Image Processing for Vehicular Applications Applying Image Comparisons

• US 20080051957A1
• Filed: 10/11/2007
• Published: 02/28/2008
• Est. Priority Date: 09/03/2002
• Status: Active Grant

• Current Assignee: American Vehicular Sciences LLC (Acacia Research Corporation)
• Original Assignee: Automotive Technologies International Incorporated
• Inventors: Breed, David, Chen, Tie-Qi
• Granted Patent: US 7,676,062 B2
• Time in Patent Office: Days
• US Class Current: 701/36
• CPC Class Codes: B60R 21/0152 using strain gauges
• CPC Class Codes: B60R 21/0153 using field detection presence sensors
• CPC Class Codes: B60R 21/01552 detecting position of specific human body parts, e.g. face, eyes or hands
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths
• CPC Class Codes: G06V 20/593 Recognising seat occupancy

Abstract

Vehicle including a compartment receivable of an object and a system for tracking the object includes at least one imaging device each arranged to receive an image of a portion of the compartment containing the object and a control unit coupled to each imaging device and which controls the imaging device to obtain a first set of images without the object and at least one second image including the object. The control unit analyzes the second image(s) in consideration of the first set of images to derive information about the object. This information may be the type, size and/or position of the object or a part thereof. The information may be used to control vehicular components which have a variable use based on the type, size or position of the object or part thereof.

Claims

1. A vehicle including a compartment receivable of an object of interest and a system for tracking the object of interest in the compartment, comprising:

at least one imaging device each arranged to receive an image of a portion of the compartment containing the object of interest; and

a control unit coupled to said at least one imaging device and arranged to control said at least one imaging device to obtain a first set of images without the object of interest and a second image including the object of interest, said control unit being arranged to analyze the second image in consideration of the first set of images to derive information about the object of interest,

whereby the derived information can be used to control a vehicular component.

2. The vehicle of claim 1, wherein said at least one imaging device comprises a plurality of imaging devices arranged at different locations to obtain different images of a common portion of the compartment.

3. The vehicle of claim 1, wherein said control unit analyzes the second image in consideration of the first set of images to derive information about the object of interest by subtracting the second image from each image in the first set of images to determine which provides a best match and then deriving the information about the object of interest based on a differential image between the second image and the determine one of the first images which provides the best match after the subtraction process.

4. The vehicle of claim 1, wherein said control unit analyzes the second image in consideration of the first set of images to derive information about the object of interest by comparing the second image to each image in the first set of images to determine the presence of static and dynamic reflections and deriving information about the object of interest by analyzing the dynamic reflections.

5. The vehicle of claim 1, wherein said control unit analyzes the second image in consideration of the first set of images to derive information about the object of interest by inputting a matrix derived from the second image into a trained pattern recognition algorithm which is trained on images of different objects in different positions in the compartment.

6. The vehicle of claim 5, wherein the trained pattern recognition algorithm is a neural network.

7. The vehicle of claim 1, wherein said control unit is arranged to derive information about occupancy or a seating state of a seat in the compartment.

8. The vehicle of claim 1, wherein when the object of interest is a human, said control unit is arranged to determine information about a location of eyes, ears, a chest or a head of the human.

9. The vehicle of claim 1, wherein when the object of interest is a human, said control unit is arranged to determine information about a location of a chest or a head of the human and track motion of the head or chest through analysis of images subsequently obtained by said at least one imaging device.

10. The vehicle of claim 9, wherein said control unit is arranged to analyze images to track motion of the head or chest using differential motion or a correlation technique.

11. The vehicle of claim 1, wherein said control unit is arranged to apply a feature extraction technique to images obtained by said at least one imaging device.

12. The vehicle of claim 11, wherein said feature extraction technique is arranged to divide an entire image into small rectangular blocks, calculate feature values from each block, line up the calculated feature values from individual blocks, and then normalize the feature values.

13. The vehicle of claim 11, wherein said feature extraction technique is arranged to analyze pixels of the images by obtaining a sum of edge strengths defined as the largest differences between neighboring pixels and an orientation of an edge determined by the neighboring pixel that produces the largest difference between itself and the pixel of interest.

14. A method for tracking a human in a compartment in a vehicle, comprising:

arranging at least one imaging device in a position in which it receives an image of a portion of the compartment containing the human;

controlling the at least one imaging device to obtain a first set of images without the human and a second image including the human;

analyzing the second image in consideration of the first set of images to determine information about a location of a chest or a head of the human;

once the location of the chest or head is determined, tracking motion of the head or chest through analysis of images subsequently obtained by the at least one imaging device; and

optionally controlling a component in the vehicle based on the tracking of the motion of the head or chest.

15. The method of claim 14, wherein the tracking of the motion of the head or chest comprises using differential motion or a correlation technique.

16. The method of claim 14, wherein the at least one imaging device comprises a plurality of imaging devices, further comprising arranging the imaging devices at different locations to obtain different images of a common portion of the compartment.

17. The method of claim 14, wherein the analysis of the second image in consideration of the first set of images comprises applying a trained pattern recognition algorithm to an array derived from the second image.

18. The method of claim 14, wherein the analysis of the second image in consideration of the first set of images comprises extracting features from an array derived from the second image.

19. The method of claim 18, wherein the step of extracting features comprises dividing the array into small rectangular blocks, calculating feature values from each block, lining up the calculated feature values from individual blocks, and then normalizing the feature values.

20. The method of claim 18, wherein the step of extracting features comprises analyze pixels of the array by obtaining a sum of edge strengths defined as the largest differences between neighboring pixels and an orientation of an edge determined by the neighboring pixel that produces the largest difference between itself and the pixel of interest.