Image processing system and driving support system

• US 7,065,230 B2
• Filed: 05/24/2002
• Issued: 06/20/2006
• Est. Priority Date: 05/25/2001
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Yuasa, Mayumi, Yamaguchi, Osamu, Fukui, Kazuhiro
• Primary Examiner(s): Wu, Jingge
• Assistant Examiner(s): Tabatabai, Abolfazl
• Application Number:
  US10/153,806
• Publication Number:
  US 20030039378A1
• Time in Patent Office: 1,488 Days
• Field of Search: 382/103, 382/104, 382/115, 382/118, 382/218, 351/200, 351/206, 351/209, 351/210, 351/221, 351/246, 701/23, 340/5.53, 340/5.58, 340/5.83
• US Class Current: 382/104
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads

Abstract

An image input unit inputs an image including a user'"'"'s face area to specify the user'"'"'s gaze or face direction. A pattern generation unit generates an input pattern that can be compared with each dictionary pattern from the image. A dictionary unit previously stores a plurality of dictionary patterns each of which differently corresponds to the user'"'"'s gaze or face direction. A pattern comparison unit respectively calculates a similarity degree between the input pattern and each of the dictionary patterns by comparing the input pattern with each of the dictionary patterns. A gaze detection unit detects the user'"'"'s gaze or face direction based on the similarity degree.

Claims

1. Driving support apparatus for a driver of a vehicle, comprising;

an image input unit configured to input an image including the driver'"'"'s face area to specify the driver'"'"'s gaze or face direction;

a pattern generation unit configured to generate an input pattern from the image that can be compared with each of a plurality of dictionary patterns each of which differently corresponds to the driver'"'"'s gaze or face direction;

a dictionary unit configured to store the plurality of dictionary patterns;

a pattern comparison unit configured to respectively calculate a similarity degree between the input pattern and each of the dictionary patterns by comparing the input pattern with each of the dictionary patterns;

a gaze detection unit configured to detect the driver'"'"'s gaze or face direction based on the similarity degree;

an obstacle detection area selection unit configured to select an area outside of the driver'"'"'s view based on the driver'"'"'s gaze or face direction, the area representing an obstacle detection area for the vehicle; and

an obstacle detection unit configured to detect an obstacle existing in the obstacle detection area.

2. The driving support apparatus according to claim 1,

wherein said image input unit previously inputs a plurality of images of the driver'"'"'s face area while the driver gazes at each of a plurality of targets being differently located in the driver'"'"'s view space, and

wherein said dictionary unit generates the dictionary pattern using the plurality of images for each of the targets.

3. The driving support apparatus according to claim 2,

wherein the dictionary pattern of each of the targets is a subspace generated from the plurality of images for each of the targets, and

wherein said dictionary unit generates a difference subspace between two classes, the two classes being two of the targets that are closely located.

4. The driving support apparatus according to claim 3,

wherein said dictionary unit generates a common subspace from each difference subspace between each combination of two classes of the targets.

5. The driving support apparatus according to claim 4,

wherein said image input unit previously inputs the plurality of images of the driver'"'"'s face area whenever the driver gazes at the same target at different times, the plurality of images being input for each of the targets, and

wherein said dictionary unit generates a second difference subspace between two different times for the same target.

6. The driving support apparatus according to claim 5,

wherein said dictionary unit generates a second common subspace from each of the second difference subspaces between two different times for all of the targets, the second common subspace being a self-variation subspace for the driver.

7. The driving support apparatus according to claim 6,

wherein said dictionary unit generates a constraint subspace by subtracting the self-variation subspace from the difference subspace between the targets.

8. The driving support apparatus according to claim 7,

wherein said pattern comparison unit respectively projects a subspace of the input pattern and a subspace of the dictionary pattern onto the constraint subspace, and calculates an angle between the projected subspace of the input pattern and the projected subspace of the dictionary pattern, the angle representing the similarity degree between the input pattern and the dictionary pattern.

9. The driving support apparatus according to claim 8,

wherein said pattern comparison unit calculates the similarity degree between the projected subspace of the input pattern and the projected subspace of the dictionary pattern by using a subspace method or a mutual subspace method.

10. Driving support method for a driver of a vehicle, comprising:

inputting an image including the driver'"'"'s face area to specify the driver'"'"'s gaze or face direction;

generating an input pattern from the image that can be compared with each of a plurality of dictionary patterns each of which differently corresponds to the driver'"'"'s gaze or face direction;

storing the plurality of dictionary patterns;

respectively calculating a similarity degree between the input pattern and each of the dictionary patterns by comparing the input pattern with each of the dictionary patterns;

detecting the driver'"'"'s gaze or face direction based on the similarity degree;

selecting an area outside of the driver'"'"'s view based on the driver'"'"'s gaze or face direction, the area representing an obstacle detection area for the vehicle; and

detecting an obstacle existing in the obstacle detection area.

11. A computer program product, comprising:

a computer readable program code embodied in said product for causing a computer to support driving of a vehicle, said computer readable program code having;

a first program code to input an image including the driver'"'"'s face area to specify the driver'"'"'s gaze or face direction;

a second program code to generate an input pattern from the image that can be compared with each of a plurality of dictionary patterns each of which differently corresponds to the driver'"'"'s gaze or face direction;

a third program code to store the plurality of dictionary patterns;

a fourth program code to respectively calculate a similarity degree between the input pattern and each of the dictionary patterns by comparing the input pattern with each of the dictionary patterns;

a fifth program code to detect the driver'"'"'s gaze or face direction based on the similarity degree;

a sixth program code to select an area outside of the driver'"'"'s view based on the driver'"'"'s gaze or face direction, the area representing an obstacle detection area for the vehicle; and

a seventh program code to detect an obstacle existing in the obstacle detection area.