Image Processing for Vehicular Applications Applying Edge Detection Technique

• US 20070282506A1
• Filed: 08/20/2007
• Published: 12/06/2007
• 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: Chen, Tie-Qi, Breed, David
• Granted Patent: US 7,769,513 B2
• Time in Patent Office: Days
• US Class Current: 701/45
• CPC Class Codes: B60N 2/028 Elastically mounted back-rests
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths
• CPC Class Codes: G06V 20/593 Recognising seat occupancy

Abstract

Arrangement and method for obtaining information about objects in an environment in or around a vehicle includes one or more optical imagers for obtaining images of the environment and a processor coupled to the imager(s) for obtaining information about an object in one or more images obtained by the imager(s). The processor is arranged to process the obtained images to determine edges of objects in the images and input data about the edges into a trained pattern recognition algorithm which has been trained to provide information about the object as output. The pattern recognition algorithm may include a neural network or variation thereof. The information about the object may be used to control a vehicular component such as an airbag or light filter.

Claims

1. An arrangement for obtaining information about objects in an environment in or around a vehicle, comprising:

at least one optical imager for obtaining images of the environment; and

a processor coupled to said at least one imager for obtaining information about an object in one or more images obtained by said at least one imager, said processor being arranged to process the obtained images to determine edges of objects in the images and input data about the edges into a trained pattern recognition algorithm which has been trained to provide information about the object as output.

2. The arrangement of claim 1, wherein said pattern recognition algorithm is a neural network.

3. The arrangement of claim 1, wherein said at least one imager is arranged to obtain images of an interior compartment in the vehicle.

4. The arrangement of claim 1, wherein said at least one imager is arranged to obtain images of an environment exterior to the vehicle.

5. The arrangement of claim 1, wherein said processor includes an edge detection algorithm arranged to receive the obtained images and output a vector containing the location of edges of the object.

6. The arrangement of claim 1, wherein said processor includes an edge detection algorithm arranged to receive the obtained images and output a vector containing the location of edges of the object and their orientation.

7. The arrangement of claim 1, wherein each of said at least one imager is a CCD array or a CMOS array.

8. The arrangement of claim 1, wherein said pattern recognition algorithm is trained to provide as output, a classification, identification and/or location of an object.

9. A vehicle, comprising:

the arrangement of claim 1;

and

at least one component having a variable condition, state or position depending on the presence of objects in the environment in or around the vehicle, said processor being arranged to control said at least one component based on the information about the object obtained as output from said pattern recognition algorithm.

10. The vehicle of claim 9, wherein said at least one component comprises an airbag arranged to deploy prior to or during a crash involving the vehicle.

11. A method for obtaining information about objects in an environment in or around a vehicle, comprising:

in a training stage, training a pattern recognition algorithm to provide information about objects in the environment upon receiving as input information about edges of unknown objects;

installing the pattern recognition algorithm in a processor on the vehicle;

operatively obtaining images of the environment;

deriving data about edges of objects in the obtained images; and

providing the data to the pattern recognition algorithm in the processor to receive as output, information about the object.

12. The method of claim 11, wherein the pattern recognition algorithm is a neural network.

13. The method of claim 11, further comprising arranging at least one imager on the vehicle to obtain images of an interior compartment in the vehicle.

14. The method of claim 11, further comprising arranging at least one imager on the vehicle to obtain images of an environment exterior to the vehicle.

15. The method of claim 11, wherein the step of deriving data about edges of objects in the obtained images comprising inputting the obtained images into an edge detection algorithm which outputs a vector containing the location of edges of the object, the vector being the data provided to the pattern recognition algorithm.

16. The method of claim 11, wherein the images are obtained by means of at least one CCD array or at least one CMOS array.

17. The method of claim 11, wherein the pattern recognition algorithm is trained to provide as output, a classification, identification and/or location of an object.

18. A method for controlling at least one component having a variable condition, state or position depending on the presence of objects in the environment in or around the vehicle, the method comprising:

obtaining information in accordance with the method of claim 11;

and

controlling the at least one component based on the information about the object obtained as output from the pattern recognition algorithm.

19. The method of claim 18, wherein the at least one component comprises an airbag arranged to deploy prior to or during a crash involving the vehicle.

20. The method of claim 11, wherein the step of training the pattern recognition algorithm comprises:

arranging a known object in a field of view of at least one optical images, obtaining images of the object, applying an edge detection algorithm to the images to obtain data about the edges of the object in the images and associating the data with information about the object;

then

changing the position of the object, obtaining additional images, applying an edge detection algorithm to the additional images to obtain data about the edges of the object in the additional images and associating the data with the information about the object;

then

changing the object, obtaining images of the new object, applying an edge detection algorithm to the new images to obtain data about the edges of the new object in the images and associating the data with information about the new object;

then

changing the position of the new object, obtaining additional images, applying an edge detection algorithm to the images to obtain data about the edges of the new object in the additional images and associating the data with the information about the new object;

then

repeating the process of changing the object or its position until sets of data about objects in images is obtained; and

inputting the data to a pattern recognition algorithm training program to obtain a pattern recognition algorithm which will receive data about edges of unknown objects in images and provide as output, information about the objects.