Intelligent management system based on mass image data

Abstract

The invention relates to the technical field of data processing, in particular to an intelligent management system based on mass image data, which comprises a receiver, a cloud database and a retrieval engine, the receiver obtains the monitoring image data interaction authority of the monitoring equipment, receives the monitoring image data of the monitoring equipment in real time after obtaining the monitoring image data interaction authority, and sends the monitoring image data to the cloud database in real time; the cloud database obtains the monitoring image data fed back by the receiver in real time and recognizes monitoring parameters of the monitoring equipment from the monitoring image data, centralized receiving of the monitoring image data of the multiple monitoring equipment is achieved by obtaining the authority of the monitoring equipment, and the monitoring image data can be obtained through setting and screening processing of monitoring image data screening logic. A large amount of monitoring image data is simplified into image data, so that when a system end user has a tracking requirement, a retrieval target trend can be quickly retrieved in the large amount of image data autonomously according to a retrieval target uploaded by the system end user based on the system.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent management system based on massive image data. Background Art

[0002] The video and image library management requirements of the safe city scenario involve multiple aspects, including ensuring the quality and stability of data transmission to support the transmission of high-definition, real-time video streams; solving compatibility issues to ensure smooth communication between multiple protocols and multiple devices; strengthening security measures to protect video and image data from being leaked or tampered with; at the same time, attention must be paid to data standardization and consistency to achieve cross-platform and cross-system data sharing and exchange; in the face of massive data, an efficient data storage and management mechanism must be established; finally, relevant laws, regulations and privacy protection policies must be strictly observed to ensure the legality of technology applications and the protection of user privacy.

[0003] At present, as urban residents' awareness of anti-theft and safety increases, most residents install monitoring equipment at the entrances of their homes, such as the entrance doors, to provide monitoring protection for their homes. The images collected by these monitoring devices are likely to capture some special images such as security incidents and arrest targets. These images are discovered as investigators speculate and track them, and are further retrieved and reviewed by investigators.

[0004] However, the process of investigators seeking to access surveillance footage is inefficient, and the speculation and tracking stages also take a certain amount of time;

[0005] To this end, we propose an intelligent management system based on massive image data. Summary of the invention

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides an intelligent management system based on massive image data, which solves the technical problems raised in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] An intelligent management system based on massive image data, comprising: a receiver, a cloud database and a retrieval engine;

[0009] The receiver obtains the monitoring image data interaction permission of the monitoring device. After obtaining the monitoring image data interaction permission, it receives the monitoring image data of the monitoring device in real time and sends it to the cloud database in real time; the cloud database obtains the monitoring image data fed back by the receiver in real time, identifies the monitoring parameters of the monitoring device from which the monitoring image data comes, sets the monitoring image data screening logic based on the monitoring parameters of the monitoring device, applies the monitoring image data screening logic to perform picture frame screening on the monitoring image data, further stores the screened picture frames, and makes identifiable marks on the picture frames; the search engine uploads the search target and the search target attribute parameters in real time, confirms the search target group in the cloud database based on the search target attribute parameters, and further performs search of the search target in the search target group.

[0010] Furthermore, the receiver includes a protocol unit, a receiving module, and a forwarding module:

[0011] The protocol unit is used to access the monitoring device user terminal and request the monitoring device monitoring image data interaction permission. The receiving module is used to obtain the monitoring device monitoring image data interaction permission request result in the protocol unit and perform real-time monitoring image data receiving operations on the authorized monitoring device. The forwarding module is used to forward the monitoring image data received in the receiving module to the cloud database.

[0012] Among them, the operation of accessing the monitoring device user end in the protocol unit and requesting the monitoring device monitoring image data interaction permission is to contact the monitoring device user end with a communication device and request the monitoring device monitoring image data interaction permission, or during the monitoring device installation stage, the monitoring device network access protocol includes the monitoring device monitoring image data interaction permission and is confirmed by the monitoring device user end.

[0013] Furthermore, after the protocol unit confirms the authority to interact with the monitoring image data between the receiver and the monitoring device, the monitoring device is deployed based on the optical fiber network, and receives the monitoring image data through the receiving module and transmits it to the receiver. When receiving the monitoring image data, the receiving module synchronously obtains the time information of the source monitoring device of the monitoring image data and the location information of the source monitoring device.

[0014] Furthermore, when forwarding the monitoring image data held in the receiver, the forwarding module synchronously sets the monitoring image data segmentation time domain, performs segmentation processing on the monitoring image data held in the receiver based on the set monitoring image data segmentation data, and further forwards the monitoring image data belonging to each time domain to the cloud database in sequence based on the time sequence;

[0015] Among them, the surveillance image data segmentation time domain set in the forwarding module is not less than five seconds and not more than fifteen seconds.

[0016] Furthermore, the cloud database includes an identification module, a screening module, and a storage module:

[0017] The identification module is used to receive the monitoring image data forwarded by the forwarding module, identify the monitoring parameters of the monitoring device of the monitoring image data source, the screening module is used to obtain the monitoring parameters of the monitoring device identified in the identification module, set the monitoring image data screening logic based on the monitoring parameters of the monitoring device, and screen the image frames of the monitoring image data of the monitoring device through the monitoring image data screening logic, and the storage module is used to receive the remaining monitoring image data screened by the screening module and store the monitoring image data;

[0018] Among them, the monitoring image data identified in the identification module comes from the monitoring parameters of the health equipment, that is, the monitoring range of the monitoring equipment, and the monitoring image data screening logic set in the screening module is:

[0019]

[0020] Where: S(α) is the estimated monitoring range of the monitoring device; d is the vertical distance from the monitoring device to the monitoring plane; f is the focal length of the monitoring device; p v 、p h S is the horizontal pixel span and vertical pixel span of any static object in the monitoring image data of the monitoring device; h , S v P is the horizontal and vertical size of the image sensor of the monitoring device; h , P v is the number of horizontal pixels and vertical pixels of the monitoring screen of the monitoring device; t is the time interval for picking up the screen frames in the monitoring image data; t0 is the base number of the picking up time interval; λ is a constant;

[0021] After the frame picking time interval t in the monitoring image data is obtained, the monitoring image data is segmented based on t to obtain several groups of sub-monitoring image data, and the last group of frames in each sub-monitoring image data is selected as the screening result of the screening module.

[0022] Furthermore, the constant λ≥1, and the picking time interval base t0 is manually set by the system end user;

[0023] After the screening module runs and screens the picture frames, it synchronously performs dynamic target recognition in continuous frames, and the picture frames of the recognized dynamic targets are used as the final picture frame screening results of the screening module;

[0024] Among them, when the screening module identifies the dynamic target in the continuous frames, it applies the background subtraction algorithm or the feature point tracking algorithm to identify the dynamic target.

[0025] Furthermore, the storage module is internally provided with submodules, including:

[0026] The marking unit is used to traverse the picture frames stored in the storage module, obtain the location information and time stamp of the source monitoring device of each picture frame, mark each corresponding picture frame based on the obtained location information and time stamp of the source monitoring device of the picture frame, and control the storage module to perform a differentiated storage once according to the location information of the monitoring device and perform a differentiated storage once according to the time period corresponding to the source time stamp for the internally stored picture frames;

[0027] The time period corresponding to the timestamp is in hours.

[0028] Furthermore, the search engine includes an upload module and a search unit:

[0029] The upload module is used to upload the search target and the search target attribute parameters, and use the search target attribute parameters to confirm the search target group in the cloud database. The search unit is used to obtain the search target group confirmed in the upload module, and search for a picture frame containing a dynamic target similar to the search target in the search target group.

[0030] Among them, one or several groups of retrieval targets uploaded in the upload module contain image data of people, and the retrieval target attribute parameters include: retrieval target source location, retrieval target source timestamp, and the retrieval unit retrieves the picture frames containing dynamic targets similar to the retrieval target and determines that the similarity is not less than 85%.

[0031] Furthermore, the search target group is composed of storage intervals distinguished in a plurality of groups of storage modules;

[0032] When the search target group is determined, the search target source position in the search target attribute parameters and the corresponding time period of the search target source timestamp are used to obtain a differentiated storage interval in the storage module that is the same as the search target source position and the corresponding time period of the search target source timestamp, so as to differentiate each picture frame in the storage interval as an object for similarity comparison with the search target, and further perform the search operation.

[0033] Furthermore, the receiver, the cloud database and the search engine are interactively connected via a wireless network;

[0034] The receiver is internally connected with a protocol unit through a wireless network, and the receiver is interactively connected with a receiving module and a forwarding module through a wireless network;

[0035] The cloud database is interactively connected to the identification module, the screening module and the storage module via a wireless network, and the storage module is interactively connected to the marking unit via a wireless network;

[0036] The search engine is interactively connected with an upload module and a search unit via a wireless network.

[0037] Compared with the known public technology, the technical solution provided by the present invention has the following beneficial effects:

[0038] The present invention provides an intelligent management system based on massive image data. During operation, the system realizes centralized reception of monitoring image data from multiple monitoring devices by acquiring monitoring device permissions, and further condenses a large amount of monitoring image data into image data by setting and filtering logic for monitoring image data. Therefore, when a system-side user has a tracking demand, based on the system, the system can autonomously and quickly retrieve the search target trend in massive image data based on the search target uploaded by the system-side user, thereby improving the system-side user's search target tracking efficiency and bringing convenience to the system-side user's tracking work. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 It is a structural schematic diagram of an intelligent management system based on massive image data;

[0041] Figure 2 This is an example diagram of an application example of the system in the present invention;

[0042] Figure 3 This is a diagram of an example application scenario and test environment of the system in the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] The present invention will be further described below in conjunction with the embodiments.

[0045] Embodiment 1:

[0046] This embodiment is an intelligent management system based on massive image data, such as Figure 1 As shown, it includes: a receiver, a cloud database and a search engine;

[0047] The receiver obtains the monitoring image data interaction authority of the monitoring device, and after obtaining the monitoring image data interaction authority, receives the monitoring image data of the monitoring device in real time and sends it to the cloud database in real time;

[0048] The receiver includes a protocol unit, a receiving module, and a forwarding module:

[0049] The protocol unit is used to access the monitoring device user terminal and request the monitoring device monitoring image data interaction permission. The receiving module is used to obtain the monitoring device monitoring image data interaction permission request result in the protocol unit and perform real-time monitoring image data receiving operations on the authorized monitoring device. The forwarding module is used to forward the monitoring image data received in the receiving module to the cloud database.

[0050] Among them, the operation of accessing the monitoring device user end in the protocol unit and requesting the monitoring device monitoring image data interaction permission is to contact the monitoring device user end with a communication device and request the monitoring device monitoring image data interaction permission, or in the monitoring device installation stage, the monitoring device network access agreement includes the monitoring device monitoring image data interaction permission and is confirmed by the monitoring device user end;

[0051] The cloud database obtains the monitoring image data fed back by the receiver in real time, identifies the monitoring parameters of the monitoring equipment of the monitoring image data source, sets the monitoring image data screening logic based on the monitoring parameters of the monitoring equipment, applies the monitoring image data screening logic to perform picture frame screening on the monitoring image data, further stores the screened picture frames, and marks the picture frames in an identifiable manner;

[0052] The cloud database includes identification module, screening module and storage module:

[0053] The identification module is used to receive the monitoring image data forwarded by the forwarding module, identify the monitoring parameters of the monitoring device of the monitoring image data source, the screening module is used to obtain the monitoring parameters of the monitoring device identified in the identification module, set the monitoring image data screening logic based on the monitoring parameters of the monitoring device, and screen the image frames of the monitoring image data of the monitoring device through the monitoring image data screening logic, and the storage module is used to receive the remaining monitoring image data screened by the screening module and store the monitoring image data;

[0054] Among them, the monitoring image data identified in the identification module comes from the monitoring parameters of the health equipment, that is, the monitoring range of the monitoring equipment, and the monitoring image data screening logic set in the screening module is:

[0055]

[0056] Where: S(α) is the estimated monitoring range of the monitoring device; d is the vertical distance from the monitoring device to the monitoring plane; f is the focal length of the monitoring device; p v 、p hS is the horizontal pixel span and vertical pixel span of any static object in the monitoring image data of the monitoring device; h , S v P is the horizontal and vertical size of the image sensor of the monitoring device; h , P v is the number of horizontal pixels and vertical pixels of the monitoring screen of the monitoring device; t is the time interval for picking up the screen frames in the monitoring image data; t0 is the base number of the picking up time interval; λ is a constant;

[0057] After the frame picking time interval t in the surveillance image data is obtained, the surveillance image data is segmented based on t to obtain a plurality of groups of sub-surveillance image data, and the last group of frames in each sub-surveillance image data is selected as the screening result of the screening module;

[0058] The storage module is internally provided with submodules, including:

[0059] The marking unit is used to traverse the picture frames stored in the storage module, obtain the location information and time stamp of the source monitoring device of each picture frame, mark each corresponding picture frame based on the obtained location information and time stamp of the source monitoring device of the picture frame, and control the storage module to perform a differentiated storage once according to the location information of the monitoring device and perform a differentiated storage once according to the time period corresponding to the source time stamp for the internally stored picture frames;

[0060] Among them, the time period corresponding to the timestamp is in hours;

[0061] The search engine uploads the search target and the search target attribute parameters in real time, confirms the search target group in the cloud database based on the search target attribute parameters, and further performs the search of the search target in the search target group;

[0062] The search engine includes an upload module and a search unit:

[0063] The upload module is used to upload the search target and the search target attribute parameters, and use the search target attribute parameters to confirm the search target group in the cloud database. The search unit is used to obtain the search target group confirmed in the upload module, and search for a picture frame containing a dynamic target similar to the search target in the search target group.

[0064] Among them, one or more groups of search targets uploaded in the upload module contain image data of people, and the search target attribute parameters include: search target source location, search target source timestamp, and the search unit searches for a picture frame containing a dynamic target similar to the search target and determines that the similarity is not less than 85%;

[0065] The receiver, the cloud database and the search engine are interactively connected via a wireless network;

[0066] The receiver is internally connected with a protocol unit through a wireless network, and the receiver is interactively connected with a receiving module and a forwarding module through a wireless network;

[0067] The cloud database is interactively connected to the identification module, the screening module and the storage module via a wireless network, and the storage module is interactively connected to the marking unit via a wireless network;

[0068] The search engine is interactively connected with an upload module and a search unit via a wireless network.

[0069] In this embodiment, the protocol unit runs to access the monitoring device user terminal and requests the monitoring device monitoring image data interaction permission. The receiving module synchronously obtains the monitoring device monitoring image data interaction permission request result in the protocol unit, and performs real-time monitoring image data receiving operations on the authorized monitoring device. The forwarding module post-operates to forward the monitoring image data received in the receiving module to the cloud database, and then the identification module receives the monitoring image data forwarded by the forwarding module, identifies the monitoring parameters of the monitoring device from which the monitoring image data comes, and the screening module further obtains the monitoring parameters of the monitoring device identified in the identification module, sets the monitoring image data screening logic based on the monitoring device monitoring parameters, and screens the monitoring image data of the monitoring device through the monitoring image data screening logic. The storage module receives the screening module in real time. The remaining monitoring image data is filtered in the block, and the monitoring image data is stored. During the operation stage of the storage module, the marking unit synchronously traverses the picture frames stored in the storage module, obtains the location information and timestamp of the monitoring device from which each picture frame comes, and marks each corresponding picture frame based on the obtained location information and timestamp of the monitoring device from which the picture frame comes. The storage module is controlled to perform a differentiated storage for the internally stored picture frames by the location information of the monitoring device, and a differentiated storage by the time period corresponding to the source timestamp. Finally, the retrieval target and the retrieval target attribute parameters are uploaded through the upload module, and the retrieval target attribute parameters are used to confirm the retrieval target group in the cloud database. The retrieval unit obtains the retrieval target group confirmed in the upload module, and retrieves the picture frames containing dynamic targets similar to the retrieval target in the retrieval target group.

[0070] Through the operation of the system in the above embodiment, an intelligent, fast and convenient target tracking technology is provided to the system end user;

[0071] See also Figure 2 As shown:

[0072] Figure 2 A brief description:

[0073] GA / T1400 is the standard for the public security video image information application system. Tianyi Vision's snapshot and push service is based on the public security GA / T1400 protocol standard. It will install 1400 cameras in public places, road checkpoints, and key monitoring areas to capture faces, vehicle size images, and structured data and push them to the public security platform to meet the public security's requirements for social video image resource management.

[0074] The Tianyi Video 1400 protocol is one of the important interface protocols in the video surveillance system. It defines the information interaction between platforms at all levels of the video surveillance system, including key links such as device registration, keep-alive, subscription, and feature attribute push. By following the Tianyi Video 1400 protocol, you can achieve seamless docking with the public security 1400 view library, dock with the public security superior platform with the standard 1400 protocol, and realize the processes of registration, keep-alive, issue subscriptions, and push images.

[0075] The following steps are used to implement the snapshot function:

[0076] 1. Device registration and keep alive: ensure normal communication between the lower-level platform and the upper-level platform.

[0077] 2. Subscription management: The lower-level platform configures the corresponding feature attribute push service based on the subscription request of the upper-level platform.

[0078] Feature attribute collection and push: The lower-level platform collects key target feature attributes in the video in real time, and pushes the feature attribute information to the upper-level platform according to the subscription information.

[0079] Embodiment 2:

[0080] In terms of specific implementation, based on Example 1, this example refers to Figure 1 The intelligent management system based on massive image data in Example 1 is further described in detail:

[0081] After the protocol unit confirms the receiver and the monitoring device's monitoring image data interaction authority, the monitoring device is deployed based on the optical fiber network, and receives the monitoring image data through the receiving module and transmits it to the receiver. When receiving the monitoring image data, the receiving module synchronously obtains the time and location information of the monitoring device from which the monitoring image data originated;

[0082] When forwarding the monitoring image data stored in the receiver, the forwarding module synchronously sets the monitoring image data segmentation time domain, performs segmentation processing on the monitoring image data stored in the receiver based on the set monitoring image data segmentation data, and further forwards the monitoring image data belonging to each time domain to the cloud database in sequence based on the time sequence;

[0083] Among them, the surveillance image data segmentation time domain set in the forwarding module is not less than five seconds and not more than fifteen seconds.

[0084] like Figure 1 As shown, the constant λ≥1, and the picking time interval base t0 is manually set by the system end user;

[0085] After the screening module runs and screens the picture frames, it synchronously performs dynamic target recognition in continuous frames, and the picture frames of the recognized dynamic targets are used as the final picture frame screening results of the screening module;

[0086] Among them, when the screening module identifies the dynamic target in the continuous frames, the background subtraction algorithm or the feature point tracking algorithm is applied to identify the dynamic target;

[0087] The search target group is composed of storage intervals distinguished in a plurality of storage modules;

[0088] When the search target group is determined, the search target source position in the search target attribute parameters and the corresponding time period of the search target source timestamp are used to obtain a differentiated storage interval in the storage module that is the same as the search target source position and the corresponding time period of the search target source timestamp, so as to distinguish each picture frame in the differentiated storage interval as an object for similarity comparison with the search target, and further perform the search operation.

[0089] In this embodiment, through the above settings, further operation logic support is provided for the system in Example 1, ensuring the stable operation of the system in Example 1 and providing image retrieval services for system end users.

[0090] In summary, during the operation of the system in the above embodiment, the centralized reception of monitoring image data from multiple monitoring devices is achieved by obtaining the permissions of the monitoring devices, and a large amount of monitoring image data is further simplified into image data through the setting and screening processing of the monitoring image data screening logic. Therefore, when the system-side user has a tracking demand, based on the system, it can autonomously search for the search target based on the search target uploaded by the system-side user, and quickly search for the search target trend in the massive image data, thereby improving the system-side user's search target tracking efficiency and bringing convenience to the system-side user's tracking work.

[0091] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent management system based on massive image data, characterized in that: include: Receiver, cloud database and search engine; The receiver obtains the monitoring image data interaction authority of the monitoring device, and after obtaining the monitoring image data interaction authority, receives the monitoring image data of the monitoring device in real time and sends it to the cloud database in real time; The cloud database obtains the monitoring image data fed back by the receiver in real time, identifies the monitoring parameters of the monitoring equipment of the monitoring image data source, sets the monitoring image data screening logic based on the monitoring parameters of the monitoring equipment, applies the monitoring image data screening logic to perform picture frame screening on the monitoring image data, further stores the screened picture frames, and marks the picture frames in an identifiable manner; The search engine uploads the search target and the search target attribute parameters in real time, confirms the search target group in the cloud database based on the search target attribute parameters, and further performs the search of the search target in the search target group.

2. The intelligent management system based on massive image data according to claim 1, characterized in that: The receiver includes a protocol unit, a receiving module, and a forwarding module: The protocol unit is used to access the monitoring device user terminal and request the monitoring device monitoring image data interaction permission. The receiving module is used to obtain the monitoring device monitoring image data interaction permission request result in the protocol unit and perform real-time monitoring image data receiving operations on the authorized monitoring device. The forwarding module is used to forward the monitoring image data received in the receiving module to the cloud database. Among them, the operation of accessing the monitoring device user end in the protocol unit and requesting the monitoring device monitoring image data interaction permission is to contact the monitoring device user end with a communication device and request the monitoring device monitoring image data interaction permission, or during the monitoring device installation stage, the monitoring device network access protocol includes the monitoring device monitoring image data interaction permission and is confirmed by the monitoring device user end.

3. The intelligent management system based on massive image data according to claim 2, characterized in that: After the protocol unit confirms the receiver and the monitoring device's authority to interact with monitoring image data, the monitoring device is deployed based on a fiber optic network, and receives the monitoring image data through a receiving module and transmits it to the receiver. When receiving the monitoring image data, the receiving module synchronously obtains the time information of the source monitoring device of the monitoring image data and the location information of the source monitoring device.

4. The intelligent management system based on massive image data according to claim 2, characterized in that: When forwarding the monitoring image data stored in the receiver, the forwarding module synchronously sets the monitoring image data segmentation time domain, performs segmentation processing on the monitoring image data stored in the receiver based on the set monitoring image data segmentation data, and further forwards the monitoring image data belonging to each time domain to the cloud database in sequence based on the time sequence; Among them, the surveillance image data segmentation time domain set in the forwarding module is not less than five seconds and not more than fifteen seconds.

5. The intelligent management system based on massive image data according to claim 1, characterized in that: The cloud database includes an identification module, a screening module, and a storage module: The identification module is used to receive the monitoring image data forwarded by the forwarding module, identify the monitoring parameters of the monitoring device of the monitoring image data source, the screening module is used to obtain the monitoring parameters of the monitoring device identified in the identification module, set the monitoring image data screening logic based on the monitoring parameters of the monitoring device, and screen the image frames of the monitoring image data of the monitoring device through the monitoring image data screening logic, and the storage module is used to receive the remaining monitoring image data screened by the screening module and store the monitoring image data; Among them, the monitoring image data identified in the identification module comes from the monitoring parameters of the health equipment, that is, the monitoring range of the monitoring equipment, and the monitoring image data screening logic set in the screening module is: Where: S(α) is the estimated monitoring range of the monitoring device; d is the vertical distance from the monitoring device to the monitoring plane; f is the focal length of the monitoring device; p v 、p h S is the horizontal pixel span and vertical pixel span of any static object in the monitoring image data of the monitoring device; h , S v P is the horizontal and vertical size of the image sensor of the monitoring device; h , P v is the number of horizontal pixels and vertical pixels of the monitoring screen of the monitoring device; t is the time interval for picking up the screen frames in the monitoring image data; t0 is the base number of the picking up time interval; λ is a constant; After obtaining the frame picking time interval t in the monitoring image data, the monitoring image data is segmented based on t to obtain several groups of sub-monitoring image data, and the last group of frames in each sub-monitoring image data is selected as the screening result of the screening module.

6. The intelligent management system based on massive image data according to claim 5, characterized in that: The constant λ≥1, and the picking time interval base t0 is manually set by the system end user; After the screening module runs and screens the picture frames, it synchronously performs dynamic target recognition in continuous frames, and the picture frames of the recognized dynamic targets are used as the final picture frame screening results of the screening module; Among them, when the screening module identifies the dynamic target in the continuous frames, it applies the background subtraction algorithm or the feature point tracking algorithm to identify the dynamic target.

7. The intelligent management system based on massive image data according to claim 5, characterized in that: The storage module is internally provided with submodules, including: The marking unit is used to traverse the picture frames stored in the storage module, obtain the location information and time stamp of the source monitoring device of each picture frame, mark each corresponding picture frame based on the obtained location information and time stamp of the source monitoring device of the picture frame, and control the storage module to perform a differentiated storage once according to the location information of the monitoring device and perform a differentiated storage once according to the time period corresponding to the source time stamp for the internally stored picture frames; The time period corresponding to the timestamp is in hours.

8. The intelligent management system based on massive image data according to claim 1, characterized in that: The search engine includes an upload module and a search unit: The upload module is used to upload the search target and the search target attribute parameters, and use the search target attribute parameters to confirm the search target group in the cloud database. The search unit is used to obtain the search target group confirmed in the upload module, and search for a picture frame containing a dynamic target similar to the search target in the search target group. Among them, one or several groups of retrieval targets uploaded in the upload module contain image data of people, and the retrieval target attribute parameters include: retrieval target source location, retrieval target source timestamp, and the retrieval unit retrieves the picture frames containing dynamic targets similar to the retrieval target and determines that the similarity is not less than 85%.

9. The intelligent management system based on massive image data according to claim 8, characterized in that: The search target group is composed of storage intervals distinguished in a plurality of storage modules; When the search target group is determined, the search target source position in the search target attribute parameters and the corresponding time period of the search target source timestamp are used to obtain a differentiated storage interval in the storage module that is the same as the search target source position and the corresponding time period of the search target source timestamp, so as to differentiate each picture frame in the storage interval as an object for similarity comparison with the search target, and further perform the search operation.

10. The intelligent management system based on massive image data according to claim 1, characterized in that: The receiver, cloud database and search engine are interactively connected via a wireless network; The receiver is internally connected with a protocol unit through a wireless network, and the receiver is interactively connected with a receiving module and a forwarding module through a wireless network; The cloud database is interactively connected to the identification module, the screening module and the storage module via a wireless network, and the storage module is interactively connected to the marking unit via a wireless network; The search engine is interactively connected with an upload module and a search unit via a wireless network.

Images