IP Library Granted Patent US 12677072
Granted Patent B2
US 12677072 · App. 18/661,423 · Granted Jul 7, 2026

Method and systems for adaptive pixel correction

Inventor: Shiou-jyh Ja (Portland, OR)
Assignee: ARACELI BIOSCIENCES INC.
H04N23/76H04N23/52H04N23/72H04N23/73H04N25/47H04N25/50
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 12677072
App. No.
18/661,423
Granted
Jul 7, 2026
Kind
B2
Abstract

Methods and systems are provided for correcting pixel intensity variation in an image. In one example, a method comprises: acquiring reference images at different sensor temperatures and/or exposure times; generating a plurality of parameters via identifying a cut-off pixel number using a break-up point between a linear region and a non-linear region of an intensity curve for each reference image; sending coordinates of identified pixels to be corrected to an image sensor; capturing a pixel-corrected image via the image sensor by replacing a pixel intensity at coordinates of the pixels to be corrected with a local statistic intensity; and outputting the pixel-corrected image for display and/or storage.

Claims (20)

1 . A method for pixel correction of an imaging system, comprising:

obtaining a reference image with the imaging system configured according to imaging conditions;

removing pixels of the reference image having a brightness equal to a maximum intensity from a dataset of the reference image;

plotting an intensity distribution curve of the dataset of the reference image;

modeling the intensity distribution curve using a curve fitting algorithm;

identifying a break-up point between a linear region and a non-linear region of the intensity distribution curve using a decision tree algorithm;

obtaining a set of model parameters for the intensity distribution curve in the linear region based on results of the decision tree algorithm;

generating a parameter model using the set of model parameters; and

storing the parameter model in at least one memory of a computing device in the imaging system.

2 . The method of claim 1 , further comprising obtaining multiple reference images with the imaging system configured according to different imaging conditions.

3 . The method of claim 2 , wherein the different imaging conditions include different sensor temperature and fixed exposure time of the imaging system, fixed sensor temperature and different exposure time of the imaging system, and/or different sensor temperature and different exposure time of the imaging system.

4 . The method of claim 1 , wherein plotting the intensity distribution curve comprises numbering each pixel, storing coordinates of a pixel with a corresponding pixel number, and sorting numbered pixels of the reference image by intensity.

5 . The method of claim 1 , wherein modeling the intensity distribution curve using the curve fitting algorithm comprises identifying a transition between the linear region and the non-linear region of the intensity distribution curve using decision tree algorithm.

6 . The method of claim 1 , wherein obtaining the set of model parameters comprises obtaining a slope and a y-intercept of the linear region.

7 . The method of claim 6 , wherein generating the parameter model using the model parameters comprises modeling the slope and the y-intercept of linear regions of intensity curves captured for different imaging conditions derived by a single, three dimensional (3D) model.

8 . The method of claim 6 , further comprising:

identifying sensor operating conditions of an image sensor of the imaging system;

identifying coordinates of pixels to be corrected using the sensor operating conditions and the set of model parameters; and

automatically correcting pixels captured in a raw image by the imaging system via the image sensor by replacing a pixel intensity at coordinates of the pixels to be corrected with a local statistic intensity.

9 . The method of claim 8 , further comprising outputting a pixel-corrected image for display and/or storage.