An edge computing method and mechanism based on LBS

Abstract

The present invention proposes an edge computing method and mechanism in the public security field based on LBS. The present invention installs a face acquisition terminal device on a mobile device and dynamically obtains location information through an identifier that accesses the network, transmits the location information to the cloud, and pushes the face information to the edge computing device based on the location update and the current adjacent area calculation, thereby performing face comparison and calculation at the edge node. The present invention can reduce the cloud computing cost, avoid the service quality degradation caused by long-distance round-trip delays, network congestion, etc., and can be widely applied to the needs of edge computing in mobile scenarios, provide faster response speed, and better ensure security and privacy.

Description

Technical Field

[0001] The present invention relates to the field of edge computing, and mainly to edge computing methods and mechanisms based on LBS (Location Based Services). Background Art

[0002] In traditional face recognition scenarios, face collection terminal devices (such as cameras, camcorders, etc.) transmit the collected face photos back to the cloud platform for comparison, recognition and business calculation, and feed back the calculation results to the front-end business layer or business personnel. This method can better support business needs in scenarios where the collected data transmission load is moderate and the timeliness requirements are relatively weak. However, with the rapid development of mobile Internet and Internet of Things technologies, the types and types of terminal IoT devices are becoming more and more diverse, and data traffic is showing an explosive exponential growth trend. Traditional face recognition business calculation methods have gradually failed to cope with the needs of real-time, flexible and multi-concurrency scenarios. The computing rate and timeliness of the cloud have become a major factor restricting business development. This situation can be alleviated to some extent by increasing service clusters and improving computing power, but it also brings about a substantial increase in system construction and operation and maintenance costs. How to solve this problem in a low-cost and effective way in a mobile computing environment has become one of the problems to be solved in public security face recognition applications.

[0003] Edge computing technology is defined as providing computing, storage and other services to users at the edge of the network closer to the data source, which provides a prerequisite for effectively solving this problem. If cloud computing technology is centralized big data processing, edge computing technology can be understood as edge big data processing. Edge computing technology has gradually been applied in the Internet of Things, AR / VR scenarios, and the big data and artificial intelligence industries.

[0004] The patent "Face recognition architecture design method based on edge computing and cloud coordination (CN 110197128 A)" provides a face recognition architecture with fast response time and high accuracy. It applies edge computing technology and cloud coordination to face recognition tasks under big data. At each edge node, local data is used to train the edge node face recognition model through task requests, and the data is transmitted with the basic task result data and collected to the cloud server. At the same time, the local models of each edge node are integrated to improve the overall face recognition accuracy. This method can effectively reduce the amount of front-end data transmission and reduce response delays, but it needs to upload data and models to each edge node for training and learning. It is more suitable for scenarios where each edge node is relatively fixed and the business calculation is relatively stable. It is not suitable for situations where the edge node position is continuously updated in mobile scenarios.

[0005] The patent "Face recognition method applied to edge computing devices (CN 110298296A)" proposes to load the face detection model and the face feature model in advance into the AI ​​chip of the edge computing device, perform face recognition on the input face picture, and use MTCNN's ONet to obtain face position information and key point coordinate information for the face snapshot captured by the front-end snapshot device, and compare and sort the captured face feature information with the face feature information of the base map loaded into the memory. This method can indeed solve the problems of long-distance round-trip delay, network congestion, and reduced service quality in the past to a certain extent, but there are several problems: 1) The implementation cost is high: it needs to rely on the implantation of a proprietary AI chip, and the cost of modifying or replacing the existing face acquisition equipment is high; 2) Position recognition depends on image processing and extraction, which requires a certain amount of computing and causes delays; 3) The base map in the memory does not mention how to dynamically update based on the position, and is not suitable for scenes where the position and base map face library change at the same time in mobile scenes, and there are certain limitations in its application.

[0006] Therefore, how to improve face recognition efficiency while reducing cloud computing is a problem that deserves further optimization and resolution. Summary of the invention

[0007] This Summary is provided to introduce some concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.

[0008] According to one embodiment of the present invention, a method for location-based edge computing is provided, comprising: obtaining a current location of an edge node; obtaining area information and a sub-area corresponding to the current location of the edge node; calculating a set of adjacent areas adjacent to the sub-area based on the area information and the sub-area corresponding to the current location of the edge node; receiving a face library subset and a target person information library subset corresponding to the sub-area and the adjacent area set; performing local face comparison on the collected face information of a current person based on the received face library subset; if the local face comparison is successful, searching the target person information library subset to determine whether the current person belongs to the target person; and performing business processing associated with the current person if the current person belongs to the target person.

[0009] According to another embodiment of the present invention, a system for location-based edge computing is provided, comprising: a cloud-based business platform, the cloud-based business platform storing and managing a face library and a target person information library, wherein the face library contains complete face information or a wider range of face information within an area associated with the cloud-based business platform, and the target person information library contains complete target person information or a wider range of target person information within an area associated with the cloud-based business platform; wherein the cloud-based business platform comprises: a location management module, the location management module being configured to manage area information of an area associated with the cloud-based business platform and to divide the area into a plurality of sub-areas based on a grid domain; a cloud location calculation module, the cloud location calculation module is configured to calculate the sub-area corresponding to the location information based on the received location information; a cloud face comparison module, the cloud face comparison module is configured to perform cloud face comparison based on the face information of the current person in combination with the face library; and a data push module, the data push module is configured to push a face library subset and a target person information library subset based on the received sub-area, the face library subset is extracted from the face library, the target person information library subset is extracted from the target person information library, and the face library subset and the target person information library subset only include content corresponding to the received sub-area.

[0010] According to another embodiment of the present invention, there is provided a computing device for location-based edge computing, comprising: a processor; a memory, wherein the memory stores instructions, and the instructions, when executed by the processor, can execute the method described above.

[0011] These and other features and advantages will become apparent by reading the following detailed description and by reference to the associated drawings.It is to be understood that the foregoing general description and the following detailed description are illustrative only and are not restrictive of the aspects of what is claimed. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to understand the manner in which the above features of the present invention are used in detail, the above briefly summarized contents can be described in more detail with reference to various embodiments, some of which are shown in the accompanying drawings. However, it should be noted that the accompanying drawings only show some typical aspects of the present invention and should not be considered to limit its scope, because the description may allow for other equally effective aspects.

[0013] Figure 1 A block diagram of a system 100 for location-based edge computing according to an embodiment of the present invention is shown;

[0014] Figure 2 A flowchart of a method 200 for location-based edge computing according to an embodiment of the present invention is shown;

[0015] Figure 3 A region diagram 300 is shown in the form of a grid;

[0016] Figure 4 A region connectivity simulation diagram 400 is shown; and

[0017] Figure 5 A block diagram of a computing device 500 is shown that illustrates a hardware device applicable to various aspects of the present invention according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be described in detail below in conjunction with the accompanying drawings, and the features of the present invention will be further revealed in the following specific description.

[0019] With the increase of face acquisition devices, the interaction between front-end data acquisition devices and the cloud is becoming more and more frequent, the amount of data uploaded is also increasing, and the computing power of the cloud is getting higher and higher. At this time, if the computing power of the cloud cannot be upgraded in time, it will often lead to business delays, network congestion, and reduced service quality, which becomes a limiting factor restricting business development. The present invention proposes a location-based edge computing method and mechanism in a mobile scenario. The present invention installs a face acquisition terminal device on a mobile device (for example, a mobile vehicle, etc.) and dynamically obtains location information through an identifier that accesses the network, and transmits the location information to the cloud. Based on the location update and the current adjacent area calculation, the face information is pushed to the edge computing device, thereby performing face comparison and calculation at the edge node. The present invention can reduce the cloud computing cost, avoid the previous long-distance round-trip delay, network congestion, etc. caused by the reduction in service quality, and can be widely applied to the needs of edge computing in mobile scenarios, providing a faster response speed, and better ensuring security and privacy.

[0020] Figure 1A block diagram of a system 100 for location-based edge computing according to an embodiment of the present invention is shown. The system 100 mainly includes a cloud business platform 101 and an edge node 102. According to an embodiment of the present invention, the cloud business platform 101 is a business management system deployed on a cloud server, which can support business requests of the edge node 102 and centralized data storage. For example, the stored data generally includes one or more of a face library 111, a business library 112, a location information library 113, a target personnel information library 114, and an edge device information library 115. According to an embodiment of the present invention, the edge node 102 may include a face collection device with a communication function, a communication device with a face collection function, or a combination of a face collection device and a communication device, such as a mobile vehicle (such as various mobile inspection equipment, a vehicle with a specific task, etc.). As known to those skilled in the art, the edge node 102 can access the cloud business platform 101 through different communication methods (such as Ethernet, Wi-Fi, 802.x, cellular).

[0021] By combining the cloud service platform 101 with the edge node 102, some tasks (e.g., local face comparison) can be performed on the edge node 102, while other tasks (e.g., face comparison not processed in the edge node 102, etc.) can be performed in the cloud service platform 101. Thus, by prepending part of the calculation to the edge node 102, the pressure of cloud data transmission can be reduced, while reducing the data storage cost.

[0022] refer to Figure 1 According to one embodiment of the present invention, the cloud business platform 101 includes one or more of an edge device management module 103, a location management module 104, a cloud location calculation module 105, a cloud face comparison module 106, a personnel information search module 107, a business association module 108, a data push module 109 and / or a configuration management module 110. Among them, any module can communicate with any other module, but all connections are not shown for the purpose of easy explanation. In addition, it is fully understood by those skilled in the art that this article only explains the above modules for the purpose of explanation, and the functions of one or more of the above modules can be merged into a single module or split into multiple modules. In addition, one or more of the above modules can be implemented in software, hardware or a combination thereof.

[0023] According to one embodiment of the present invention, the edge device management module 103 can be configured to manage information of all edge nodes 102 (such as edge nodes 102-1 to 102-N) that interact with the cloud business platform 101, including status information of each edge node in the edge node 102, configuration information of the edge node, etc., and store this information in the edge device information library 115.

[0024] According to an embodiment of the present invention, the edge device management module 103 may be further configured to perform online and offline management and remote control on the edge node 102. For example, if the current positions of two edge nodes 102 are too close, which may result in a high repetition rate of processed face information, one of the edge nodes 102 may be offlined to save resources.

[0025] According to one embodiment of the present invention, the cloud business platform 101 may be responsible for storing and managing data associated with a business-related area (for ease of explanation, "business-related area" and "area" may be used interchangeably hereinafter), so that the edge device management module 103 may manage the edge nodes 102 in the area and interact with the edge nodes 102 in the area. For example, assuming that the area associated with the cloud business platform 101 is set to City A, it may be configured to store data associated with City A (e.g., facial information of City A, map information of City A, etc.), thereby responding to requests issued by the edge nodes 102 in City A. Of course, the cloud business platform 101 may also be configured to be associated with a larger area or a smaller area as needed.

[0026] According to one embodiment of the present invention, the location management module 104 may be configured to manage the basic map, regional information, etc. of the business-related area associated with the cloud business platform 101. In addition, the location management module 104 may be configured to divide the area into multiple sub-areas based on the grid, and form a common mapping relationship between the sub-areas and the edge nodes. For example, if the business-related area associated with the cloud business platform 101 is City A, the location management module 104 may divide City A into multiple sub-areas based on administrative area division, geographic area size, target area division, etc. Further, in the case of target area division, since the target area to which the mobile device is assigned generally does not change frequently (for example, the mobile device may be specified to move within the target area), the location management module 104 may map the sub-area with the edge node 102 on the mobile device moving in the sub-area and maintain the mapping relationship, so that it can respond more quickly when the edge node 102 makes an area search request.

[0027] According to another embodiment of the present invention, the above-mentioned basic map, area information, mapping relationship, etc. may be stored in the location information library 113 .

[0028] According to one embodiment of the present invention, the cloud location calculation module 105 may be configured to search for the corresponding sub-area based on the location of a certain edge node 102 or a set of edge nodes 102 based on the area search algorithm. According to one embodiment of the present invention, the area search algorithm may search for the sub-area corresponding to the current location of the edge node 102. For example, if the location of the edge node 102 is identified as Point-N {longitude A, latitude B}, the cloud location calculation module 105 may calculate the sub-area where the location Point-N is located, for example, based on the location information library 113.

[0029] According to one embodiment of the present invention, the area search algorithm is executed by the cloud, and retrieves the sub-area where the location is located based on the location reported by the edge node, while the area search algorithm described below is executed by the edge node, and is used to calculate the adjacent area based on the sub-area returned by the cloud.

[0030] According to one embodiment of the present invention, the cloud face comparison module 106 may be configured to process, identify and / or compare faces in the cloud based on the face information uploaded by the edge node 102 in combination with the face library 111. According to one embodiment of the present invention, the face library 111 included in the cloud business platform 101 may include complete face information or a wider range of face information within the business area associated with the cloud business platform 101. According to one embodiment of the present invention, the "face information" described in the present invention generally includes face data or face training data.

[0031] For example, for a certain person, the face database 111 may include the face information of the person. Further, the face information may include one or more of the following fields: face number, face key feature-1, face key feature-2, ..., face key feature-N, face photo position index-1, face photo position index-2, ..., face photo position index-N, face identification index tag (e.g., vector index of image feature vector).

[0032] Those skilled in the art will appreciate that the methods for processing, identifying and comparing faces have become increasingly mature, and the specific face comparison method is not within the scope of protection of the present invention.

[0033] According to one embodiment of the present invention, the personnel information search module 107 may be configured to retrieve personnel based on the results of the face comparison of the cloud face comparison module 106 in combination with the target personnel information library 114. According to one embodiment of the present invention, the target personnel information library 114 may only include the personnel information of the target personnel (for example, missing persons, lost children, etc.) defined in advance, or the target personnel information library 114 may only include the personnel information of the target personnel defined in other ways. According to an implementation example of the present invention, the target personnel information library 114 may contain complete target personnel information or a wider range of target personnel information within the business area associated with the cloud business platform 101. According to one embodiment of the present invention, the personnel information may include one or more of the following fields: personnel ID, basic personnel information, personnel tags and / or related business.

[0034] For example, if the face comparison result of the cloud face comparison module 106 is that the face currently captured by the edge node 102 matches a person in the face library 114, the person is searched in the target person information library 114 to determine whether the person is the target person for the next step of processing.

[0035] According to one embodiment of the present invention, the business association module 108 can be configured to associate specific personnel information with actual business, and the business in the module can be expanded and configured according to actual conditions. According to one embodiment of the present invention, "actual business" can be defined by public institutions such as community organizations and banks, so that when a person is identified through face matching, related subsequent specific business can be carried out. For example, if the person is a lost child or a missing person, the specific business associated with it can be reporting to relevant public institutions.

[0036] According to one embodiment of the present invention, the specific items and contents of "actual business" may be maintained in a business library, and the target personnel information library 114 may include an index linking a specific person to a specific business. According to another embodiment of the present invention, the target personnel information library 114 may also directly include specific business for a specific person without using a business library.

[0037] According to one embodiment of the present invention, the data push module 109 is configured to receive the data request and location information of the edge node 102, and push the face library subset and the target person information library subset to the edge node 102 according to the request of the edge node 102. The face library subset and the target person information library subset are extracted from the face library 111 and the target person information library 114, and are only related to the following two types of areas: (1) the sub-area corresponding to the location information of the edge node 102 (for example, the sub-area calculated by the cloud location calculation module 105); (2) the adjacent area of ​​the sub-area. It should be noted that the face data in the face library subset only includes the face data of the target person defined by the organization, and does not need to include the face data or face training data of all residents in the sub-area corresponding to the location information of the edge node 102 and the adjacent area. In addition, the target person information in the target person information library subset only includes the information of the target person corresponding to the sub-area and the adjacent area. According to one embodiment of the present invention, if the target person information library subset is large, it can be further filtered according to the administrative area and / or the level of the target person.

[0038] According to an embodiment of the present invention, the configuration management module 110 is configured as a basic system information management module. Specifically, the configuration management module 110 is configured to perform location initialization, edge node basic information initialization and setting, and basic parameter setting such as search depth.

[0039] Now further reference Figure 1 1 to describe the edge node 102. As shown in the figure, the cloud service platform 101 can manage and communicate with a plurality of edge nodes 102 (eg, edge nodes 102-1 to 102-N). According to one embodiment of the present invention, the plurality of edge nodes 102 can be located in different sub-regions or in the same sub-region.

[0040] According to one embodiment of the present invention, the edge node 102 may include a position acquisition module 116, a portrait acquisition module 117, a local position calculation module 118 and a local face comparison module 119. According to another embodiment of the present invention, the edge node 102 may store a face library subset 120 and a target person information library subset 121.

[0041] According to an embodiment of the present invention, the location acquisition module 116 is configured to obtain the current location information of the edge node 102. For example, the location acquisition module 116 may obtain the current location information based on the GPS information of the current edge node 102 or using the access network identifier provided by the base station.

[0042] According to an embodiment of the present invention, the portrait acquisition module 117 is configured to obtain face information within the field of view of the portrait acquisition module 117. For example, the portrait acquisition module 117 may be embodied as a camera or the like.

[0043] According to one embodiment of the present invention, the local location calculation module 118 is configured to calculate the adjacent area adjacent to the sub-area where the edge node 102 is currently located according to the area search algorithm (for example, the sub-area is provided by the cloud business platform 101).

[0044] According to one embodiment of the present invention, the local face comparison module 119 is configured to compare the face image of the current person obtained by the portrait acquisition module 117 with the face library subset 120, and search in the target person information library subset 121 based on the comparison result to determine whether the current person is the target person.

[0045] Figure 2 A flow chart of a method 200 for location-based edge computing according to an embodiment of the present invention is shown. The method 200 starts at step 201 .

[0046] In step 201, the current location of the edge node is obtained. According to one embodiment of the present invention, the edge node is initialized and obtains the current location through an identification of the access network or GPS.

[0047] In step 202, the edge node obtains regional information and a sub-region corresponding to the current position of the edge node. According to one embodiment of the present invention, the edge node transmits the current position to the cloud service platform 101, and the cloud service platform 101 returns the regional information of the area associated with the cloud service platform 101 (for example, the regional map, regional division, etc. from the location information library 113) and the sub-region corresponding to the current position (for example, the sub-region where the edge node is located) to the edge node. For example, if the area associated with the cloud service platform 101 is Shanghai, and the current position identifier transmitted by the edge node is Point-N {longitude A, latitude B}, the cloud service platform 101 can return the regional information of Shanghai and the sub-region corresponding to Point-N to the edge node.

[0048] According to another embodiment of the present invention, the region information of the region associated with the cloud service platform 101 may also be stored in the edge node to facilitate faster sub-region positioning.

[0049] In step 203, based on the area information obtained in step 202 and the sub-area corresponding to the current position of the edge node, a set of adjacent areas adjacent to the sub-area is calculated.

[0050] According to one embodiment of the present invention, the edge node calculates the adjacent region set according to the region search algorithm. Figure 3 and Figure 4 to be described in detail. Figure 3 The region diagram 300 is shown in the form of a grid. Figure 4 A region connectivity simulation graph 400 is shown.

[0051] Assume that the area associated with the cloud service platform 101 is divided into n sub-areas, such as Figure 3 As shown in the figure (for example, the area is divided into 38 sub-areas), each sub-area is marked as Loc-i, and its adjacent area is marked as NearA{i}={Loc-m,Loc-n,Loc-j....}, where m,n,j belong to i, then Loc-i can be expressed as Loc-i=[i,NearA{i}]. The graphical modeling of {Loc-1, Loc-2, Loc-3....Loc-n} is obtained as follows Figure 4 Assume that the location of the current edge node N is marked as Point-N {longitude A, latitude B}, upload Point-N information to the cloud service platform 101 for positioning, and obtain its corresponding sub-area as Loc-k ( Figure 3 , if the current search depth is set to t, the search algorithm is described as follows:

[0052]

[0053] like Figure 4 As shown, according to the above algorithm, if the search depth is 2, the adjacent areas corresponding to sub-area 14 are {7, 8, 9, 10, 11, 12, 13, 20, 19, 21, 22, 15, 18}. Those skilled in the art will appreciate that the number of adjacent areas searched depends on the depth of the area search algorithm. The greater the depth, the greater the number of adjacent areas searched. According to one embodiment of the present invention, as described above, the depth of the area search algorithm can be configured by the cloud business platform 101.

[0054] In step 204 , the adjacent region set is uploaded to the cloud service platform. According to an embodiment of the present invention, as in the above example, {7, 8, 9, 10, 11, 12, 13, 20, 19, 21, 22, 15, 18} may be uploaded to the cloud service platform 101 .

[0055] In step 205, the cloud business platform calculates a face library subset and a target person information library subset from the face library and the target person information library, and the face library subset and the target person information library subset only include content corresponding to the following two types of areas: (1) a sub-area corresponding to the current location of the edge node; (2) a set of adjacent areas of the sub-area. As described above, the face library included in the cloud business platform may include complete face information or face information of a wider range within the business area associated with the cloud business platform. The target person information library may include complete target person information or target person information of a wider range within the business area associated with the cloud business platform.

[0056] In step 206, the cloud service platform sends the face database subset and the target person information database subset to the edge node. According to one embodiment of the present invention, the cloud calculates and generates the relevant face database and the relevant business database according to the business needs and sends the data. The sent face database subset and the target person information database subset can be sent based on the association rule, and the association rule can be as follows:

[0057] (region number, region name, region identification boundary list {<longitude-1, latitude-1>, <longitude-2, latitude-2>, <longitude-3, latitude-3>, ..., <longitude-N, latitude-N>}}, face identification index label, target person label)

[0058] Target person (personnel ID, basic information of the person (......), person tag list (target person tag 1, target person tag 2, target person tag 3))

[0059] Face record (face number, face key feature-1, face key feature-2, ... face key feature-N, face photo position index-1, face photo position index-2, ... face photo position index-N, face identification index label)

[0060] In step 207, the edge node performs local face comparison on the collected face information of the current person based on the received face library subset. According to one embodiment of the present invention, the face similarity is set to φ. If the face comparison algorithm can obtain that the difference between the similarity of the currently collected face Fcur and the person P in the face library subset is less than φ, it can be determined that the face comparison is successful. If the local face comparison is successful, step 208 is entered. If the local face comparison is unsuccessful, step 210 is entered.

[0061] In step 208, the target person information library subset is searched to determine whether the current person P belongs to the target person. In step 209, if the current person P belongs to the target person, the current person is directly processed for business. According to one embodiment of the present invention, the edge node can perform business processing based on the business associated with the current person recorded in the target person information library subset or the business library.

[0062] In step 210, the facial information of the current person is transmitted to the cloud service platform, and the cloud service platform performs cloud face comparison and business processing. According to one embodiment of the present invention, the cloud service platform compares the facial information with the face library. If the cloud face comparison is successful, the current person is searched in the target person information library to determine whether the current person belongs to the target person, and if so, the corresponding business processing is performed.

[0063] In step 211, the cloud service platform feeds back the cloud face comparison result and / or service processing result to the edge node. According to another embodiment of the present invention, this step is optional, that is, the cloud service platform does not need to feed back the processing result to the edge node.

[0064] Compared with the prior art, the present invention has the following significant advantages:

[0065] (1) Practicality: According to the present invention, the edge node can make dynamic requests based on the actual business scenario and location, and by pre-placing the calculation, it reduces the pressure of cloud data transmission and reduces the data storage cost. At the same time, the method of the present invention does not require the modification of edge camera equipment, and the promotion and application cost is low.

[0066] (2) Wide range of application scenarios: The present invention can be applied to scenarios where edge nodes are relatively fixed, as well as scenarios where edge nodes are constantly updating their locations in a mobile environment. By initiating task requests based on location by edge nodes, the present invention can significantly reduce the data uploaded to the cloud and reduce service latency.

[0067] Figure 5 A block diagram of a computing device 500 of a hardware device applicable to various aspects of the present invention according to an embodiment of the present invention is shown. The computing device 500 can be used as the edge node 102 or the cloud service platform 101 in the present invention.

[0068] refer to Figure 5The computing device 500 may be any machine that can be configured to perform processing and / or computing, and may be, but is not limited to, a workstation, a server, a desktop computer, a laptop computer, a tablet computer, a personal digital assistant, a smart phone, a vehicle-mounted computer, a home camera, a conference recording device, a road camera, or any combination thereof. The aforementioned various methods / devices / servers / client devices may be implemented in whole or in part by the computing device 500 or a similar device or system.

[0069] The computing device 500 may include components that can be connected or communicated via one or more interfaces and a bus 502. For example, the computing device 500 may include a bus 502, one or more processors 504, one or more input devices 506, and one or more output devices 308. The one or more processors 504 may be any type of processor and may include, but are not limited to, one or more general-purpose processors and / or one or more special-purpose processors (e.g., a dedicated processing chip). The input device 506 may be any type of device capable of inputting information to the computing device and may include, but are not limited to, a mouse, a keyboard, a touch screen, a microphone, a camera, and / or a remote controller. The output device 508 may be any type of device capable of presenting information and may include, but are not limited to, a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The computing device 500 may also include or be connected to a non-transient storage device 510, which may be any storage device that is non-transient and capable of data storage, and may include, but is not limited to, a disk drive, an optical storage device, a solid-state memory, a floppy disk, a floppy disk, a hard disk, a tape or any other magnetic medium, an optical disk or any other optical medium, a ROM (read-only memory), a RAM (random access memory), a cache memory, and / or any memory chip or cassette, and / or any other medium from which a computer can read data, instructions, and / or code. The non-transient storage device 510 may be detachable from the interface. The non-transient storage device 510 may have data / instructions / code for implementing the above methods and steps. The computing device 500 may also include a communication device 512. The communication device 512 can be any type of device or system that can communicate with internal devices and / or with a network and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication device and / or a chipset, such as a Bluetooth device, an IEEE1302.11 device, a WiFi device, a WiMax device, a cellular communication device and / or the like.

[0070] The bus 502 may include, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0071] The computing device 500 may also include a working memory 514 , which may be any type of working memory capable of storing instructions and / or data that facilitate the operation of the processor 504 and may include, but is not limited to, a random access memory and / or a read-only storage device.

[0072] Software components may be located in the working memory 514, including but not limited to an operating system 516, one or more application programs 518, drivers and / or other data and codes. Instructions for implementing the above methods and steps of the present invention may be included in the one or more application programs 518, and the processor 504 may read and execute the instructions of the one or more application programs 518 to implement the above method 200 of the present invention.

[0073] It should also be recognized that changes can be made according to specific needs. For example, custom hardware can also be used, and / or specific components can be implemented in hardware, software, firmware, middleware, microcode, hardware description language or any combination thereof. In addition, connections with other computing devices, such as network input / output devices, etc. can be adopted. For example, programming hardware (e.g., programmable logic circuits including field programmable gate arrays (FPGAs) and / or programmable logic arrays (PLAs)) with assembly language or hardware programming languages ​​(e.g., VERILOG, VHDL, C++) can be used to implement part or all of the disclosed methods and devices using logic and algorithms according to the present invention.

[0074] Although various aspects of the present invention have been described so far with reference to the accompanying drawings, the above-described methods, systems and devices are merely examples, and the scope of the present invention is not limited to these aspects, but is limited only by the appended claims and their equivalents. Various components may be omitted or replaced by equivalent components. In addition, the steps may be implemented in an order different from the order described in the present invention. In addition, various components may be combined in various ways. It is also important that, as technology develops, many of the components described may be replaced by equivalent components that appear later.

Claims

1. A method for location-based edge computing, comprising: Get the current position of the edge node; Obtaining area information and a sub-area corresponding to the current position of the edge node; Based on the area information and the sub-area corresponding to the current position of the edge node, calculate a set of adjacent areas adjacent to the sub-area; Receiving a face library subset and a target person information library subset corresponding to the sub-region and the adjacent region set; Based on the received subset of the face library, perform local face comparison on the collected face information of the current person; If the local face comparison is successful, searching the target person information database subset to determine whether the current person belongs to the target person; and In the case where the current person is a target person, a business process associated with the current person is executed.

2. The method according to claim 1, characterized in that Obtaining the area information and the sub-area corresponding to the current position of the edge node further includes: Transmitting the current location to a cloud service platform; and The cloud service platform returns the region information and a sub-region corresponding to the current position of the edge node to the edge node, wherein the region information corresponds to a region associated with the cloud service platform, and the region is divided into a plurality of sub-regions.

3. The method according to claim 1, characterized in that Receiving a face library subset and a target person information library subset corresponding to the sub-region and the adjacent region set further includes: Uploading the adjacent area set to the cloud service platform; The cloud service platform calculates the face library subset and the target person information library subset from the face library and the target person information library, and the face library subset and the target person information library subset only include content corresponding to the following two types of areas: (1) the sub-area; (2) the adjacent area set; The face database at the cloud business platform contains complete face information or a wider range of face information within the area associated with the cloud business platform, and the target person information database contains complete target person information or a wider range of target person information within the area associated with the cloud business platform.

4. The method according to claim 1, characterized in that The method further comprises: If the local face comparison is unsuccessful, the collected face information of the current person is transmitted to the cloud service platform; The cloud service platform performs a cloud face comparison on the face information with a face database, wherein the face database contains complete face information or face information of a wider range in an area associated with the cloud service platform; If the cloud face matching is successful, searching the target person information database for the current person to determine whether the current person belongs to the target person, wherein the target person information database contains complete target person information or a wider range of target person information in the area associated with the cloud business platform; and In the case where the current person is a target person, a business process associated with the current person is executed.

5. The method according to claim 4, characterized in that The method further comprises: The cloud service platform feeds back the cloud face comparison result to the edge node; and / or The cloud service platform feeds back the result of the service processing to the edge node.

6. A system for location-based edge computing, comprising: A cloud-based business platform, wherein the cloud-based business platform stores and manages a face database and a target person information database, wherein the face database contains complete face information or face information of a wider range within an area associated with the cloud-based business platform, and the target person information database contains complete target person information or target person information of a wider range within an area associated with the cloud-based business platform; The cloud business platform includes: a location management module, the location management module being configured to manage area information of an area associated with the cloud service platform and to divide the area into a plurality of sub-areas based on a grid; A cloud location calculation module, wherein the cloud location calculation module is configured to calculate a sub-area corresponding to the location information based on the received location information; A cloud face comparison module, wherein the cloud face comparison module is configured to perform cloud face comparison based on the face information of the current person in combination with the face database; and A data push module, wherein the data push module is configured to push a face library subset and a target person information library subset based on a received sub-area, wherein the face library subset is extracted from the face library, the target person information library subset is extracted from the target person information library, and the face library subset and the target person information library subset only include content corresponding to the received sub-area.

7. The system according to claim 6, characterized in that The cloud business platform further includes: The personnel information search module is configured to search the target personnel information database according to the result of the cloud face comparison to determine whether the current person belongs to the target person.

8. The system according to claim 6, characterized in that The system further comprises one or more edge nodes, each of the one or more edge nodes comprising: A location acquisition module, wherein the location acquisition module is configured to obtain a current location of the edge node; A portrait acquisition module, wherein the portrait acquisition module is configured to obtain facial information of a current person; A local location calculation module, the local location calculation module is configured to calculate a set of adjacent areas adjacent to the sub-area based on the sub-area where the edge node is currently located; and A local face comparison module is configured to perform a face comparison between the face information of the current person and the subset of the face library, wherein the subset of the face library only includes content corresponding to the sub-region.

9. The system according to claim 8, characterized in that Each of the one or more edge nodes is mounted on a mobile tool.

10. A computing device for location-based edge computing, comprising: processor; A memory storing instructions, wherein the instructions, when executed by the processor, can execute the method according to any one of claims 1 to 5.

Images