CN-122019830-A - Video resource positioning method of camera gun and related equipment

CN122019830ACN 122019830 ACN122019830 ACN 122019830ACN-122019830-A

Abstract

The application discloses a video resource positioning method and related equipment of a camera, the method comprises the steps of obtaining camera information of each financial website, obtaining structured risk model data aiming at a daily-repeated-disc high-risk model investigation index, determining target financial websites corresponding to each piece of risk model data in the structured risk model data, determining a plurality of current retrieval cameras according to the structured risk model data, the target financial websites corresponding to each piece of risk model data, the camera information of each financial website and the current heat value of each camera, updating the current heat value corresponding to each camera according to a current user operation log, and determining a plurality of updated current retrieval cameras according to the updated current heat value corresponding to each camera. The method can realize accurate pushing of video resources required by site daily re-coiling, improves video verification efficiency, and can be widely applied to the technical field of artificial intelligence.

Inventors

LIU ZONGPEI
FANG XI
LUO ZHENGXIANG
LIAO YUN
YI LAN
Lun Yuehui
TANG WEIQIANG
WU WENSHU
Zhong Yinna
WEN SHENG
YANG CHAO
HE WENJIAN

Assignees

中国邮政集团有限公司广东省分公司

Dates

Publication Date: 20260512
Application Date: 20260116

Claims (10)

1. The method for positioning the video resource of the camera gun is characterized by comprising the following steps of: acquiring original camera gun information of a plurality of financial sites, and performing data cleaning on the original camera gun information of each financial site to obtain target camera gun information of each financial site; acquiring original risk model data aiming at a daily compound disk high risk model investigation index, and carrying out data analysis processing on the original risk model data to obtain structured risk model data; according to a preset institution matching method, performing financial website matching on each piece of risk model data in the structured risk model data, and determining a target financial website corresponding to each piece of risk model data in the structured risk model data; Determining a plurality of current retrieval shooting guns according to the structured risk model data, target financial sites corresponding to each piece of risk model data in the structured risk model data, target shooting gun information of each financial site and a current heat value of each shooting gun; Acquiring a current user operation log, and updating a current heat value corresponding to each camera gun according to the current user operation log; And determining a plurality of updated current retrieval shooting guns according to the updated current heat value corresponding to each shooting gun, wherein each shooting gun shoots corresponding video resources.
2. The method of claim 1, wherein the obtaining the original camera gun information of the plurality of financial sites and performing data cleaning on the original camera gun information of each financial site to obtain the target camera gun information of each financial site comprises: Acquiring the original camera gun information of a plurality of financial websites; According to a preset shooting gun naming rule, carrying out information integration processing on the original shooting gun information of each financial website to obtain an original shooting gun website mapping table; And cleaning the data of the original shooting gun website mapping table to obtain a target shooting gun website mapping table, wherein the target shooting gun website mapping table is used for representing the target shooting gun information corresponding to each financial website, and the target shooting gun website mapping table is a relation mapping table between shooting guns and the financial website.
3. The method of claim 1, wherein the obtaining the original risk model data for the daily complex high risk model investigation index and performing data analysis processing on the original risk model data to obtain the structured risk model data includes: acquiring the original risk model data aiming at the daily compound high risk model investigation index, and inputting the original risk model data into a corresponding risk model; Extracting key fields from the original risk model data by adopting a regular matching technology and a semantic analysis technology to obtain key field data in the original risk model data; and constructing the structured risk model data according to the key field data and the risk dimension of the risk model, and displaying the structured risk model data in a video retrieval interface, wherein the structured risk model data comprises the key field data and a core inspection element.
4. The method according to claim 1, wherein the performing, according to a preset institution matching method, financial website matching on each piece of risk model data in the structured risk model data, and determining a target financial website corresponding to each piece of risk model data in the structured risk model data, includes: performing mechanism information matching on each piece of risk model data in the structured risk model data according to a preset mechanism mapping relation table; If the risk model data with the matching failure exists, carrying out compound matching processing on the risk model data with the matching failure by adopting a multi-mode intelligent matching method to obtain a compound matching result, wherein the compound matching result comprises an accurate matching result and a fuzzy matching result, and the fuzzy matching result comprises a first fuzzy matching result and a second fuzzy matching result; Performing intersection operation on the first fuzzy matching result and the second fuzzy matching result to generate a double fuzzy matching result; Carrying out uniqueness verification on the accurate matching result, and if the accurate matching result passes the uniqueness verification, determining a target financial website corresponding to each piece of risk model data in the structured risk model data according to the accurate matching result; If the accurate matching result does not pass the uniqueness verification, carrying out the uniqueness verification on the double fuzzy matching result; And if the dual fuzzy matching result passes the uniqueness verification, determining a target financial website corresponding to each piece of risk model data in the structured risk model data according to the dual fuzzy matching result.
5. The method of claim 1, wherein determining a number of current retrieval cameras based on the structured risk model data, the target financial website for each piece of risk model data in the structured risk model data, the target camera information for each financial website, and the current heat value for each camera, comprises: According to a preset double screening mechanism, combining the structured risk model data, target financial sites corresponding to each piece of risk model data in the structured risk model data and target camera gun information of each financial site, and determining all cameras in a target site area; Sequencing the heat value of each camera gun in the target network point area to obtain a heat value sequencing result; and comprehensively evaluating according to the heat value sequencing result and the historical judging result of each piece of risk model data in the structured risk model data, and determining a plurality of current retrieval shooting guns.
6. The method of claim 1, wherein the obtaining a current user operation log and updating the current heat value corresponding to each camera gun according to the current user operation log comprise: The current user operation log is obtained, wherein the current user operation log is a log for an inspector to actually retrieve a camera gun; calculating a model weight value corresponding to each camera gun according to the current user operation log; and calculating the updated current heat value of each camera gun by adopting a linear weighting model and combining the user operation log corresponding to each camera gun and the model weight value.
7. A camera gun video resource positioning device, characterized in that the device comprises the following modules: The shooting gun information acquisition module is used for acquiring original shooting gun information of a plurality of financial sites, and carrying out data cleaning on the original shooting gun information of each financial site to acquire target shooting gun information of each financial site; The risk model data analysis module is used for acquiring original risk model data aiming at the daily-repeated-disc high-risk model investigation index, and carrying out data analysis processing on the original risk model data to obtain structured risk model data; The financial website matching module is used for carrying out financial website matching on each piece of risk model data in the structured risk model data according to a preset institution matching method, and determining a target financial website corresponding to each piece of risk model data in the structured risk model data; The current retrieval camera gun determining module is used for determining a plurality of current retrieval camera guns according to the structured risk model data, target financial sites corresponding to each piece of risk model data in the structured risk model data, target camera gun information of each financial site and a current heat value of each camera gun; the camera gun heat value updating module is used for acquiring a current user operation log and updating the current heat value corresponding to each camera gun according to the current user operation log; the current retrieval camera gun updating module is used for determining a plurality of updated current retrieval camera guns according to the updated current heat value corresponding to each camera gun, wherein each camera gun shoots corresponding video resources.
8. An electronic device comprising a memory storing a computer program and a processor implementing the method of any of claims 1 to 6 when the computer program is executed by the processor.
9. A computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 6.
10. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the method of any one of claims 1 to 6.

Description

Video resource positioning method of camera gun and related equipment Technical Field The application relates to the technical field of artificial intelligence, in particular to a video resource positioning method of a camera gun and related equipment. Background In the field of financial security, site daily re-coiling is a core link of risk management and control, and the operation process is generally to check the possible safety risk by retrieving the video of a site shooting gun, performing site management of re-coiling sites, counter business handling, cash operation, receiving and delivering cash and other important businesses. At present, a daily re-coiling in the industry mainly adopts an implementation mode of manual leading, namely, a checker possibly extracts website videos randomly to develop re-coiling, a differential re-coiling flow can be formed based on experience accumulated by individuals, the mode mainly depends on manual experience, obvious limitation exists in practical application, incomplete re-coiling website can be caused, and a few websites really at risk can be missed finally. In addition, part of scenes (such as a mobile phone operated by a customer) can be automatically extracted from video data by using a YOLO video dynamic detection algorithm, and video clips conforming to the scenes are submitted to manual review, however, the video detection scheme mainly has the problems of high requirement on model training professionals and high deployment cost. In summary, the technical problems in the related art are to be improved. Disclosure of Invention Embodiments of the present application aim to solve at least one of the technical problems in the related art to some extent. Therefore, the main purpose of the embodiment of the application is to provide a video resource positioning method of a camera gun and related equipment, which can realize accurate pushing of video resources required by site day rewinding and improve video checking efficiency. In order to achieve the above object, an aspect of the embodiments of the present application provides a method for positioning video resources of a camera, including the following steps: acquiring original camera gun information of a plurality of financial sites, and performing data cleaning on the original camera gun information of each financial site to obtain target camera gun information of each financial site; acquiring original risk model data aiming at a daily compound disk high risk model investigation index, and carrying out data analysis processing on the original risk model data to obtain structured risk model data; according to a preset institution matching method, performing financial website matching on each piece of risk model data in the structured risk model data, and determining a target financial website corresponding to each piece of risk model data in the structured risk model data; Determining a plurality of current retrieval shooting guns according to the structured risk model data, target financial sites corresponding to each piece of risk model data in the structured risk model data, target shooting gun information of each financial site and a current heat value of each shooting gun; Acquiring a current user operation log, and updating a current heat value corresponding to each camera gun according to the current user operation log; And determining a plurality of updated current retrieval shooting guns according to the updated current heat value corresponding to each shooting gun, wherein each shooting gun shoots corresponding video resources. In some embodiments, the obtaining the original camera gun information of a plurality of financial sites, and performing data cleaning on the original camera gun information of each financial site to obtain target camera gun information of each financial site includes: Acquiring the original camera gun information of a plurality of financial websites; According to a preset shooting gun naming rule, carrying out information integration processing on the original shooting gun information of each financial website to obtain an original shooting gun website mapping table; And cleaning the data of the original shooting gun website mapping table to obtain a target shooting gun website mapping table, wherein the target shooting gun website mapping table is used for representing the target shooting gun information corresponding to each financial website, and the target shooting gun website mapping table is a relation mapping table between shooting guns and the financial website. In some embodiments, the obtaining the original risk model data of the daily complex high risk model investigation index, and performing data analysis processing on the original risk model data, to obtain structured risk model data, includes: acquiring the original risk model data aiming at the daily compound high risk model investigation index, and inputting the original risk model data into a correspondi