IMAGE PROCESSING DEVICE, IMAGE PROCESSING METHOD

• US 20130250107A1
• Filed: 11/05/2012
• Published: 09/26/2013
• Est. Priority Date: 03/26/2012
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Mizutani, Masami, Okada, Yasutaka
• Granted Patent: US 9,390,336 B2
• Time in Patent Office: Days
• US Class Current: 348/148
• CPC Class Codes: G06V 10/255 Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
• 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
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles

Abstract

An image processing device includes a processor; and a memory which stores a plurality of instructions that cause the processor to execute, capturing an image including vehicle side information, dividing the image into a plurality of areas, and defining a first divided area image including a center of a side end of the image and a second divided area image not including the center of the side end of the image; smoothing the first divided area image using a first filter, or smoothing the second divided area image using a second filter having a greater coefficient than that of the first filter; extracting a plurality of feature points from the first divided area image or the second divided area image; calculating an optical flow from the plurality of feature points of the first divided area image or the second divided area image.

Claims

1. An image processing device comprising:

a processor; and

a memory which stores a plurality of instructions that cause the processor to execute,

capturing an image including vehicle side information;

dividing the image into a plurality of areas, and defining a first divided area image including a center of a side end of the image and a second divided area image not including the center of the side end of the image;

smoothing the first divided area image using a first filter, or smoothing the second divided area image using a second filter having a greater coefficient than that of the first filter;

extracting a plurality of feature points from the first divided area image or the second divided area image;

calculating an optical flow from the plurality of feature points of the first divided area image or the second divided area image having different acquisition time; and

determining an approaching object approaching the vehicle on the basis of the optical flow.

2. The image processing device according to claim 1,

wherein the defining defines the first divided area image such that a height thereof decreases from the side end toward a center of the image, and defines the second divided area image such that an area thereof increases from the side end toward the center of the image.

3. The image processing device according to claim 1,

wherein the extracting extracts feature points from the first divided area image before extracting feature points from the second divided area image.

4. The image processing device according to claim 1, further comprising:

a storage unit configured to store positional information of the approaching object determined by the determining,

wherein the smoothing performs smoothing using the second filter on a part of the first divided area image having been subjected to the smoothing using the first filter on the basis of the positional information.

5. The image processing device according to claim 1,

wherein the extracting extracts the feature points by scanning the first divided area image in an upward direction or in a downward direction of the image starting from the center of the side end of the image.

6. The image processing device according to claim 1,

wherein the extracting extracts the feature points by scanning the second divided area image in an upward direction or in a downward direction of the image so as to break apart the image from the center of the side end of the image.

7. The image processing device according to claim 5,

wherein the extracting defines an approaching direction in which the approaching object approaches the vehicle on the basis of a defined travelling direction of the vehicle, and gives priority of the scanning in the upward direction or in the downward direction on the basis of the approaching direction.

8. The image processing device according to claim 1,

wherein the plurality of instructions cause the processor to further execute,

detecting a speed of the vehicle;

and

stopping the capturing the image on the basis of the speed of the vehicle.

9. The device according to claim 1,

wherein the plurality of instructions cause the processor to further execute,

detecting a steering angle of the vehicle;

and

stopping the capturing the image on the basis of the steering angle.

10. The device according to claim 1,

wherein the image including the vehicle side information is captured by using a fisheye lens.

11. An image processing method comprising:

capturing an image including vehicle side information;

dividing the image into a plurality of areas, and defining a first divided area image including a center of a side end of the image and a second divided area image not including the center of the side end of the image;

smoothing the first divided area image using a first filter, or smoothing the second divided area image using a second filter having a greater coefficient than that of the first filter;

extracting a plurality of feature points from the first divided area image or the second divided area image;

calculating, by a computer processor, an optical flow from the plurality of feature points of the first divided area image or the second divided area image having different acquisition time; and

determining an approaching object approaching the vehicle on the basis of the optical flow.

12. The method according to claim 11,

wherein the defining defines the first divided area image such that a height thereof decreases from the side end toward a center of the image, and defines the second divided area image such that an area thereof increases from the side end toward the center of the image.

13. The method according to claim 11,

wherein the extracting extracts feature points from the first divided area image before extracting feature points from the second divided area image.

14. The method according to claim 11, further comprising:

Obtaining positional information from a storage unit configured to store the positional information of the approaching object determined by the determining,

wherein the smoothing performs smoothing using the second filter on a part of the first divided area image having been subjected to the smoothing using the first filter on the basis of the positional information.

15. The method according to claim 11,

wherein the extracting extracts the feature points by scanning the first divided area image in an upward direction or in a downward direction of the image starting from the center of the side end of the image.

16. The method according to claim 11,

wherein the extracting extracts the feature points by scanning the second divided area image in an upward direction or in a downward direction of the image so as to break apart the image from the center of the side end of the image.

17. The method according to claim 15,

wherein the extracting defines an approaching direction in which the approaching object approaches the vehicle on the basis of a defined travelling direction of the vehicle, and gives priority of the scanning in the upward direction or in the downward direction on the basis of the approaching direction.

18. A computer-readable storage medium storing an image processing program that causes a computer to execute a process comprising:

capturing an image including vehicle side information;

dividing the image into a plurality of areas, and defining a first divided area image including a center of a side end of the image and a second divided area image not including the center of the side end of the image;

smoothing the first divided area image using a first filter, or smoothing the second divided area image using a second filter having a greater coefficient than that of the first filter;

extracting a plurality of feature points from the first divided area image or the second divided area image;

calculating an optical flow from the plurality of feature points of the first divided area image or the second divided area image having different acquisition time; and

determining an approaching object approaching the vehicle on the basis of the optical flow.