Monitoring an area using multiple networked video cameras

• Main Authors: Ackerman, Nathan, Labrosse, Jean-Paul, Hodge, Andrew
• Format: Patent
• Language: English
• Published: 03.03.2020
• Subjects: ALARM SYSTEMS; CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRIC DIGITAL DATA PROCESSING; ELECTRICITY; HANDLING RECORD CARRIERS; INFORMATION STORAGE; INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORDCARRIER AND TRANSDUCER; ORDER TELEGRAPHS; PHYSICS; PICTORIAL COMMUNICATION, e.g. TELEVISION; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SIGNALLING; SIGNALLING OR CALLING SYSTEMS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

A system and method to connect several cameras located in a given location, such as a parking garage, or area, such as in a neighborhood, is provided. The system allows each camera to function independently from each other with the purpose of recording video within the field of view of the particular camera. Image recognition processes the video to identify objects of interest and generate classification data. A search request can be generated, for example, by touching on a specific object shown on a display of the video, and networked cameras are searched for classification data that matches the search request, e.g., the selected object. Networked cameras can determine the relative spatial views of other cameras. Within a prescribed location or area, a camera or other user can access other networked cameras to extend the range of the search for the requested classification data in a "peer-to-peer" manner. Artificial intelligence software and logic is used to determine which of the surrounding networked cameras is likely to provide requested classification data.