Park monitoring data storage and sharing method and system based on cloud computing
By processing and identifying campus monitoring data privacy on the cloud computing platform, the problems of data sharing security and identification of suspicious persons among campuses are solved, and the performance and responsiveness of the monitoring system are improved.
Patent Information
- Application Number
- CN202411241018.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-09-05
AI Technical Summary
In the prior art, due to poor security of data sharing, the privacy of personnel in different parks is at risk of leakage, and the accuracy of identification of suspicious people is not high, resulting in limited overall performance of the monitoring system.
The park monitoring data storage and sharing method based on cloud computing is adopted. By collecting monitoring image sequences in adjacent parks, privacy personnel identification and encryption processing are carried out, suspicious personnel characteristics are identified, and data is uploaded to the cloud computing platform for storage and sharing. Ensure data security and identification accuracy.
It realizes secure cross-park monitoring data storage and shared identification, improves the overall performance of the park monitoring system, and improves the efficiency and response capabilities of security monitoring.
Smart Images

Figure CN118762333B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of data sharing technology, and in particular to a cloud computing-based campus monitoring data storage and sharing method and system. Background Art
[0002] With the acceleration of urbanization, various industrial parks, science and technology parks, and industrial parks are constantly emerging. These parks usually occupy a large area, with a large number of enterprises and personnel, and the demand for monitoring in terms of security, environmental protection, and energy management is growing. With the improvement of the resolution of monitoring equipment and the increase in monitoring points, the amount of data has shown explosive growth. Traditional data storage methods often rely on local servers or hard disks. The scalability, reliability, and maintenance costs of this method are difficult to meet the storage needs of large-scale monitoring data.
[0003] At present, due to the poor security of data sharing in existing technologies, there is a risk of privacy leakage in different parks, and the accuracy of identifying suspicious persons is not high, which limits the overall performance of the monitoring system and further affects the safety management and emergency response capabilities of the park. Summary of the invention
[0004] The purpose of this application is to provide a cloud computing-based campus monitoring data storage and sharing method and system to solve the problems of the prior art due to poor security of data sharing, the risk of privacy leakage of personnel in different parks, and low accuracy in identifying suspicious persons, which limits the overall performance of the monitoring system.
[0005] In view of the above problems, the present application provides a cloud computing-based campus monitoring data storage and sharing method and system.
[0006] In a first aspect, the present application provides a cloud computing-based campus monitoring data storage and sharing method, which is implemented by a cloud computing-based campus monitoring data storage and sharing system, wherein the cloud computing-based campus monitoring data storage and sharing method includes: in adjacent parks, based on monitoring equipment, continuously collecting monitoring images to obtain multiple park monitoring image sequences; in multiple park nodes in the adjacent parks, performing privacy personnel identification and privacy personnel encryption processing on the multiple park monitoring image sequences to obtain multiple encrypted park monitoring image sequences; in the multiple park nodes, based on the multiple encrypted park monitoring image sequences, respectively, suspicious personnel identification is performed to obtain suspicious appearance features and suspicious action features; based on the suspicious appearance features, a suspicious monitoring image sequence including suspicious personnel is screened and obtained, and based on the suspicious action features, a suspicious movement time frame of the adjacent park is determined; the suspicious monitoring image sequence and the suspicious movement time frame are uploaded to the cloud computing platform, and the cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification.
[0007] In a second aspect, the present application further provides a cloud computing-based campus monitoring data storage and sharing system for executing the cloud computing-based campus monitoring data storage and sharing method as described in the first aspect, wherein the cloud computing-based campus monitoring data storage and sharing system includes: a multi-park monitoring image sequence acquisition module for continuously collecting monitoring images based on monitoring equipment in adjacent parks to obtain multiple park monitoring image sequences; an encryption module for performing privacy personnel identification and privacy personnel encryption processing on the multiple park monitoring image sequences in multiple park nodes in the adjacent parks to obtain multiple encrypted park monitoring image sequences; a personnel identification module for In the multiple park nodes, suspicious persons are identified respectively according to the multiple encrypted park monitoring image sequences to obtain suspicious appearance features and suspicious action features; an image screening module is used to screen and obtain suspicious monitoring image sequences including suspicious persons according to the suspicious appearance features, and determine the suspicious movement time frame of the adjacent park according to the suspicious action features; a monitoring identification module is used to upload the suspicious monitoring image sequence and the suspicious movement time frame to a cloud computing platform, and the cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification.
[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0009] The invention obtains a plurality of park monitoring image sequences by continuously collecting monitoring images based on monitoring equipment in adjacent parks respectively; performs privacy personnel identification and privacy personnel encryption processing on the plurality of park monitoring image sequences in the plurality of park nodes in the adjacent parks respectively, and obtains a plurality of encrypted park monitoring image sequences; performs suspicious personnel identification in the plurality of park nodes respectively according to the plurality of encrypted park monitoring image sequences, and obtains suspicious appearance features and suspicious action features; screens and obtains a suspicious monitoring image sequence including suspicious persons according to the suspicious appearance features, and determines the suspicious movement time frame of the adjacent park according to the suspicious action features; uploads the suspicious monitoring image sequence and the suspicious movement time frame to a cloud computing platform, and the cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification, which effectively solves the problem that the existing technology has poor security of data sharing, the risk of leakage of personnel privacy in different parks, and the low accuracy of identifying suspicious persons, which limits the overall performance of the monitoring system, realizes safe cross-park monitoring data storage and shared monitoring identification, improves the overall performance of the park monitoring system, and improves the efficiency and responsiveness of security monitoring.
[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.
[0012] Figure 1 This is a flow chart of the cloud computing-based campus monitoring data storage and sharing method of this application.
[0013] Figure 2 This is a schematic diagram of the structure of the campus monitoring data storage and sharing system based on cloud computing in this application.
[0014] Description of reference numerals:
[0015] A multi-park monitoring image sequence acquisition module 11, an encryption module 12, a personnel identification module 13, an image screening module 14, and a monitoring identification module 15. DETAILED DESCRIPTION
[0016] The present application provides a cloud computing-based campus monitoring data storage and sharing method and system, which solves the problems of the prior art due to poor security of data sharing, the risk of privacy leakage of personnel in different parks, and low accuracy in identifying suspicious persons, which limits the overall performance of the monitoring system. It realizes secure cross-park monitoring data storage and shared monitoring identification, improves the overall performance of the campus monitoring system, and enhances the efficiency and responsiveness of security monitoring.
[0017] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.
[0018] For example, please refer to the attached Figure 1 The present application provides a cloud computing-based campus monitoring data storage and sharing method, wherein the cloud computing-based campus monitoring data storage and sharing method is applied to a cloud computing-based campus monitoring data storage and sharing system, and the cloud computing-based campus monitoring data storage and sharing method specifically includes the following steps:
[0019] S1: In adjacent parks, monitoring images are continuously collected based on monitoring equipment to obtain multiple park monitoring image sequences.
[0020] Specifically, analyze the key areas and targets that need to be monitored in the park, such as entrances, exits, parking lots, main roads, important facilities, etc. Select appropriate monitoring equipment according to the monitoring target, such as high-definition cameras, thermal imaging cameras, motion detection cameras, etc. Install the camera at the planned location, ensure that the equipment is stable, and take appropriate protective measures. Connect the monitoring equipment to the park network, which can be a wired connection or a wireless connection, to ensure that the network is stable and reliable, adjust the camera parameters, such as resolution, frame rate, exposure time, etc., to ensure that the image quality meets the monitoring requirements, and the monitoring equipment continuously collects image data according to the set parameters, and arranges the collected images in chronological order to form a monitoring image sequence of multiple parks. In two adjacent parks, give priority to the two closest monitoring areas for image acquisition.
[0021] S2: In the multiple park nodes in the adjacent park, privacy personnel identification and privacy personnel encryption processing are performed on the multiple park monitoring image sequences respectively to obtain multiple encrypted park monitoring image sequences.
[0022] Specifically, use deep learning algorithms, such as convolutional neural networks (CNNs), to recognize faces. Use human pose estimation or pedestrian re-identification technology to identify individuals. Preprocess surveillance images, such as scaling and grayscale conversion. Use recognition algorithms to extract facial or human features in images, compare the extracted features with known features in the database, and identify privacy-sensitive persons. Select appropriate encryption algorithms, such as AES, RSA, etc. Determine the granularity of encryption, which can be the entire image or a specific area in the image, such as a face or body. Based on the recognition results, determine the privacy areas in the image that need to be encrypted, generate a unique encryption key for each park or each monitoring node, and use the encryption algorithm and key to encrypt the privacy areas.
[0023] S3: In the multiple park nodes, suspicious persons are identified respectively according to the multiple encrypted park monitoring image sequences to obtain suspicious appearance features and suspicious action features.
[0024] Specifically, decrypt the privacy area in the surveillance image sequence, and use the corresponding decryption algorithm to decrypt the encrypted privacy area. Select appropriate image processing and machine learning algorithms to identify suspicious persons. Define suspicious appearance features, such as clothing, accessories, hairstyle, etc. and suspicious action features, such as abnormal speed, abnormal path, wandering behavior, etc. Adjust the image brightness and contrast, eliminate noise, and ensure image quality. Use image processing technology to extract the appearance features of the person, compare the extracted appearance features with the known suspicious feature database, and mark the person with suspicious appearance features based on the comparison results. Establish a normal behavior model for comparison with the actual behavior, extract the movement trajectory of the person through image sequence analysis, compare the movement trajectory with the normal behavior model, detect abnormal action features, and mark the detected abnormal action features as suspicious. Suspicious appearance feature records include identified suspicious appearance features, such as color, pattern, size, clothing, body shape, carried items, facial features, etc. Suspicious action feature records include identified suspicious action features, such as abnormal speed, abnormal path, wandering behavior, sudden stop, etc.
[0025] S4: According to the suspicious appearance features, a suspicious surveillance image sequence including suspicious persons is screened and obtained, and according to the suspicious action features, a suspicious movement time frame of an adjacent park is determined.
[0026] Specifically, suspicious appearance features include but are not limited to clothing, body shape, items carried, facial features, etc. The appearance features extracted from the surveillance image are matched with suspicious features in multiple encrypted park surveillance image sequences, and based on the matching results, the surveillance image sequences containing suspicious appearance features are screened out. Suspicious action features include abnormal speed, abnormal path, wandering behavior, sudden stops, etc. Based on the occurrence of suspicious actions, the time frame of suspicious movement is marked. Based on the suspicious action speed and suspicious action direction in the suspicious action features, determine whether to go to the adjacent park. If so, calculate the time frame to reach the monitoring range of the adjacent park based on the speed and direction as the suspicious movement time frame.
[0027] S5: The suspicious monitoring image sequence and the suspicious movement time frame are uploaded to the cloud computing platform. The cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification.
[0028] Specifically, the suspicious surveillance image sequence is sorted and annotated to ensure the integrity and identifiability of the image. The suspicious movement time frame includes the specific time frame of the suspicious movement, including the start and end time. The image sequence is compressed, and the data is encrypted before uploading to ensure the security of data transmission. A secure data transmission connection is established with the cloud computing platform, and the packaged suspicious surveillance image sequence and suspicious movement time frame are uploaded to the cloud computing platform. The cloud computing platform stores the received data in a secure storage service, such as object storage or block storage. According to the uploaded suspicious movement time frame, the platform algorithm extracts the image of the corresponding time frame from the encrypted surveillance image sequence of the adjacent park. The surveillance image within a period of time after the suspicious movement time frame is extracted, and the extracted encrypted surveillance image is decrypted using the corresponding key. A database is established on the cloud computing platform to store the extracted and decrypted surveillance images, and the features in the extracted images are compared with the suspicious feature database to identify suspicious persons, and the identification results are shared with authorized security and management personnel.
[0029] Furthermore, step S2 of the present application also includes:
[0030] In the multiple park nodes in the adjacent park, personnel image segmentation and privacy personnel identification are performed on the park monitoring images in the multiple park monitoring image sequences to obtain multiple privacy personnel identification result sequences; based on the multiple privacy personnel identification result sequences, the privacy personnel images in the multiple park monitoring image sequences are encrypted to obtain multiple encrypted park monitoring image sequences.
[0031] Specifically, use deep learning segmentation algorithms, such as Mask R-CNN, U-Net, etc., to accurately segment the characters in the image. Obtain a sequence of campus surveillance images, adjust the image size, contrast, etc. to prepare for segmentation, and apply the segmentation algorithm to separate the characters in the image from the background. Extract features from the segmented character images, compare the features with the privacy personnel database, record the comparison results, and generate a sequence of privacy personnel identification results. Select a suitable encryption algorithm, such as AES or RSA, and determine the granularity of encryption, which can be the entire segmented character image or a specific area, to ensure the secure generation, storage, and distribution of encryption keys, apply the encryption algorithm to the image area identified as the privacy personnel, and reintegrate the encrypted privacy personnel image into the original surveillance image to form an encrypted campus surveillance image sequence.
[0032] Furthermore, step S2 of the present application also includes:
[0033] According to the historical monitoring data of the park, a plurality of sample park monitoring image sets are collected, and the personnel images in each sample park monitoring image are segmented and labeled to obtain a plurality of sample personnel image segmentation result sets; the plurality of sample park monitoring image sets and the plurality of sample personnel image segmentation result sets are used to train a plurality of personnel image segmentors, and personnel image segmentation is performed on the park monitoring images in the plurality of park monitoring image sequences to obtain a plurality of personnel image segmentation results; based on the plurality of privacy personnel databases of the adjacent parks, a plurality of privacy personnel images are extracted, and a plurality of privacy personnel image identifiers are trained in the plurality of park nodes respectively; the plurality of privacy personnel image identifiers are used respectively to perform privacy personnel identification and labeling on the plurality of personnel image segmentation results to obtain a plurality of privacy personnel labeling result sequences.
[0034] Specifically, multiple sample park monitoring image sets are collected from the park's historical monitoring data, images with poor quality and incomplete information are removed, the resolution and format of the images are unified, the personnel images in the sample park monitoring images are segmented, the segmented personnel images are annotated, and the boundaries and other relevant information of the personnel are marked. The segmentation and annotation results are organized into a sample set for subsequent training, and a training data set is constructed using the sample park monitoring image set and the sample personnel image segmentation result set. A suitable deep learning model, such as Mask R-CNN, U-Net, etc., is selected, and the model is trained using the training data set until a satisfactory segmentation effect is achieved. The trained model is tested using the validation set, and the model is tuned according to the test results. Private personnel images are extracted from multiple private personnel databases of adjacent parks, and the extracted private personnel images are used to construct a training data set. A suitable face recognition or body recognition model is selected, which can be a neural network model, and a private personnel image recognizer is trained in each park node. The trained personnel image segmentor is used to segment the park monitoring images in multiple park monitoring image sequences, and the segmented personnel images are recognized using the trained private personnel image recognizer. The identified privacy personnel are identified, and the identification results are recorded to generate a privacy personnel identification result sequence.
[0035] Furthermore, the present application also includes:
[0036] According to the multiple privacy personnel identification result sequences, multiple privacy personnel image sets identified as privacy personnel are extracted; the privacy personnel images in the multiple park monitoring image sequences are encrypted to obtain multiple encrypted park monitoring image sequences.
[0037] Specifically, the sequence of private person identification results is analyzed to determine which images contain private persons. Based on the identification results, images identified as private persons are extracted from the park monitoring image sequence, and the extracted private person images are organized into a set. A suitable encryption algorithm, such as AES, RSA, etc., is selected to prepare the necessary encryption keys and parameters for the encryption process, and each image in the private person image set is encrypted using the encryption algorithm. The encrypted private person images are reintegrated into the original monitoring images to form an encrypted park monitoring image sequence.
[0038] Furthermore, step S3 of the present application also includes:
[0039] According to the historical surveillance images of the adjacent park, a plurality of sample encrypted park surveillance image sequences are collected, and suspicious persons are marked to obtain a plurality of sample suspicious appearance feature sets and a plurality of sample suspicious behavior feature sets, wherein the sample suspicious behavior features include suspicious behavior type, suspicious action speed and suspicious action direction; the plurality of sample encrypted park surveillance image sequences are respectively used as input data, and the plurality of sample suspicious appearance feature sets and the plurality of suspicious behavior feature sets are used as supervision data to train a plurality of suspicious feature identifiers; the plurality of suspicious feature identifiers are used to respectively perform suspicious person identification on the plurality of encrypted park surveillance image sequences, and obtain suspicious appearance features and suspicious action features in the encrypted park surveillance image sequence including suspicious person features.
[0040] Specifically, multiple sample encrypted park surveillance image sequences are collected from historical surveillance images of adjacent parks, suspicious persons are identified based on suspicious behaviors in historical surveillance images, and the identified suspicious person images are annotated, including suspicious appearance features and suspicious behavior features. The sample suspicious appearance feature set includes suspicious appearance features of suspicious persons, such as clothing, body shape, and items carried. The sample suspicious behavior feature set includes suspicious behavior features of suspicious persons, including behavior type, movement speed, and direction. The sample encrypted park surveillance image sequence is used as input data for training, and the sample suspicious appearance feature set and suspicious behavior feature set are used as supervision data. A suitable machine learning model is selected, such as support vector machine, random forest, etc., and the model is trained using the training data set until a satisfactory recognition effect is achieved. The trained model is tested using the validation set, and the model is tuned according to the test results. The trained suspicious feature identifier is used to identify suspicious persons in the encrypted park surveillance image sequence, identify suspicious persons in the image sequence, and extract their suspicious appearance features and suspicious behavior features.
[0041] Furthermore, step S4 of the present application also includes:
[0042] According to the suspicious appearance features, the park surveillance images including the suspicious appearance features are screened in the park surveillance image sequence to obtain a suspicious surveillance image sequence; according to the suspicious action features, it is determined whether the suspicious movement direction is an adjacent park. If so, the time frame in the park surveillance image of the suspicious person entering the adjacent park is calculated based on the suspicious action speed and the suspicious action direction as the suspicious movement time frame. If not, the suspicious movement time frame is not obtained.
[0043] Specifically, based on the extracted suspicious appearance features, images containing these features in the park monitoring image sequence are screened out. The features in the monitoring image sequence are compared with the suspicious appearance feature set, and based on the comparison results, the park monitoring images containing suspicious appearance features are screened out, and the screened images are combined into a suspicious monitoring image sequence. Based on the suspicious action speed and suspicious action direction, it is determined whether the suspicious person is moving to an adjacent park, the suspicious action features are analyzed, the suspicious movement direction is determined, and it is confirmed whether the suspicious person is moving to an adjacent park. The suspicious action features include suspicious action speed and suspicious action direction. If the suspicious person moves to an adjacent park, the time frame of the suspicious person entering the adjacent park is calculated based on the suspicious action speed and direction, and the suspicious movement time frame is marked in the suspicious monitoring image sequence.
[0044] Furthermore, step S5 of the present application also includes:
[0045] The cloud computing platform receives the suspicious monitoring image sequence and the suspicious movement time frame; the cloud computing platform sends the suspicious movement time frame to the park node of the adjacent park, and the park node of the adjacent park uploads the suspicious movement time frame in the encrypted park monitoring image sequence and the subsequent encrypted park monitoring images to the cloud computing platform for monitoring identification of monitoring data sharing.
[0046] Specifically, the cloud computing platform receives uploaded suspicious surveillance image sequences and suspicious movement time frames, and stores the received data in the cloud computing platform's storage services, such as object storage or block storage. The cloud computing platform sends the suspicious movement time frames to the park nodes of the adjacent parks to ensure that the clocks of the park nodes of the adjacent parks are synchronized so as to accurately track the movement of suspicious persons. The park nodes of the adjacent parks search for images of suspicious movement time frames and later in the encrypted park surveillance image sequence, and upload the found images to the cloud computing platform. The uploaded images remain encrypted during transmission to ensure data security. The cloud computing platform receives the uploaded encrypted images and shares the surveillance data. At the same time, the shared surveillance images are analyzed to identify suspicious persons.
[0047] In summary, the cloud computing-based campus monitoring data storage and sharing method provided by this application has the following technical effects:
[0048] The invention obtains a plurality of park monitoring image sequences by continuously collecting monitoring images based on monitoring equipment in adjacent parks respectively; performs privacy personnel identification and privacy personnel encryption processing on the plurality of park monitoring image sequences in the plurality of park nodes in the adjacent parks respectively, and obtains a plurality of encrypted park monitoring image sequences; performs suspicious personnel identification in the plurality of park nodes respectively according to the plurality of encrypted park monitoring image sequences, and obtains suspicious appearance features and suspicious action features; screens and obtains a suspicious monitoring image sequence including suspicious persons according to the suspicious appearance features, and determines the suspicious movement time frame of the adjacent park according to the suspicious action features; uploads the suspicious monitoring image sequence and the suspicious movement time frame to a cloud computing platform, and the cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification, which effectively solves the problem that the existing technology has poor security of data sharing, the risk of leakage of personnel privacy in different parks, and the low accuracy of identifying suspicious persons, which limits the overall performance of the monitoring system, realizes safe cross-park monitoring data storage and shared monitoring identification, improves the overall performance of the park monitoring system, and improves the efficiency and responsiveness of security monitoring.
[0049] Embodiment 2, based on the cloud computing-based park monitoring data storage and sharing method in the previous embodiment, the same inventive concept, this application also provides a cloud computing-based park monitoring data storage and sharing system, please refer to the attached Figure 2 The cloud computing-based park monitoring data storage and sharing system includes:
[0050] The multi-park monitoring image sequence acquisition module 11 is used to continuously collect monitoring images based on monitoring equipment in adjacent parks to obtain multiple park monitoring image sequences.
[0051] The encryption module 12 is used to perform privacy personnel identification and privacy personnel encryption processing on the multiple park monitoring image sequences in multiple park nodes in the adjacent parks, so as to obtain multiple encrypted park monitoring image sequences.
[0052] The personnel identification module 13 is used to identify suspicious persons in the multiple park nodes according to the multiple encrypted park monitoring image sequences, and obtain suspicious appearance characteristics and suspicious action characteristics.
[0053] The image screening module 14 is used to screen and obtain a suspicious surveillance image sequence including suspicious persons according to the suspicious appearance characteristics, and determine the suspicious movement time frame of the adjacent parks according to the suspicious action characteristics.
[0054] The monitoring identification module 15 is used to upload the suspicious monitoring image sequence and the suspicious movement time frame to the cloud computing platform. The cloud computing platform extracts the corresponding time frame and the subsequent encrypted park monitoring images in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification.
[0055] Furthermore, the encryption module 12 in the cloud computing-based park monitoring data storage and sharing system is also used for:
[0056] In the multiple park nodes in the adjacent park, personnel image segmentation and privacy personnel identification are performed on the park monitoring images in the multiple park monitoring image sequences to obtain multiple privacy personnel identification result sequences; based on the multiple privacy personnel identification result sequences, the privacy personnel images in the multiple park monitoring image sequences are encrypted to obtain multiple encrypted park monitoring image sequences.
[0057] Furthermore, the encryption module 12 in the cloud computing-based park monitoring data storage and sharing system is also used for:
[0058] According to the historical monitoring data of the park, a plurality of sample park monitoring image sets are collected, and the personnel images in each sample park monitoring image are segmented and labeled to obtain a plurality of sample personnel image segmentation result sets; the plurality of sample park monitoring image sets and the plurality of sample personnel image segmentation result sets are used to train a plurality of personnel image segmentors, and personnel image segmentation is performed on the park monitoring images in the plurality of park monitoring image sequences to obtain a plurality of personnel image segmentation results; based on the plurality of privacy personnel databases of the adjacent parks, a plurality of privacy personnel images are extracted, and a plurality of privacy personnel image identifiers are trained in the plurality of park nodes respectively; the plurality of privacy personnel image identifiers are used respectively to perform privacy personnel identification and labeling on the plurality of personnel image segmentation results to obtain a plurality of privacy personnel labeling result sequences.
[0059] Furthermore, the cloud computing-based park monitoring data storage and sharing system also includes an encrypted park image sequence acquisition module, which is used to:
[0060] According to the multiple privacy personnel identification result sequences, multiple privacy personnel image sets identified as privacy personnel are extracted; the privacy personnel images in the multiple park monitoring image sequences are encrypted to obtain multiple encrypted park monitoring image sequences.
[0061] Furthermore, the personnel identification module 13 in the cloud computing-based park monitoring data storage and sharing system is also used for:
[0062] According to the historical surveillance images of the adjacent park, a plurality of sample encrypted park surveillance image sequences are collected, and suspicious persons are marked to obtain a plurality of sample suspicious appearance feature sets and a plurality of sample suspicious behavior feature sets, wherein the sample suspicious behavior features include suspicious behavior type, suspicious action speed and suspicious action direction; the plurality of sample encrypted park surveillance image sequences are respectively used as input data, and the plurality of sample suspicious appearance feature sets and the plurality of suspicious behavior feature sets are used as supervision data to train a plurality of suspicious feature identifiers; the plurality of suspicious feature identifiers are used to respectively perform suspicious person identification on the plurality of encrypted park surveillance image sequences, and obtain suspicious appearance features and suspicious action features in the encrypted park surveillance image sequence including suspicious person features.
[0063] Furthermore, the image screening module 14 in the cloud computing-based park monitoring data storage and sharing system is also used for:
[0064] According to the suspicious appearance features, the park surveillance images including the suspicious appearance features are screened in the park surveillance image sequence to obtain a suspicious surveillance image sequence; according to the suspicious action features, it is determined whether the suspicious movement direction is an adjacent park. If so, the time frame in the park surveillance image of the suspicious person entering the adjacent park is calculated based on the suspicious action speed and the suspicious action direction as the suspicious movement time frame. If not, the suspicious movement time frame is not obtained.
[0065] Further, the monitoring identification module 15 in the cloud computing-based park monitoring data storage and sharing system is also used for:
[0066] The cloud computing platform receives the suspicious monitoring image sequence and the suspicious movement time frame; the cloud computing platform sends the suspicious movement time frame to the park node of the adjacent park, and the park node of the adjacent park uploads the suspicious movement time frame in the encrypted park monitoring image sequence and the subsequent encrypted park monitoring images to the cloud computing platform for monitoring identification of monitoring data sharing.
[0067] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. Figure 1 The cloud computing-based campus monitoring data storage and sharing method and specific examples in Example 1 are also applicable to the cloud computing-based campus monitoring data storage and sharing system in this example. Through the above detailed description of the cloud computing-based campus monitoring data storage and sharing method, those skilled in the art can clearly understand the cloud computing-based campus monitoring data storage and sharing system in this example, so for the sake of brevity, it will not be described in detail here. For the system disclosed in the example, since it corresponds to the method disclosed in the example, the description is relatively simple, and the relevant parts can be referred to the method description.
[0068] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
[0069] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.
Claims
1. A cloud computing-based park monitoring data storage and sharing method, characterized in that: The cloud computing-based park monitoring data storage and sharing method includes: In adjacent parks, monitoring images are continuously collected based on monitoring equipment to obtain multiple park monitoring image sequences; In the plurality of park nodes in the adjacent park, the plurality of park monitoring image sequences are respectively subjected to privacy personnel identification and privacy personnel encryption processing to obtain a plurality of encrypted park monitoring image sequences; In the plurality of park nodes, suspicious persons are identified respectively according to the plurality of encrypted park monitoring image sequences to obtain suspicious appearance features and suspicious action features; According to the suspicious appearance features, a suspicious surveillance image sequence including suspicious persons is screened and obtained, and according to the suspicious action features, a suspicious movement time frame of an adjacent park is determined; The suspicious monitoring image sequence and the suspicious movement time frame are uploaded to the cloud computing platform. The cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification; Wherein, in the plurality of park nodes in the adjacent park, the plurality of park monitoring image sequences are respectively subjected to privacy personnel identification and privacy personnel encryption processing, including: In the plurality of park nodes in the adjacent park, respectively performing personnel image segmentation and privacy personnel identification on the park monitoring images in the plurality of park monitoring image sequences to obtain a plurality of privacy personnel identification result sequences; According to the multiple privacy personnel identification result sequences, encrypting the privacy personnel images in the multiple park monitoring image sequences to obtain multiple encrypted park monitoring image sequences; Among them, in the plurality of park nodes in the adjacent park, the park monitoring images in the plurality of park monitoring image sequences are respectively subjected to personnel image segmentation and privacy personnel identification, including: According to the historical monitoring data of the park, a plurality of sample park monitoring image sets are collected, and the person images in each sample park monitoring image are segmented and labeled to obtain a plurality of sample person image segmentation result sets; Using the plurality of sample park surveillance image sets and the plurality of sample personnel image segmentation result sets, training a plurality of personnel image segmentors, performing personnel image segmentation on the park surveillance images in the plurality of park surveillance image sequences, and obtaining a plurality of personnel image segmentation results; Extracting multiple private person images according to multiple private person databases of the adjacent parks, and training multiple private person image recognizers in the multiple park nodes respectively; Using the multiple private person image identifiers respectively, perform private person identification and labeling on the multiple person image segmentation results to obtain multiple private person labeling result sequences; The method of encrypting the images of the private persons in the plurality of park monitoring image sequences according to the plurality of private persons identification result sequences comprises: Extracting a plurality of sets of images of private persons identified as private persons according to the plurality of private person identification result sequences; The private personnel images in the plurality of park monitoring image sequences are encrypted to obtain a plurality of encrypted park monitoring image sequences.
2. The cloud computing-based park monitoring data storage and sharing method according to claim 1 is characterized in that: In the plurality of park nodes, suspicious persons are identified respectively according to the plurality of encrypted park monitoring image sequences to obtain suspicious appearance features and suspicious action features, including: According to the historical surveillance images of the adjacent parks, multiple sample encrypted park surveillance image sequences are collected, and suspicious persons are marked to obtain multiple sample suspicious appearance feature sets and multiple sample suspicious behavior feature sets, wherein the sample suspicious behavior features include suspicious behavior type, suspicious action speed and suspicious action direction; The plurality of sample encrypted park monitoring image sequences are respectively used as input data, and the plurality of sample suspicious appearance feature sets and the plurality of suspicious behavior feature sets are used as supervision data to train a plurality of suspicious feature identifiers; The multiple suspicious feature identifiers are used to identify suspicious persons in the multiple encrypted park monitoring image sequences respectively, and suspicious appearance features and suspicious action features in the encrypted park monitoring image sequences including features of suspicious persons are obtained.
3. The cloud computing-based park monitoring data storage and sharing method according to claim 2 is characterized in that: According to the suspicious appearance features, a suspicious surveillance image sequence including suspicious persons is screened and obtained, and according to the suspicious action features, a suspicious movement time frame of an adjacent park is determined, including: According to the suspicious appearance features, a park monitoring image sequence is screened to obtain a park monitoring image including the suspicious appearance features, to obtain a suspicious monitoring image sequence; According to the suspicious action characteristics, determine whether the suspicious movement direction is an adjacent park. If so, calculate the time frame in the park monitoring image of the suspicious person entering the adjacent park based on the suspicious action speed and the suspicious action direction as the suspicious movement time frame. If not, do not obtain the suspicious movement time frame.
4. The cloud computing-based park monitoring data storage and sharing method according to claim 1 is characterized in that: The suspicious monitoring image sequence and the suspicious movement time frame are uploaded to the cloud computing platform. The cloud computing platform extracts the encrypted park monitoring images of the corresponding time frame and thereafter in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring identification for monitoring data sharing, including: The cloud computing platform receives the suspicious surveillance image sequence and the suspicious movement time frame; The cloud computing platform sends the suspicious movement time frame to the park node of the adjacent park, and the park node of the adjacent park uploads the suspicious movement time frame and the subsequent encrypted park monitoring images in the encrypted park monitoring image sequence to the cloud computing platform for monitoring identification of monitoring data sharing.
5. The cloud computing-based park monitoring data storage and sharing system is characterized by: The steps for implementing the cloud computing-based campus monitoring data storage and sharing method according to any one of claims 1 to 4, the cloud computing-based campus monitoring data storage and sharing system comprising: A multi-park monitoring image sequence acquisition module is used to continuously collect monitoring images based on monitoring equipment in adjacent parks to obtain multiple park monitoring image sequences; An encryption module is used to perform privacy personnel identification and privacy personnel encryption processing on the plurality of park monitoring image sequences in the plurality of park nodes in the adjacent park, respectively, to obtain a plurality of encrypted park monitoring image sequences; A personnel identification module is used to identify suspicious persons in the multiple park nodes according to the multiple encrypted park monitoring image sequences, and obtain suspicious appearance characteristics and suspicious action characteristics; An image screening module is used to screen and obtain a suspicious surveillance image sequence including suspicious persons according to the suspicious appearance characteristics, and determine the suspicious movement time frame of the adjacent park according to the suspicious action characteristics; The monitoring identification module is used to upload the suspicious monitoring image sequence and the suspicious movement time frame to the cloud computing platform. The cloud computing platform extracts the corresponding time frame and the subsequent encrypted park monitoring images in the encrypted park monitoring image sequence of the adjacent park according to the suspicious movement time frame, and performs monitoring data storage and shared monitoring identification.
