reduce the regression bias in the classification process, but also can improve the generalization ability of the classifier to new data, and ultimately achieve more accurate classification results.

[0052] In summary, the embodiment of the present application first obtains the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform, and then uses deep learning technology to perform feature extraction and correlation analysis on the two, and finally obtains the classification result through the classifier to determine whether it is necessary to issue an abnormal event alarm, so as to detect the abnormal situation in time and take corresponding measures, which will help improve the overall security of the city and reduce criminal incidents and other security risks.

[0053] Furthermore, by inputting the optimized feature vector into the classifier, the system can use the learning ability and pattern recognition ability of the classifier to distinguish and classify different types of features. The classifier can automatically identify and learn the characteristic laws of abnormal events based on the characteristic patterns in the training data, thereby classifying and judging new feature vectors. Secondly, the classifier can help the system establish a classification model for abnormal events and map different types of feature vectors to corresponding category labels. Through the classification results of the classifier, the system can accurately determine whether there are abnormal events in the current monitoring situation and decide whether an alarm needs to be issued. This automated classification and judgment process can greatly reduce the burden of manual intervention and improve the efficiency and reliability of urban safety management.

[0054] As described above, the cloud computing-based intelligent monitoring data analysis system 100 according to the embodiment of the present application can be implemented in various terminal devices. In one example, the cloud computing-based intelligent monitoring data analysis system 100 can be integrated into the terminal device as a software module and / or a hardware module. For example, the cloud computing-based intelligent monitoring data analysis system 100 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 cloud computing-based intelligent monitoring data analysis system 100 can also be one of the many hardware modules of the terminal device.

[0055] Alternatively, in another example, the cloud computing-based intelligent monitoring data analysis system 100 and the terminal device may also be separate devices, and the cloud computing-based intelligent monitoring data analysis system 100 may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0056] Figure 4 FIG. 1 is a flow chart of an intelligent monitoring data analysis method based on cloud computing according to an embodiment of the present application. Figure 4 As shown, according to the cloud computing-based intelligent monitoring data analysis method of the embodiment of the present application, it includes: S110, obtaining the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform; S120, extracting the urban security monitoring interest attention feature vector and the urban security basic text semantic association feature vector from the urban security monitoring video collected by the camera and the urban security basic information collected by the cloud computing platform; S130, judging whether it is necessary to issue an abnormal event alarm based on the urban security monitoring interest attention feature vector and the urban security basic text semantic association feature vector.

[0057] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned cloud computing-based intelligent monitoring data analysis method have been referred to above. Figures 1 to 3 The description of the cloud computing-based intelligent monitoring data analysis system has been introduced in detail, and therefore, its repeated description will be omitted.

[0058] Below, reference Figure 5 To describe an electronic device according to an embodiment of the present application.

[0059] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned drone-based image target detection when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.

[0060] See also Figure 5 , Figure 5 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0061] The processor 11 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solution provided in the embodiment of the present application; the memory 12 can be implemented in the form of a read-only memory (Read Only Memory, ROM), a static storage device, a dynamic storage device or a random access memory (Random Access Memory, RAM), etc.

[0062] The memory 12 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 12, and the processor 11 calls and executes the cloud computing-based intelligent monitoring data analysis method of the embodiment of this application;

[0063] Input / output interface 13, used to implement information input and output;

[0064] The communication interface 14 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WIFI, Bluetooth, etc.);

[0065] A bus 15 that transmits information between the various components of the device (e.g., the processor 11, the memory 12, the input / output interface 13, and the communication interface 14);

[0066] The processor 11 , the memory 12 , the input / output interface 13 and the communication interface 14 are connected to each other in communication within the device via a bus 15 .

[0067] An embodiment of the present application also provides a storage medium, which is a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned cloud computing-based intelligent monitoring data analysis method is implemented.

[0068] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0069] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0070] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0071] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0072] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0073] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0074] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0075] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0076] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0077] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-On lyMemory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs. The preferred embodiment of the embodiment of the present application is described above with reference to the accompanying drawings, and the scope of rights of the embodiment of the present application is not limited thereto. Any modifications, equivalent substitutions and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of rights of the embodiments of the present application.

Claims

1. An intelligent monitoring data analysis system based on cloud computing, It is characterized in that include: The city security data acquisition module is used to obtain the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform; A city security data extraction module, used to extract city security monitoring interest attention feature vectors and city security basic text semantic association feature vectors from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform; The security abnormal event alarm module is used to determine whether it is necessary to issue an abnormal event alarm based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector.

2. According to the cloud computing-based intelligent monitoring data analysis system of claim 1, It is characterized in that The urban security data extraction module comprises: A city security monitoring data preprocessing unit, used for performing data preprocessing on the city security monitoring video collected by the camera to obtain a plurality of city security monitoring key frame images after noise reduction; A key frame image feature encoding unit, used for performing feature encoding on the plurality of noise-reduced urban security monitoring key frame images to obtain the urban security monitoring attention feature vector of interest; The city security basic information semantic encoding unit is used to semantically encode the city security basic information collected by the cloud computing platform to obtain the city security basic text semantic association feature vector.

3. According to the cloud computing-based intelligent monitoring data analysis system of claim 2, It is characterized in that The urban security monitoring data preprocessing unit comprises: Extracting a plurality of city security monitoring key frames from the city security monitoring video collected by the camera; The multiple city security monitoring key frames are passed through a city security monitoring key frame denoising module based on an auto-encoder to obtain the multiple denoised city security monitoring key frame images.

4. According to claim 3, the intelligent monitoring data analysis system based on cloud computing, It is characterized in that The key frame image feature encoding unit comprises: Passing the plurality of noise-reduced urban security monitoring key frame images through an urban security monitoring target detection model to obtain a plurality of urban security monitoring regions of interest; The multiple urban security monitoring regions of interest are passed through an urban security monitoring feature of interest encoder based on a temporal attention mechanism to obtain the urban security monitoring feature of interest attention vector.

5. According to the cloud computing-based intelligent monitoring data analysis system of claim 4, It is characterized in that The city security basic information semantic coding unit includes: Passing the basic city security information collected by the cloud computing platform through a basic city security text semantic context encoder to obtain a plurality of basic city security text semantic feature vectors; The plurality of city security basic text semantic feature vectors are cascaded to obtain the city security basic text semantic association feature vector.

6. According to claim 5, the intelligent monitoring data analysis system based on cloud computing, It is characterized in that The security abnormal event alarm module comprises: A security data feature fusion unit, used to fuse the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector to obtain an abnormal event alarm judgment classification feature vector; A security data feature optimization unit, used for optimizing the abnormal event alarm judgment classification feature vector to obtain an optimized abnormal event alarm judgment classification feature vector; The abnormal event alarm judgment generating unit is used to pass the optimized abnormal event alarm judgment classification feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether it is necessary to issue an abnormal event alarm.

7. The cloud computing-based intelligent monitoring data analysis system according to claim 6, It is characterized in that The security data feature optimization unit comprises: Performing a category probability description significance improvement based on feature attribute layout on the abnormal event alarm judgment classification feature vector to obtain an optimization factor; and The optimization factor is used as a weight to weight the abnormal event alarm judgment classification feature vector to obtain an optimized abnormal event alarm judgment classification feature vector.

8. According to claim 7, the intelligent monitoring data analysis system based on cloud computing, It is characterized in that The abnormal event alarm judgment classification feature vector is improved based on the category probability description significance of the feature attribute layout to obtain an optimization factor, including: Multiplying the abnormal event alarm judgment classification feature vector by the classification weight matrix of the classifier to obtain the abnormal event alarm judgment classification weight feature vector; Cascading the abnormal event alarm judgment classification feature vector and the abnormal event alarm judgment classification weight feature vector to obtain an abnormal event alarm judgment classification feature-weight information joint vector; The abnormal event alarm judgment classification feature-weight information joint vector passes through the first fully connected layer and then passes through the sigmoid function to obtain a first activation output value; Add the abnormal event alarm judgment classification feature vector and the abnormal event alarm judgment classification weight feature vector according to position to obtain an abnormal event alarm judgment classification feature-weight information sum vector; The abnormal event alarm judgment classification feature-weight information sum vector passes through the second fully connected layer and then passes through the sigmoid function to obtain a second activation output value; Calculating a mean of the first activation output value and the second activation output value, and subtracting the mean from 1 to obtain a first weighting coefficient; Calculating a mean of the first activation output value and the second activation output value to obtain a second weighting coefficient; Calculate the value of a natural exponential function by exponentiating the eigenvalue of the abnormal event alarm judgment classification weight eigenvector to obtain a first exponential abnormal event alarm judgment classification weight eigenvector; Calculate the value of a natural exponential function by using the eigenvalue of the abnormal event alarm judgment classification feature vector as a power to obtain a second exponential abnormal event alarm judgment classification feature vector; Based on the first weighting coefficient and the second weighting coefficient, the weighted sum of the first index abnormal event alarm judgment classification weight feature vector and the second index abnormal event alarm judgment classification feature vector is calculated to obtain a weighted sum feature vector, and the bi-norm of the weighted sum feature vector is calculated to obtain an optimization factor.

9. An intelligent monitoring data analysis method based on cloud computing, It is characterized in that include: Obtain urban security surveillance videos collected by cameras and basic urban security information collected by cloud computing platforms; Extracting the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform; Based on the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector, it is determined whether it is necessary to issue an abnormal event alarm.

10. The intelligent monitoring data analysis method based on cloud computing according to claim 9, It is characterized in that Extracting the city security monitoring interest attention feature vector and the city security basic text semantic association feature vector from the city security monitoring video collected by the camera and the city security basic information collected by the cloud computing platform, including: Performing data preprocessing on the urban security monitoring video collected by the camera to obtain a plurality of noise-reduced urban security monitoring key frame images; Performing feature encoding on the plurality of noise-reduced urban security monitoring key frame images to obtain the urban security monitoring attention feature vector of interest; The city security basic information collected by the cloud computing platform is semantically encoded to obtain the city security basic text semantic association feature vector.

Citation Information

Cited By

  • Intelligent education management system and method based on AI and cloud computing

    CN119918806A

  • Regional risk determination method and device, computer program product and electronic equipment

    CN119920070A

  • Monitoring video abnormity alarm platform and alarm system based on big data analysis

    CN119967157A

  • Surveillance video anomaly alarm platform and alarm system based on big data analysis

    CN119967157B

  • Intelligent analysis system based on artificial intelligence

    CN120578945A