Audio and video multi-mode identification method and system
Through the multimodal recognition method and system of audio and video, the targets in the law enforcement scenario are locked and tracked in real time, and the staff are assisted through frame-level annotation and direction indication, which solves the problem of poor auxiliary effect of recorders in target tracking in the prior art, and improves the accuracy and efficiency of tracking.
CN119723428AActive Publication Date: 2025-03-28ZHIYE ELECTRONICS
5 Cites 0 Cited by
Patent Information
- Application Number
- CN202510230366.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-28
- Publication Date
- 2025-03-28
- Estimated Expiration
- 2045-02-28
AI Technical Summary
Technical Problem
When the prior art is used for target tracking in law enforcement scenarios, the recorder cannot effectively assist staff, resulting in easy loss of tracking targets in complex environments.
Method used
A multimodal recognition method and system for audio and video is adopted to collect and process video data in real time through the collaborative work of the perception layer, identification layer and indication layer, lock and track targets, and assist staff through frame-level annotation and direction indication.
Benefits of technology
It improves the accuracy and efficiency of target tracking in law enforcement scenarios, reduces the risk of target loss, and enhances the functionality and purpose of the recorder in actual use.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention relates to the technical field of data processing, in particular to an audio and video multi-mode identification method and system, and the system comprises a sensing layer, an identification layer, and an indication layer. The collection device operates collected video data and uploads the collected video data to the perception layer in real time, the perception layer synchronously perceives the posture of the collection device, locks a tracking target based on the collected video and the posture of the collection device, and synchronously extracts feature parameters of the tracking target, and the identification layer further receives the feature parameters of the tracking target extracted from the perception layer. According to the invention, the collection device is used as a main body to process the video data collected by the collection device, the tracking target is obtained from the video data, the specific tracking target is finally determined by using the specified repeated locking logic, and further, the frame-level image processing and analysis technology is used to determine the tracking target. According to the invention, the tracking target is continuously tracked, so that a system end user instruction is provided, and an auxiliary tracking effect is brought to the system end user in a tracking scene of the tracking target.
Need to check novelty before this filing date? Find Prior Art