System or method for enhancing an image

• US 7,492,962 B2
• Filed: 08/25/2005
• Issued: 02/17/2009
• Est. Priority Date: 08/25/2005
• Status: Active Grant

• Current Assignee: Aptiv Technologies AG (Aptiv PLC)
• Original Assignee: Delphi Technologies, Inc. (Aptiv PLC)
• Inventors: Bauson, William A., Zhang, Yan, Perisho, Robert A. Jr.
• Primary Examiner(s): Kassa; Yosef
• Application Number:
  US11/212,052
• Publication Number:
  US 20070047833A1
• Time in Patent Office: 1,272 Days
• Field of Search: 382/274, 382/275, 382/284, 382/294, 358/3.26, 358/3.27
• US Class Current: 382/274
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 5/50 using two or more images, e.g. averaging or subtraction
• CPC Class Codes: G06T 5/92 based on global image properties

Abstract

A system and method for enhancing the contrast within an image. An enhanced image can be generated in a real-time or substantially real-time manner from an initial image. A low-contrast image is enhanced, including the visibility of text within the low-contrast image. A logarithm transform method is used to enhance the overall contrast and visibility within the image. A power transform method is used to enhance the visibility and contrast of written text, symbols and other forms of signs within the enhanced image.

Claims

1. An image enhancement system that is configured to create an enhanced image from an initial image, comprising:

a power transform subsystem, said power transform subsystem providing for a power transform method for creating a power image from the initial image;

a logarithm transform subsystem, said logarithm transform subsystem providing for a logarithm transform method for creating a log image from the initial image; and

a fusion subsystem, said fusion subsystem providing for a fusion method for creating said enhanced image, wherein said enhanced image is influenced by said log image and said power image.

2. The system of claim 1, further comprising a linear scaling method, wherein said linear scaling method influences said log image and said power image.

3. The system of claim 1, further comprising a thresholding method, wherein said power image is influenced by said thresholding method.

4. The system of claim 1, further comprising a mask image refinement method, wherein said power image is influenced by said mask image refinement method.

5. The system of claim 1, wherein the initial image includes a sign image, wherein the power transform subsystem enhanced said sign image within said enhanced image.

6. The system of claim 1, wherein a plurality of initial images are captured by a CMOS camera at a rate of between about 10 frames per second and 60 frames per second.

7. The system of claim 1, wherein the system is a night vision application for use in a vehicle.

8. The system of claim 1, further comprising a first Laplacian pyramid representation built from the log image and a second Laplacian pyramid representation built from the power image, wherein the fusion subsystem, further comprises a pyramid fusion of the first and second Laplacian pyramid transformations.

9. A system for creating an enhanced image from an initial image, comprising:

a computer, said computer providing for;

identifying a plurality of pixel values within the initial image;

performing a logarithm transform method and a power transform method on the initial image to create a plurality of interim images;

invoking a linear transform method on at least two of said interim images; and

fusing at least two of said interim images into the enhanced image.

10. The system of claim 9, wherein the linear transform method includes a plurality of potential pixel values from 0 through 255.

11. The system of claim 9, wherein said interim images are grayscale or color images.

12. The system of claim 9, wherein at least one interim image is influenced by said power transform method and is subsequently subjected to a threshold method before being fused into the enhanced image.

13. The system of claim 9, wherein at least one interim image is influenced by said power transform method and is subsequently subjected to a mask image refinement method before being fused into the enhanced image.

14. A method for enhancing an image, comprising:

identifying a plurality of pixel values in an initial image;

creating a first copy and a second copy of the plurality of pixel values;

performing a logarithm transform method on the first copy of pixel values to create a log image;

performing a power transform method on the second copy of pixel values to create a power image;

combine the power image and log image into the enhanced image.

15. The method of claim 14, wherein a fusion method selectively combines the power image and the log image into the enhanced image.

16. The method of claim 14, wherein a threshold method selectively influences the power image before it is combined into the enhanced image.

17. The method of claim 14, wherein a mask image refinement method selectively influences the power image before it is combined into the enhanced image.

18. The method of claim 14, wherein the logarithm transform method is influenced by a predefined plurality of numerical parameters.

19. The method of claim 14, wherein the power transform method is influenced by a predefined plurality of numerical parameters.

20. The method of claim 14, wherein the initial image is a view from a CMOS camera loaded on a vehicle.

21. The method of claim 14, wherein the enhanced image is created from the initial image in a substantially real-time manner.

22. An image enhancement system that is configured to create an enhanced image from an initial image, comprising:

a power transform subsystem, said power transform subsystem providing for a power transform method for creating a power image from the initial image;

a logarithm transform subsystem, said logarithm transform subsystem providing for a logarithm transform method for creating a log image from the initial image;

a first Laplacian pyramid representation built from the log image and a second Laplacian pyramid representation built from the power image; and

a pyramid fusion subsystem that fuses said first and second Laplacian pyramid transformations into an enhanced, pyramid fused image.