Method for detecting target objects in a surveillance region

Patent metadata

Field	Value
Publication number	US-9607225-B2
Application number	US-201214000214-A
Country	US
Kind code	B2
Filing date	Feb 17, 2012
Priority date	Feb 18, 2011
Publication date	Mar 28, 2017
Grant date	Mar 28, 2017

Abstract

The present application presents methods and apparatuses for detecting target objects in an image sequence of a monitoring region. In some examples, such methods may include adjusting pixel values of images of the image sequence for interference components associated with at least one interfering object, generating the interference components associated with the at least one interfering object that is situated in the monitoring region, searching the image sequence for the target objects based on the adjusted pixel values, detecting a start of a predetermined sequence of motions associated with the interfering object, and computing an instantaneous position of the at least one interfering object during the predetermined sequence of motions, wherein adjusting the pixel values of the images is based upon the instantaneous position.

First claim

The invention claimed is:

1. A method for detecting target objects in an image sequence of a monitoring region, comprising:
adjusting pixel values of images of the image sequence for interference components associated with at least one interfering object;
generating the interference components associated with the at least one interfering object that is situated in the monitoring region;
searching the image sequence for the target objects based on the adjusted pixel values;
detecting a start of a predetermined sequence of motions associated with the interfering object; and
computing an instantaneous position of the at least one interfering object during the predetermined sequence of motions, wherein adjusting the pixel values of the images is based upon the instantaneous position.

2. The method of claim 1 , wherein the instantaneous position of the at least one interfering object is computed based upon a movement model, wherein the movement model includes at least one model parameter.

3. The method of claim 2 , wherein at least one of the at least one model parameters is determined during an initialization phase as an initial value.

4. The method of claim 2 , wherein at least one of the at least one model parameters is determined during a first training phase.

5. The method of claim 2 , wherein at least one of the at least one model parameters is continually optimized during a second training phase.

6. The method of claim 2 , wherein at least one of the at least one model parameters is one or more of a resting position of a sub-region of the at least one interfering object within an image, an opening region within an image belonging to one or more of the at least one interfering object, a time period between the start of the predetermined sequence of motions and a trigger signal that triggers the sequence of motions, and a velocity of the at least one interfering object during the predetermined sequence of motions of the at least one interfering object.

7. The method of claim 2 , wherein at least one of the at least one model parameters is one or more of a line index of a starting line or of an ending line of the interfering object in the height or occupancy map of the image, a gap index of a center gap between two parts of the interfering object in the height or occupancy map of the image, an average time difference between a trigger signal for the predetermined sequence of motions of the interfering object and a first detection of the predetermined sequence of motions in an occupancy map of the image, and a movement velocity of the interfering object.

8. The method of claim 1 , wherein the at least one interfering object comprises a door and the predetermined sequence of motions comprises an opening or closing process of the door.

9. The method of claim 1 , wherein the pixel values include height values and/or occupancy values of picture elements of the images.

10. A system for detecting target objects in an image sequence of a monitoring region, comprising:
a camera for recording the image sequence; and
an evaluation unit connected to the camera for data transmission and configured to:
adjust pixel values of images of the image sequence for interference components, the interference components originating from at least one interfering object situated in a control region;
search the image sequence for the target objects based on the adjusted pixel values; and
detect a start of a predetermined sequence of motions of the interfering object to determine an instantaneous position of the interfering object within the images during the sequence of motions, wherein the pixel values are adjusted based on the instantaneous position of the interfering object.

11. The system of claim 10 , wherein the evaluation unit is further configured to:
evaluate a trigger signal received from a motion sensor or an operating element to detect the start of the predetermined sequence of motions; and
detect the start of the predetermined sequence of motions by evaluating images from the image sequence of the camera.

12. The system of claim 10 , wherein the evaluation unit is further configured to compute the instantaneous position of the predetermined sequence of motions based on a movement model.

13. The system as of claim 10 , wherein the camera is a stereo camera for generating two images each comprising individual images.

14. The system of claim 10 , wherein the interfering object is a door and the sequence of motions is an opening motion of the door.

15. A vehicle for transporting persons, comprising:
at least one vehicle door;
an operating unit configured to generate a trigger signal for automatically opening the vehicle door; and
a system for detecting target objects in an image sequence of a monitoring region, the system comprising:
a camera for recording the image sequence; and
an evaluation unit connected to the camera for data transmission and configured to:
adjust pixel values of images of the image sequence for interference components, the interference components originating from at least one interfering object situated in a control region;
search the image sequence for the target objects based on the adjusted pixel values; and
detect a start of a predetermined sequence of motions of the interfering object to determine an instantaneous position of the interfering object within the images during the sequence of motions, wherein the pixel values are adjusted based on the instantaneous position of the interfering object, wherein the interfering object is the at least one vehicle door.

Assignees

• Ick Julia
• Hella Kgaa Hueck & Co

Inventors

• Ick Julia

Classifications

• G06V20/59

  inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions · CPC title

• G06V20/52 Primary

  Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69) · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06F18/24

  Classification techniques · CPC title

• G06T2207/30248

  Vehicle exterior or interior · CPC title