A novel identification and filtering method and system in an intelligent security system

Abstract

The present invention discloses a recognition filtering method and system in a novel intelligent security system, wherein the method comprises: step 1: any edge end camera device collects image information of personnel and vehicles, and uploads the image information and corresponding device information to an application server of a computing center through a high-speed network; step 2: after the application server calls a face recognition program and a license plate recognition program to recognize the image information, the recognition result with the corresponding device information is transmitted to a network server of the computing center; step 3: the network server filters and removes duplicates from the recognition result before performing business processing on the recognition result, and stores the filtered and deduplicated result in a database.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an identification and filtering method and system in a novel intelligent security system, and more particularly to a technology for filtering and processing duplicate data identified in the novel intelligent security system. Background Art

[0002] With the continuous development of smart cities, face or license plate detection and recognition technologies have been widely used in urban security and other related fields. Taking face recognition as an example, a complete face recognition process generally includes two parts: face detection and face recognition. Face detection refers to the process in which a computer detects a face in an image containing a face and gives information such as the location and size of the face area; face recognition compares the face to be recognized with a known face (such as a face stored in a database) to obtain relevant information on the degree of similarity.

[0003] At present, many cities have installed security cameras in every corner, and a large number of front-end camera devices will collect a large amount of data, which includes a certain amount of invalid data and duplicate data. For example, if a person stays at a certain collection point for a period of time, this person will be identified and recorded in the database many times during this period. If these invalid or duplicate data are processed or stored together with valid data in a preset way, the storage space of valid data will be occupied, which will cause a waste of resources to a certain extent and reduce the efficiency of the server in processing valid data. Therefore, it is particularly important to filter and deduplicate the data.

[0004] At present, there are three main methods for filtering and deduplicating data: 1) manual intervention, which is time-consuming and labor-intensive, and the security and validity of the data cannot be guaranteed; 2) database query deduplication, which filters the data based on the storage time. This method requires repeated queries of the database, which not only has low processing efficiency and long processing time, but also affects the real-time nature of the data. In addition, the code business logic used in this method is relatively complex; 3) fixed-time data deduplication, which detects that a certain camera has data and sets a fixed time to stop receiving new data. This method can easily cause data loss. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a new identification and filtering method and system in an intelligent security system, which sends the personnel information, vehicle information, etc. collected by front-end cameras and other equipment in the urban security system to the cloud, and cleans and filters the data within a set time period to remove duplication before storing it in the database, so as to reduce the waste of resource usage and improve the availability of stored data.

[0006] To achieve the above object, the present invention provides a novel identification and filtering method in an intelligent security system, which comprises the following steps:

[0007] Step 1: Any edge camera device collects image information of people and vehicles, and uploads the image information and corresponding device information to the application server of the computing center through a high-speed network;

[0008] Step 2: After the application server calls the face recognition program and the license plate recognition program to recognize the image information, it transmits the recognition result with the corresponding device information to the network server of the computing center;

[0009] Step 3: The network server filters and removes duplicates from the recognition results before performing business processing on the recognition results, and stores the filtered and duplicated results in the database.

[0010] In one embodiment of the present invention, the identification result with corresponding device information in step 2 includes: an identification code, identification type data and a unique device identifier of the corresponding device.

[0011] In one embodiment of the present invention, the specific process of filtering and deduplicating the recognition results in step 3 is as follows:

[0012] Step 301: Generate a key value of the data to be processed, key=rcg_id+stranger+cameraid, from any recognition result of step 2, where rcg_id is the recognition identification code, stranger is the recognition type data, and cameraid is the unique device identifier of the corresponding device;

[0013] Step 302: Use the key value to search and obtain data from the system's built-in redis cache.

[0014] If no data with the same key value is found and obtained in the redis cache, proceed to step 303;

[0015] Otherwise, proceed to step 305;

[0016] Step 303: store the key value generated in step 301 into the redis cache and reset the data validity period, then transmit the data to the business program for processing, store the data in the real-time data table and trigger the plan event;

[0017] Step 304: store the plan event information into the plan information table, and encapsulate the corresponding recognition result data and plan event information into a JSON data packet and send it to the client through websocket for page display;

[0018] Step 305: Determine that the identification result data of step 301 is invalid data, refresh the cache data of the corresponding key value in the redis cache and update the data validity time;

[0019] Step 306: Give up reporting the corresponding recognition result data to the business program, and return to step 301 to continue filtering and deduplicating the next recognition result.

[0020] To achieve the above object, the present invention also provides a novel identification and filtering system in an intelligent security system, which is used to implement the above identification and filtering method, and comprises:

[0021] Redis cache is used to cache the key value and data validity period corresponding to the recognition result data;

[0022] A processing unit, connected to the redis cache, for processing the key value and data validity time of the recognition result data;

[0023] Among them, the identification and filtering system is integrated into the network server and connected to the application server and the data processing module of the network server respectively, and is used to filter and de-duplicate the identification result data processed by the application server and transmit the de-duplicated data to the data processing module.

[0024] Compared with the prior art, the present invention solves the problem of excessive and duplicate data in the data collected by the front-end acquisition device by adding an identification and filtering method and system in the new intelligent security system, realizes the filtering and deduplication of duplicate data, reduces the waste of equipment resources, and improves the accuracy of stored data. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] 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 description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 It is an overall flow chart of identification and filtering in a security system according to an embodiment of the present invention;

[0027] Figure 2 A flowchart of filtering and deduplicating data in one embodiment of the present invention;

[0028] Figure 3 A security system architecture diagram including a filtering and deduplication system according to an embodiment of the present invention;

[0029] Figure 4 The structure diagram of the filtering and deduplication system according to an embodiment of the present invention.

[0030] Explanation of reference numerals: 1 - application server; 2 - network server; 40 - identification and filtering system; 401 - redis cache; 401 - processing unit; 50 - data processing module. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0032] Embodiment 1

[0033] Figure 1 FIG. 1 is an overall flow chart of identification and filtering in a security system according to an embodiment of the present invention. Figure 1 As shown, this embodiment provides a novel identification and filtering method in an intelligent security system, which includes the following steps:

[0034] Step 1: Any edge camera device collects image information of people and vehicles, and uploads the image information and corresponding device information to the application server of the computing center through a high-speed network;

[0035] Step 2: After the application server calls the face recognition program and the license plate recognition program to recognize the image information, it transmits the recognition result with the corresponding device information to the network server (Web server) of the computing center;

[0036] In this embodiment, the identification result with corresponding device information in step 2 includes: an identification code (rcg_id), identification type data (stranger), and a unique device identifier (cameraid) of the corresponding device and other information.

[0037] Step 3: The network server (Web server) filters and removes duplicates from the recognition results before performing business processing on the recognition results, and stores the filtered and duplicated results in the database.

[0038] Figure 2 FIG. 1 is a flow chart of filtering and deduplicating data in one embodiment of the present invention. Figure 2 As shown, in this embodiment, the specific process of filtering and deduplicating the recognition results in step 3 is as follows:

[0039] Step 301: Generate a key value (key value) of the data to be processed from any recognition result of step 2, key=rcg_id+stranger+cameraid, where rcg_id is the identification code, stranger is the identification type data, and cameraid is the unique device identifier of the corresponding device;

[0040] Step 302: Use the key value to search and obtain data from the system's built-in redis (Remote Dictionary Server) cache.

[0041] If no data with the same key value (key value) is found and obtained in the redis cache, proceed to step 303;

[0042] Otherwise, proceed to step 305;

[0043] Step 303: Store the key value generated in step 301 into the redis cache and reset the data validity period, then transmit the data to the business program for processing, store the data in the real-time data table and trigger the plan event;

[0044] Step 304: store the plan event information into the plan information table, and encapsulate the corresponding recognition result data and the plan event information into a json (JavaScript Object Notation, JS object notation, a lightweight data exchange format) data packet and send it to the client through websocket (a protocol for full-duplex communication on a single TCP connection) for page display;

[0045] Step 305: Determine that the identification result data of step 301 is invalid data, refresh the cache data of the corresponding key value (key value) in the redis cache and update the data validity time;

[0046] Among them, in step 305, since the redis cache supports setting the data validity time (setting TTL), that is, the data validity time (TTL) is set while storing data in the redis cache, for example, it can be set to 60 seconds, then if the data validity time (TTL) of this data has not been reset within 60 seconds, then this data will automatically disappear from the redis cache after 60 seconds. Therefore, when the result data is determined to be invalid data, the cached data of the corresponding key value (key value) in the redis cache is refreshed and reaches the preset data validity time, the corresponding data will automatically disappear from the redis cache because the data validity time will not be reset.

[0047] Step 306: Give up reporting the corresponding recognition result data to the business program, and return to step 301 to continue filtering and deduplicating the next recognition result.

[0048] Embodiment 2

[0049] Figure 3 FIG. 1 is a diagram of a security system architecture including a filtering and deduplication module according to an embodiment of the present invention. Figure 3 As shown, the "Golden Eye" city security system implemented in this embodiment is mainly divided into three parts: edge end, cloud end and client end. The system is divided into five levels: perception layer, transmission layer, storage layer, service layer and presentation layer. Among them, the edge end includes camera equipment, sensors and edge processing stations, etc., which are used to collect raw data and upload them to the cloud through the transmission layer; the storage layer in the cloud includes application servers and storage servers. The storage layer calls the application server to process the raw data, including feature extraction, matching and other operations to generate recognition result data, and sends the recognition result data to the service layer. The network server (Web server) of the service layer processes the recognition result data according to different business needs, and transmits the processed data to the storage server for storage; the presentation layer mainly includes clients such as mobile phones and PCs to display the results according to different business needs. Among them, after the application server of the storage layer processes the raw data and before the network server (Web server) of the service layer processes the recognition result data, the recognition result data is cleaned and deduplicated to reduce resource waste and improve data availability.

[0050] Combination Figure 2 and Figure 3 The specific filtering and deduplication process is described as follows:

[0051] The recognition result data processed by the application server of the storage layer includes personnel information or license plate information and corresponding recognition device information, and these recognition result data are sent to the data filtering system of the service layer. The data filtering system has a built-in redis cache. The data filtering system forms a key value through the identification code rcg_id identified in the data, the identification type data stranger and the unique device identification id of the corresponding device, key = 'rcgid' + 'stranger' + 'cameraid';

[0052] Then, use this key value to retrieve data from the redis cache. Taking face recognition as an example, if data is retrieved, it means that this person has appeared within the valid time set under this device. In this case, the data is determined to be invalid data. When the key value of the recognition result data is stored in the redis cache, the data valid time will be set accordingly. For example, if the data valid time is set to 60s, the data in the redis cache will be automatically cleared after 60s. At this time, the system will refresh the cache data corresponding to the key value in the redis cache, update the data valid time, and then execute return (end and continue to process the next recognition result data), and give up reporting the recognition result data to the application layer.

[0053] If no data is obtained, it means that this person has not appeared within the time set under this device. This recognition result data is valid data. The cache of the recognition result data corresponding to the key value in the redis cache is refreshed (that is, the key value of the recognition result data is stored in the redis cache) and the data validity time (the data validity time of the recognition result data is reset). Then, according to the normal process, the data is transmitted to the business program for subsequent processing. At the same time, the data is stored in the real-time data table and the plan event is triggered. Finally, the plan event is stored in the plan information table, and the real-time information, that is, the current recognition result data and the plan information are encapsulated into a json data packet and sent to the client (mobile phone or PC) through websocket for page display.

[0054] Embodiment 3

[0055] Figure 4 FIG. 1 is a structural diagram of a filtering and deduplication system according to an embodiment of the present invention. Figure 4 As shown, this embodiment provides a novel identification and filtering system (40) in an intelligent security system, which is used to implement the above-mentioned identification and filtering method, and includes:

[0056] Redis cache (401), used to cache the key value (key value) and data validity period corresponding to the recognition result data;

[0057] A processing unit (402), connected to the redis cache (401), is used to process the key value (key value) and data validity time of the recognition result data;

[0058] The identification and filtering system (40) is integrated into a network server (Web server) (2) and is respectively connected to the data processing modules (50) of the application server (1) and the network server (2), and is used to filter and remove duplicates from the identification result data processed by the application server and transmit the deduplicated data to the data processing module.

[0059] Compared with the prior art, the present invention solves the problem of excessive and duplicate data in the data collected by the front-end acquisition device by adding an identification and filtering method and system in the new intelligent security system, realizes the filtering and deduplication of duplicate data, reduces the waste of equipment resources, and improves the accuracy of stored data.

[0060] Those skilled in the art can understand that the accompanying drawings are only schematic diagrams of an embodiment, and the modules or processes in the accompanying drawings are not necessarily required to implement the present invention.

[0061] Those skilled in the art can understand that the modules in the device in the embodiment can be distributed in the device in the embodiment according to the description of the embodiment, or can be changed accordingly and located in one or more devices different from the embodiment. The modules in the above embodiment can be combined into one module, or can be further divided into multiple sub-modules.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A recognition and filtering method in an intelligent security system, characterized in that: The following steps are involved: Step 1: Any edge camera device collects image information of people and vehicles, and uploads the image information and device information of the edge camera device to the application server of the computing center through a high-speed network; Step 2: After the application server calls the face recognition program and the license plate recognition program to recognize the image information, the recognition result with the device information is transmitted to the network server of the computing center, wherein the recognition result with the device information includes: the recognition identification code, the recognition type data and the unique device identification of the edge camera device; Step 3: The network server filters and removes duplicates from the recognition results before processing the recognition results, and stores the filtered and deduplicated results in a database, wherein filtering and deduplicating the recognition results includes: Step 301: Generate a key value key=rcg_id+stranger+cameraid of the data to be processed using any recognition result of step 2, wherein rcg_id is the recognition identification code, stranger is the recognition type data, and cameraid is the unique device identifier of the edge camera device.

2. The identification and filtering method according to claim 1, characterized in that: The specific process of filtering and removing duplicates from the recognition results in step 3 also includes: Step 302: Use the key value to search and obtain data from the system's built-in redis cache. If no data with the same key value is found and obtained in the redis cache, proceed to step 303; Otherwise, proceed to step 305; Step 303: store the key value generated in step 301 into the redis cache and reset the data validity period, then transmit the data to the business program for processing, store the data in the real-time data table and trigger the plan event; Step 304: store the plan event information into the plan information table, and encapsulate the corresponding recognition result data and plan event information into a JSON data packet and send it to the client through websocket for page display; Step 305: Determine that the identification result data of step 301 is invalid data, refresh the cache data of the corresponding key value in the redis cache and update the data validity time; Step 306: Give up reporting the corresponding recognition result data to the business program, and return to step 301 to continue filtering and deduplicating the next recognition result.

3. An identification and filtering system in an intelligent security system, which is used to implement the identification and filtering method of any one of claims 1 to 2, characterized in that: include: Redis cache is used to cache the key value and data validity period corresponding to the recognition result data; A processing unit, connected to the redis cache, for processing the key value and data validity time of the recognition result data; Among them, the identification and filtering system is integrated into the network server and connected to the application server and the data processing module of the network server respectively, and is used to filter and de-duplicate the identification result data processed by the application server and transmit the de-duplicated data to the data processing module.

Images