IP Library Granted Patent US 7,433,532
Granted Patent B2
US 7,433,532 · App. 10/428,320 · Granted Oct 7, 2008

Max entropy optimized retinal camera

Assignee: Kestrel Corporation
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,433,532
App. No.
10/428,320
Granted
Oct 7, 2008
Kind
B2
Abstract

An enhanced entropy camera provides enhanced image quality by utilizing a custom analog-to-digital conversion function. The analog-to-digital conversion occurs according to a custom non-linear quantization function that provides enhanced entropy in the digitized image, yet avoids the harsh artifacts that tend to arise in images quantized according to a maximum entropy function. A plurality of custom non-linear functions are made available for selection depending upon imaging conditions, such as light intensity, pigmentation of the retina, or the part of the retina being imaged. Each of the plural non-linear quantization function comprises a distinct non-linear look up table for each of plural bands (e.g., red, green, and blue color channels).

Claims (25)

1. A method of producing a digital image signal having enhanced entropy, the method comprising:

selecting one of a plurality of non-linear quantization functions based on a light characteristic of a subject to be imaged;

capturing an image signal of the subject;

quantizing the captured image signal utilizing the selected non-linear quantization function, to produce a digital image signal containing information about the subject.

2. The method of claim 1 , wherein the light characteristic of the subject is color.

3. The method of claim 1 , wherein the light characteristic of the subject is iniensity.

4. The method of claim 1 , wherein the non-linear quantization function comprises a non-linear look up table.

5. The method of claim 1 , wherein the selecting is performed automatically.

6. The method of claim 1 , wherein the selecting is performed based on an input received from an operator.

7. The method of claim 1 , wherein cach of the plurality of non-linear quantization functions comprises a distinct non-linear look up table for each of plural wavelength bands.

8. The method of claim 1 , wherein there are three wavelength bands, the wavelength bands corresponding to red, green, and blue color channels.

9. A method of producing a digital image signal having enhanced entropy, the method comprising:

capturing an image signal of a subject;

quantizing a first band of the captured image signal utilizing a first non-linear quantiation function;

quantizing a second band of the captured image signal utilizing a second non-linear quantization function;

combining the first and second quantized bands to produce a digital image signal containing informalion about the subject; and “wherein the first and second bands each correspond to a distinct colir channel”.

10. The method of claim 9 , wherein the first non-linear quantiation function is distinct from the second non-linear quantization function.

11. The method of claim 9 , wherein the first non-lincar quantiation function is substantially the same as the second non-linear quantization function.

12. The method of claim 9 , wherein at least one of the first and second bands each correspond to a band of light that is not visible.

13. The method of claim 9 , wherein the first and second bands cach correspond to a distinct spectral band other than red, green, or blue.

14. A method of producing a digital image signal having enhanced entropy, the method comprising:

capturing an image signal of a subject;

quantizing the captured image signal at P-bit depth utilizing a subsantially un-enhanced quantiation function, to produce an intermediate digital image signal containing information about the subject;

re-quantizing the intermediate digital image signal at Q-bit depth utilizing a non-linear quantization function, to produce a digital image signal containing enhanced information about the subjeet, wherein Q is smaller than P.

15. The method of producing a digital image signal of claim 14 , wherein P=16 and Q=8.

Assignments (2)
CORRECTIVE ASSIGNMENT TO CORRECT THE SECOND INVENTOR'S NAME. IT IS LISTED AS "HOLIZ, PETER" AND SHOULD BE CORRECTED TO READ "SOLIZ, PETER" PREVIOUSLY RECORDED ON REEL 015889 FRAME 0702 ASSIGNOR HEREBY CONFIRMS THE INVENTOR'S CORRECT NAME IS PETER SOLIZ. Recorded Jul 1, 2005
From: TRUITT, PAUL WILEY; SOLIZ, PETER
To: KESTREL CORPORATION
Reel/Frame 016463/0295 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 8, 2005
From: WILEY, PAUL; HOLIZ, PETER
To: KESTREL CORPORATION
Reel/Frame 015689/0702 →
Continuity (2)
Provisional Application 6037706900 · May 1, 2002
Related Publication 20030228117A1 · Dec 11, 2003