Automatic multiscale image acquisition from a steerable camera

• US 7,796,154 B2
• Filed: 03/07/2005
• Issued: 09/14/2010
• Est. Priority Date: 03/07/2005
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Tian, Ying-Li, Hampapur, Arun, Brown, Lisa Marie, Senior, Andrew William, Pankanti, Sharathchandra
• Primary Examiner(s): Dastouri; Mehrdad
• Assistant Examiner(s): HALLENBECK-HUBER, JEREMIAH CHARLES
• Application Number:
  US11/074,383
• Publication Number:
  US 20060197839A1
• Time in Patent Office: 2,017 Days
• Field of Search: 348/154
• US Class Current: 348/154
• CPC Class Codes: H04N 23/62 Control of parameters via user interfaces
• CPC Class Codes: H04N 23/632 for displaying or modifying preview images prior to image capturing, e.g. variety of image resolutions or capturing parameters
• CPC Class Codes: H04N 23/662 by using master/slave camera arrangements for affecting the control of camera image capture, e.g. placing the camera in a desirable condition to capture a desired image
• CPC Class Codes: H04N 23/695 Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
• CPC Class Codes: H04N 23/90 Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast

Abstract

A system for automatically acquiring high-resolution images by steering a pan-tilt-zoom camera at targets detected in a fixed camera view is provided. The system uses automatic or manual calibration between multiple cameras. Using automatic calibration, the homography between the cameras in a home position is estimated together with the effects of pan and tilt controls and the expected height of a person in the image. These calibrations are chained together to steer a slave camera. The manual calibration scheme steers a camera to the desired region of interest and calculates the pan, tile and zoom parameters accordingly.

Claims

1. A method for acquiring multi-scale images, the method comprising:

monitoring a field of view using a master camera;

identifying a plurality of predefined regions of interest within the field of view;

associating a slave camera with each one of the regions of interest, each slave camera comprising a moveable and controllable camera;

for each region of interest;

directing the master camera and the slave camera associated with that region of interest to a common position within that region of interest;

generating tracking data on a common object located at the common position for both the master camera and the slave camera;

using the tracking data from the master camera and the slave camera to generate a homography between the master camera and the slave camera;

determining a height proportionality between the master camera and the slave camera for the common object;

moving each slave camera in accordance with at least one known pattern;

tracking points along the known pattern while the slave camera is moving;

estimating the motion of the slave camera using the tracked points; and

comparing the estimated motion of the tracked points to actual point locations from the known pattern and deriving the control parameters for each slave camera from the comparison of estimated points to actual points;

using the master camera to identify an object of interest moving through the field of view;

predicting future movement of the object of interest;

identifying the predefined regions of interest that are overlapped by the predicted future movement;

identifying slave cameras associated with the identified overlapped predefined regions of interest, and setting pan, tilt and zoom parameters for each identified slave camera so that a field of view associated with each identified slave camera is equal in size to its associated overlapped predefined region of interest; and

using the homography, height proportionality and control parameters derived from the comparison of estimated points to actual points for each identified slave camera to track future movement of the object of interest through the identified overlapped predefined regions of interest.