Browse Prior Art Database

A system for tracking object with distributed surveillance nodes

IP.com Disclosure Number: IPCOM000248945D
Publication Date: 2017-Jan-24
Document File: 4 page(s) / 185K

Publishing Venue

The IP.com Prior Art Database

Abstract

The present disclosure provides intelligent webcam tracking object function by leveraging embedded cognitive features on each intelligent webcam, which can recognize missing people or artifacts instantly in the system based on the input of picture or video clip.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

1

A system for tracking object with distributed surveillance nodes

It's pretty hard to track a people or artifact with current Closed Circuit Television (CCTV) or public surveillance system, because existing systems are designed for monitoring by people, which caused poor efficiency as no easy way to figure out missing people or artifact from tons of video clips of public surveillance system or CCTV by people. And these systems are designed for historic data analysis, they can't identify and track people or artifact instantly and interactively.

Our idea is to leverage embedded cognitive features on each intelligent webcam, the webcam could recognize missing people or artifacts instantly in the system based on the input of picture or video clip. 1) Very efficient: People/system doesn't need to go through all of the video clips that might have missing objects, instead, the new system can recognize missing object and analyze missing object's feature based on the distribution of intelligent monitors, even track if missing object could attached to any transportation tool. 2) Instant tracking: intelligent monitor could collaborate based on their cognitive feature and analyzed result, to track artifacts instantly and interactively with each other, and even with other mobilized devices (i.e. mobile phone).

This innovative method mainly includes 2 parts: 1) Intelligent cameras with embedded cognitive features: to recognize artifacts based on missing object's image and properties input (i.e a missing blue car and license number is '京XYZ0123'), camera calls different recognition Application Program Interfaces (APIs) to analyze and find out the missing objects, and then send most likely video clips and analyzed results (text) back to the monitor of cloud based platform and also send to the other cameras that may also have tracking info of the missing object (base...