Method and device for determining a snapshot image of a target vehicle, storage medium and program product
By performing lightweight detection and multi-dimensional scoring on the video stream of the target vehicle and dynamically adjusting the weights, the problem of poor vehicle capture quality on low-performance devices is solved, achieving efficient and accurate vehicle capture.
CN122116299APending Publication Date: 2026-05-29ZHEJIANG DAHUA TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ZHEJIANG DAHUA TECH CO LTD
- Filing Date
- 2026-01-30
- Publication Date
- 2026-05-29
AI Technical Summary
Technical Problem
Existing license plate recognition systems rely on high-performance computing devices, while low-performance embedded devices struggle to achieve optimal vehicle capture, resulting in poor image quality.
Method used
By performing lightweight target detection on the video stream of the target vehicle, vehicle features are extracted, and a multi-dimensional scoring strategy is adopted to dynamically adjust the weights and select high-quality captured images.
Benefits of technology
Achieve efficient and accurate vehicle capture on low-performance devices, reduce memory consumption, adapt to ultra-low-end hardware, and improve vehicle detection and capture efficiency.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122116299A_ABST
Abstract
The application discloses a method and device for determining a snapshot image of a target vehicle, a storage medium and a program product. The method comprises the following steps: performing lightweight target detection on a plurality of images contained in a video stream of a target vehicle, extracting vehicle features of the target vehicle in each image of the plurality of images, and obtaining a plurality of vehicle features corresponding to the plurality of images; performing multi-dimensional scoring on the plurality of images according to the plurality of vehicle features, and obtaining a plurality of scoring results, wherein the plurality of images correspond to the plurality of scoring results in a one-to-one manner; and determining a snapshot image of the target vehicle from the plurality of images according to the plurality of scoring results.
Need to check novelty before this filing date? Find Prior Art