Image Processor

• US 20080069399A1
• Filed: 12/22/2005
• Published: 03/20/2008
• Est. Priority Date: 12/24/2004
• Status: Abandoned Application

• Current Assignee: National University Corporation Yokohama National University, Fuji Jukogyo KK (Subaru Corp.)
• Original Assignee: National University Corporation Yokohama National University, Fuji Jukogyo KK (Subaru Corp.)
• Inventors: Torii, Tsuyoshi, Yokoi, Wataru, Nagao, Tomoharu, Kise, Katsuyuki, Ogawa, Genya
• Application Number:
  US11/658,135
• Publication Number:
  US 20080069399A1
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30261 Obstacle
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/215 Motion-based segmentation

Abstract

An image processor (1) for subjecting an image picked up by an imaging apparatus (21) to image processing to extract a specific object from the image, the image processor comprising: an image processing section (3) which subjects a plurality kinds of images (t, t−1, . . . , t−k) picked up by the imaging apparatus to the image processing based on a processing program comprising image filters (F) combined in a form of a tree structure, and which forms an output image (O) in which the specific object is extracted.

Claims

1. An image processor for subjecting an image picked up by an imaging apparatus to image processing to extract a specific object from the image, the image processor comprising:

an image processing section which subjects a plurality kinds of images picked up by the imaging apparatus to the image processing based on a processing program comprising image filters combined in a form of a tree structure, and which forms an output image in which the specific object is extracted,

wherein the plurality kinds of images constituting a plurality of dynamic images picked up by the imaging apparatus at time intervals.

2. The image processor of claim 1, further comprising a processing program forming section for forming the processing program, wherein the processing program forming section forms the processing program by genetic programming using the plurality kinds of images, a target image and a weight image.

3. The image processor of claim 2, wherein a ratio of a weight of an extraction region of the weight image and a weight of a non-extraction region of the weight image is set to be a ratio of reciprocal of an area ratio of the extraction region and the non-extraction region.

4. The image processor of claim 2, wherein the processing program forming section forms the processing program using a plurality of learning sets of the plurality kinds of images, the target image and the weight image.

5. The image processor claims 2, wherein fitness used for genetic programming at the processing program forming section is calculated such that it becomes smaller as the number of nodes in the processing program is greater.

6. The image processor of claim 5, wherein a rate of the number of nodes to the fitness is varied in accordance with the number of generations in a process of evolution in the genetic programming.

7. The image processor claims 2, wherein fitness used for genetic programming at the processing program forming section is calculated such that it becomes greater as the number of nodes of two-input image filters in the processing program is greater.

8. The image processor of claim 7, wherein a rate of the number of nodes of the two-input image filters to the fitness is varied in accordance with the number of generations in the process of evolution in the genetic programming.

9. The image processor of claims 1, wherein the processing program comprises a combination of a plurality of processing programs.

10. The image processor of claim 9, wherein the output image is formed by coupling results of processing of the plurality of processing programs nonlinearly.

11. The image processor of claims 1, wherein a mask filter is included in the image filters.

12. The image processor of claims 1, further comprising a display section for displaying an image, wherein an output image formed based on the processing program is superposed on the input image displayed on the display section and displayed.

13. The image processor of claim 1, wherein the image processing section subjects, to the image processing, a plurality of images constituting a dynamic image picked up by the imaging apparatus and an optical flow image produced from these images based on the processing program comprising image filters combined in a form of a tree structure.

14. The image processor of claim 13, further comprising a processing program forming section for forming the processing program, wherein the processing program forming section outputs a processing program optimized by genetic programming using the plurality of images, an optical flow image, a target image and a weight image.

15. The image processor of claim 13, wherein the image processing section converts the plurality of images picked up by the imaging apparatus into images viewed from above in a pseudo manner.

16. The image processor of claim 15, wherein the image processing section inputs, to the processing program, the plurality of converted images and an optical flow image produced based on the plurality of converted images.

17. The image processor of claim 15, wherein the processing program forming section carries out learning by genetic programming using the plurality of converted images, an optical flow image, a target image and a weight image produced based on the plurality of converted images, and outputs an optimized processing program.

18. The image processor of claim 13, wherein the optical flow image is an image on which information of magnitude of calculated flow is expressed as a gradation value.

19. The image processor of claim 13, wherein the optical flow image is an image on which information of a direction of calculated flow is expressed as a gradation value.

20. The image processor of claim 18, wherein the flow in the optical flow image is a flow with respect to a moving plane of the imaging apparatus calculated based on moving state of the imaging apparatus.

21. The image processor of claim 13, wherein in the optical flow image, a gradation value of a picture element portion where reliability of calculated flow is low is set to 0.

22. The image processor of claim 13, wherein the image processing section converts the plurality of images picked up by the imaging apparatus into a state in which a vantage point is moved upward.