Real-time archiving method and device

Abstract

The invention relates to the technical field of computer networks. The invention relates to an archiving method and device, in particular to a real-time archiving method and device. A device D500 is included, and the device D500 comprises an acquisition module, an analysis and comparison module, an archive feature updating module, an archive merging module and a real-time display and push module.According to the invention, archived archives are compared in real time, so that the real-time performance is improved; the archiving rate can be improved by classifying pictures accumulated in archives and then regularly fusing and updating archive features, archive splitting is reduced by regularly comparing and combining the archives, and meanwhile, the device D500 is invented on the basis of optimized real-time archiving, updating and combining strategies, and the device can be realized through software or hardware. The device is provided with an archiving method involved in the first aspect, and meanwhile an interface for pushing real-time snapshot personnel identities is supplemented. The device can flexibly carry out secondary development for early warning, troubleshooting, attendance checking, entrance guard and other scenes needing identities of landing personnel in real time.

Description

Technical field [0001] The invention relates to the technical field of computer networks, in particular to a method and device for real-time archiving. Background technique [0002] With the increasingly in-depth application of AI biometric recognition technology in the security field in recent years, the application of facial biometric technology has become more and more mature. There are various face gates in key areas of major cities, airport stations, etc., and various face recognition cameras at key entrances and exits. For a city, the face data collected every day is quite huge. [0003] Analyze the characteristics and attributes of each picture for massive historical data (such as gender, age, whether to wear glasses, whether to wear a mask, etc.), and then provide various basic applications such as picture retrieval or attribute retrieval; introduction of existing technology: further , The pictures can be classified by gathering the data for a period of time, and the pict...

AI Technical Summary

Problems solved by technology

Form a space-time archive similar to "one person, one file"; disadvantages of the existing technology: the real-time performance of historical image clustering in the existing technology is not strong, and it usually needs to accumulate data for a period of time before clustering. If there is a lot of data, the clustering waits longer time

Moreover, the analyzed clustering result files are prone to problems such as splitting, miscellaneous pictures, and low actual filing rate; if a real-time filing method similar to one-person-one-picture control is adopted, the file filing rate is usually relatively low and there are many file splits, which are specifically reflected in For all kinds of low light or strong light, the side face and other comparisons cannot be archived correctly

Images