Method and Device for Video Surveillance

Simple SummaryContent extracted from patent full text and abstract with AI.

This invention introduces a method and device for video surveillance that compares each pixel of a camera image to both short-term and long-term background models. By analyzing the differences between these models, the system can accurately detect static objects (such as abandoned bags) and recognize when objects are added or removed from the scene, without needing prior knowledge of an empty background.

Use CasesContent extracted from patent full text and abstract with AI.

• Security monitoring in airports, train stations, and public places to detect unattended or removed items.
• Surveillance in industrial or commercial facilities to identify unauthorized placement or removal of equipment.
• Automatic alarm generation in critical infrastructure (e.g., power plants) when suspicious objects appear or disappear.
• Smart city or traffic monitoring to track static obstructions or hazards on roads.
• Retail security to detect shoplifting (e.g., removal of large items) or misplaced products.

BenefitsContent extracted from patent full text and abstract with AI.

• Does not require an initial 'empty scene' for background modeling, making it robust for busy or crowded environments.
• Reduces false alarms by more reliably distinguishing between transient and persistent changes in the scene.
• Accurately detects both added and removed static objects, improving overall security.
• Can operate continuously in highly dynamic and cluttered locations, such as airports or stations.
• Automates the process of identifying suspicious items, enabling quick response and reducing manual monitoring needs.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

The invention relates to a method and a device for video surveillance, wherein, by means of at least one video camera, an image of an image excerpt of an environment to be monitored in the vicinity of the video camera is recorded, wherein at least one pixel of the image is compared with a corresponding pixel of a short-term background model assigned to the image excerpt and with a corresponding pixel of a long-term background model assigned to the image excerpt.