Sensing apparatus and method for vehicles

• US 8,044,998 B2
• Filed: 04/12/2007
• Issued: 10/25/2011
• Est. Priority Date: 10/12/2004
• Status: Expired due to Fees

• Current Assignee: TRW Limited (Zeppelin - Stiftung der Stadt Friedrichshafen)
• Original Assignee: TRW Limited (Zeppelin - Stiftung der Stadt Friedrichshafen)
• Inventors: Heenan, Adam John
• Primary Examiner(s): Maung; Zarni
• Application Number:
  US11/786,771
• Publication Number:
  US 20080036855A1
• Time in Patent Office: 1,657 Days
• Field of Search: 348/61, 348113-118
• US Class Current: 348/61
• CPC Class Codes: G06T 5/80 Geometric correction
• CPC Class Codes: G06V 10/24 Aligning, centring, orientation detection or correction of the image
• CPC Class Codes: G06V 10/247 by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle

Abstract

A sensing apparatus for a host vehicle or the like, the apparatus comprises sensing means arranged to capture a first dataset of digital data comprising a set of data points representative of an image of a scene around the host vehicle, said image including at least one salient feature;

• first data processing means arranged to identify a subset of the data points which correspond to the at least one salient feature, and second data processing means arranged to apply a transformation to the subset of data points to produce a processed subset which corresponds to a digital image of the salient features in the scene that has been corrected for at least one type of distortion in the image.

Claims

1. A sensing apparatus for a host vehicle comprising:

sensing means arranged to capture a first dataset of digital data comprising a set of data points representative of an image of a scene around the host vehicle, said image including at least one salient feature;

first data processing means arranged to identify a subset of said data points which correspond only to said at least one salient feature; and

second data processing means arranged to apply a transformation to said subset of data points to produce a processed subset which corresponds to a digital image of said salient features in said scene that has been corrected for at least one type of distortion in said image.

2. Apparatus according to claim 1 wherein the apparatus is included in a lane detection apparatus.

3. Apparatus according to claim 1 which is adapted to produce a subset of data points which only correspond to salient features and nothing else.

4. Apparatus according to claim 1 which further includes a third data processing means which is adapted to further process said processed subset of data points to provide real world image points by mapping said further processed points onto a three dimensional map of said scene around said vehicle.

5. Apparatus according to claim 1 in which said second data processing means is arranged to apply at least one transform which corrects said identified image points for barrel distortion.

6. A method of determining the position of salient features within a scene in an image around a host vehicle comprising:

(a) capturing a first dataset of digital data comprising a set of data points representative of an image of a scene ahead of said host vehicle, said image including at least one salient feature;

(b) identifying a subset of said data points which correspond only to said at least one salient feature; and

(c) applying a transformation to said subset of data points to produce a processed subset which corresponds to a digital image of said at least one salient feature in said scene that has been corrected for at least one type of distortion in said image.

7. A non-transitory computer readable medium comprising a set of program instructions which, when running on a processor of a vehicle-mounted image processing apparatus that is arranged to receive signals from a host vehicle mounted image capture device, cause said apparatus to:

(a) capture a first dataset of digital data comprising a set of data points representative of an image of a scene around said host vehicle as captured by said image capture device, said image including at least one salient feature;

(b) identify a subset of said data points which correspond to only said at least one salient feature; and

(c) apply a transformation to said subset of data points to produce a processed subset of data points which corresponds to a digital image of said at least one salient features in said scene that has been corrected for at least one type of distortion in said image.