IP Library Granted Patent US 11,816,811
Granted Patent B2
US 11,816,811 · App. 17/740,121 · Granted Nov 14, 2023

Efficient image demosaicing and local contrast enhancement

Inventor: Wenyi Zhao (Weston, FL)
Assignee: Intuitive Surgical Operations, Inc.
G06T3/4015H04N23/843H04N25/134A61B2576/00H04N23/555
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 11,816,811
App. No.
17/740,121
Granted
Nov 14, 2023
Kind
B2
Abstract

An exemplary method includes a demosaic module receiving a frame of raw pixels having values in a first color space, generating vertical and horizontal color difference signals using information from the frame of raw pixels, generating reconstructed color pixel values for raw pixels that are missing captured color pixel values of a first color component of the first color space, transforming the captured and reconstructed color pixel values of the first color component to brightness pixel values in a second color space, and transforming the vertical and horizontal color difference signals into chrominance pixel values in the second color space.

Claims (62)

1. A method comprising:

receiving, by a demosaic module, a frame of raw pixels captured by an image capture sensor in an image capture unit having a color filter array, each location in the frame of raw pixels including one captured color pixel value and missing two color pixel values of a first color space;

generating, by the demosaic module for each location in the frame of raw pixels, a vertical color difference signal and a horizontal color difference signal using information from the frame of raw pixels;

generating, by the demosaic module for each location in the frame of raw pixels that is missing a captured color pixel value of a first color component of the first color space, a reconstructed color pixel value of the first color component such that a captured color pixel value or a reconstructed color pixel value of the first color component is known for each location in the frame of raw pixels;

transforming, by the demosaic module for each location in the frame of raw pixels, the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to a brightness pixel value in a second color space that is different from the first color space; and

transforming, by the demosaic module for each location in the frame of raw pixels and without using the brightness pixel value in the second color space, the vertical color difference signal and the horizontal color difference signal of the respective location into chrominance pixel values in the second color space.

2. The method of claim 1 , wherein transforming the vertical color difference signal and the horizontal colors difference signal of the respective location into chrominance pixel values in the second color space comprises:

determining an average of the vertical color difference signal and the horizontal colors difference signal; and

using the average of the vertical color difference signal and the horizontal colors difference signal to determine the chrominance pixel values in the second color space.

3. The method of claim 1 , wherein for a particular location:

the captured color pixel value is a captured green pixel value;

the horizontal color difference signal is the captured green pixel value minus a horizontal estimated blue pixel value; and

the vertical color difference signal is the captured green pixel value minus a vertical estimated red pixel value.

4. The method of claim 1 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a second color component of the first color space.

5. The method of claim 4 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a third color component of the first color space.

6. The method of claim 1 , further comprising:

generating, by the demosaic module for each location in the frame of raw pixels, a vertical estimated color pixel value and a horizontal estimated color pixel value using information from the frame of raw pixels, wherein for a location in the frame of raw pixels having a captured color pixel value of the second color component of the first color space, the vertical estimated color pixel value is of the first color component of the first color space, and the horizontal estimated color pixel value is of the first color component of the first color space;

wherein generating the vertical color difference signal and the horizontal color difference signal for each location in the frame of pixels uses the vertical estimated color pixel value and the horizontal estimated color pixel value, wherein for the location in the frame of raw pixels having the captured color pixel value of the second color component of the first color space, the vertical color difference signal is generated based on the vertical estimated color pixel value of the first color component of the first color space, and the horizontal color difference signal is generated based on the horizontal estimated color pixel value of the first color component of the first color space.

7. A system comprising:

a processor; and

a memory communicatively coupled to the processor and storing instructions executable by the processor to:

receive a frame of raw pixels captured by an image capture sensor in an image capture unit having a color filter array, each location in the frame of raw pixels including one captured color pixel value and missing two color pixel values of a first color space;

generate, for each location in the frame of raw pixels, a vertical color difference signal and a horizontal color difference signal using information from the frame of raw pixels;

generate, for each location in the frame of raw pixels that is missing a captured color pixel value of a first color component of the first color space, a reconstructed color pixel value of the first color component such that a captured color pixel value or a reconstructed color pixel value of the first color component is known for each location in the frame of raw pixels;

transform, for each location in the frame of raw pixels, the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to a brightness pixel value in a second color space that is different from the first color space; and

transform, for each location in the frame of raw pixels and without using the brightness pixel value in the second color space, the vertical color difference signal and the horizontal color difference signal of the respective location into chrominance pixel values in the second color space.

8. The system of claim 7 , wherein transforming the vertical color difference signal and the horizontal colors difference signal of the respective location into chrominance pixel values in the second color space comprises:

determining an average of the vertical color difference signal and the horizontal colors difference signal; and

using the average of the vertical color difference signal and the horizontal colors difference signal to determine the chrominance pixel values in the second color space.

9. The system of claim 7 , wherein for a particular location:

the captured color pixel value is a captured green pixel value;

the horizontal color difference signal is the captured green pixel value minus a horizontal estimated blue pixel value; and

the vertical color difference signal is the captured green pixel value minus a vertical estimated red pixel value.

10. The system of claim 7 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a second color component of the first color space.

11. The system of claim 10 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a third color component of the first color space.

12. The system of claim 7 , the instructions further executable by the processor to:

generate, for each location in the frame of raw pixels, a vertical estimated color pixel value and a horizontal estimated color pixel value using information from the frame of raw pixels, wherein for a location in the frame of raw pixels having a captured color pixel value of the second color component of the first color space, the vertical estimated color pixel value is of the first color component of the first color space, and the horizontal estimated color pixel value is of the first color component of the first color space;

wherein generating the vertical color difference signal and the horizontal color difference signal for each location in the frame of pixels uses the vertical estimated color pixel value and the horizontal estimated color pixel value, wherein for the location in the frame of raw pixels having the captured color pixel value of the second color component of the first color space, the vertical color difference signal is generated based on the vertical estimated color pixel value of the first color component of the first color space, and the horizontal color difference signal is generated based on the horizontal estimated color pixel value of the first color component of the first color space.

13. A non-transitory computer-readable medium storing instructions executable by a processor to:

receive a frame of raw pixels captured by an image capture sensor in an image capture unit having a color filter array, each location in the frame of raw pixels including one captured color pixel value and missing two color pixel values of a first color space;

generate, for each location in the frame of raw pixels, a vertical color difference signal and a horizontal color difference signal using information from the frame of raw pixels;

generate, for each location in the frame of raw pixels that is missing a captured color pixel value of a first color component of the first color space, a reconstructed color pixel value of the first color component such that a captured color pixel value or a reconstructed color pixel value of the first color component is known for each location in the frame of raw pixels;

transform, for each location in the frame of raw pixels, the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to a brightness pixel value in a second color space that is different from the first color space; and

transform, for each location in the frame of raw pixels and without using the brightness pixel value in the second color space, the vertical color difference signal and the horizontal color difference signal of the respective location into chrominance pixel values in the second color space.

14. The non-transitory computer-readable medium of claim 13 , wherein transforming the vertical color difference signal and the horizontal colors difference signal of the respective location into chrominance pixel values in the second color space comprises:

determining an average of the vertical color difference signal and the horizontal colors difference signal; and

using the average of the vertical color difference signal and the horizontal colors difference signal to determine the chrominance pixel values in the second color space.

15. The non-transitory computer-readable medium of claim 13 , wherein for a particular location:

the captured color pixel value is a captured green pixel value;

the horizontal color difference signal is the captured green pixel value minus a horizontal estimated blue pixel value; and

the vertical color difference signal is the captured green pixel value minus a vertical estimated red pixel value.

16. The non-transitory computer-readable medium of claim 13 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a second color component of the first color space.

17. The non-transitory computer-readable medium of claim 16 , wherein transforming the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to the brightness pixel in the second color space for each location in the frame of raw pixels is performed without using a color pixel value of a third color component of the first color space.

18. A method comprising:

receiving, by a demosaic module, a frame of raw pixels captured by an image capture sensor in an image capture unit having a color filter array, each location in the frame of raw pixels including one captured color pixel value and missing two color pixel values of a first color space;

generating, by the demosaic module for each location in the frame of raw pixels, a vertical color difference signal and a horizontal color difference signal using information from the frame of raw pixels;

generating, by the demosaic module for each location in the frame of raw pixels that is missing a captured color pixel value of a first color component of the first color space, a reconstructed color pixel value of the first color component such that a captured color pixel value or a reconstructed color pixel value of the first color component is known for each location in the frame of raw pixels;

transforming, by the demosaic module for each location in the frame of raw pixels, the captured color pixel value or the reconstructed color pixel value of the first color component of the respective location to a brightness pixel value in a second color space that is different from the first color space; and

transforming, by the demosaic module for each location in the frame of raw pixels, the vertical color difference signal and the horizontal color difference signal of the respective location into chrominance pixel values in the second color space;

wherein transforming the vertical color difference signal and the horizontal colors difference signal of the respective location into chrominance pixel values in the second color space comprises:

determining an average of the vertical color difference signal and the horizontal colors difference signal; and

using the average of the vertical color difference signal and the horizontal colors difference signal to determine the chrominance pixel values in the second color space.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 9, 2022
From: ZHAO, WENYI
To: INTUITIVE SURGICAL OPERATIONS, INC.
Reel/Frame 059875/0242 →
Continuity (5)
Continuation 16917483 · Jun 30, 2020
Continuation 16239489 · Jan 3, 2019
Continuation 14451621 · Aug 5, 2014
Provisional Application 61864090 · Aug 9, 2013
Related Publication 20220261956A1 · Aug 18, 2022