Method of digitally processing an image and method of digitally processing a plurality of pixels

Patent metadata

Field	Value
Publication number	US-11729516-B2
Application number	US-202217667646-A
Country	US
Kind code	B2
Filing date	Feb 9, 2022
Priority date	Feb 10, 2021
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

A method of digitally processing an image comprises: generating an intensity distribution model ( 170 ) in respect of at least a portion of the array of sensing pixels ( 102 ) of a detector device ( 100 ). The array of sensing pixels comprises clusters of pixels. A pixel ( 140 ) from the array of sensing pixels is then selected ( 202 ) and a first distance and a second distance from the selected pixel to a first neighbouring pixel ( 142 ) and a second neighbouring pixel ( 144 ), respectively, are determined ( 402 ) and the intensity distribution model ( 170 ) referenced ( 406 ) by the first distance is used to calculate a first weight and a second weight to apply to the first and second neighbouring pixels, respectively. The first distance comprises an intra-cluster distance and the second distance comprises an inter-cluster distance, the intra-cluster distance being different from the inter-cluster distance. The first weight is applied ( 214 ) to the first neighbouring pixel ( 142 ) and the second weight is applied ( 214 ) to the second neighbouring pixel ( 144 ).

First claim

What is claimed is:

1. A method of digitally processing an image captured using an array of sensing pixels of a detector device, the method comprising:
generating an intensity distribution model in respect of at least a portion of the array of sensing pixels;
selecting a pixel from the array of sensing pixels, the array of sensing pixels comprising clusters of pixels;
determining a first distance from the selected pixel to a first neighboring pixel and using the intensity distribution model referenced by the first distance to calculate a first weight to apply to the first neighboring pixel;
determining a second distance from the selected pixel to a second neighboring pixel and using the intensity distribution model referenced by the second distance to calculate a second weight to apply to the second neighboring pixel; wherein the first distance comprises an intra-cluster distance and the second distance comprises an inter-cluster distance, the intra-cluster distance being different from the inter-cluster distance;
applying the first weight to the first neighboring pixel; and
applying the second weight to the second neighboring pixel.

2. The method according to claim 1 , wherein
the intensity distribution model models variation of intensity with angle of incidence with respect to a notional pixel; and the method further comprises:
translating the first distance to a first corresponding angle of incidence; and
determining a first intensity from the intensity distribution model using the first corresponding angle of incidence.

3. The method according to claim 2 , further comprising:
translating the second distance to a second corresponding angle of incidence; and
determining a second intensity from the intensity distribution model using the second corresponding angle of incidence.

4. The method according to claim 1 , wherein the intensity distribution model models variation of intensity with an angle of incidence with respect to a notional pixel.

5. The method according to claim 4 , wherein the angle of incidence is linearly related to a distance from a peak of the intensity distribution model.

6. The method according to claim 4 , wherein a distance between the notional pixel and a neighboring pixel is:
c×p o +d×p i ,

where p o is an inter-cluster distance and p i is an intra-cluster distance, and c and d are constants.

7. The method according to claim 1 , wherein the intensity distribution model models variation of intensity with distance.

8. The method according to claim 1 , wherein the intensity distribution model in respect of the at least a portion of the array of sensing pixels is in respect of a notional pixel and a predetermined number of pixels peripheral to the notional pixel.

9. The method according to claim 1 , wherein the intensity distribution model comprises normalized intensity values.

10. The method according to claim 1 , further comprising:
generating a kernel using the first weight and the second weight.

11. The method according to claim 10 , further comprising:
deconvolving a pixel of the array of pixels with the kernel.

12. The method according to claim 10 , further comprising:
generating another kernel using different weights; and
deconvolving another pixel of the array of pixels using the another kernel.

13. The method according to claim 1 , further comprising:
generating a set of kernels, each kernel of the set of kernels respectively corresponding to a different pixel index of a predetermined cluster pattern of pixels in the array of sensing pixels, the predetermined cluster pattern repeating throughout the array of sensing pixels;
separately applying each kernel of the set of kernels as a deconvolution filter to the captured image to yield a set of deconvolved images, respectively corresponding to each pixel index of the predetermined cluster pattern;
selecting a plurality of processed pixels respectively from each deconvolved image of the set of deconvolved images, each plurality of the plurality of processed pixels selected respectively corresponding to the each pixel index of the predetermined cluster pattern; and
constructing a processed image having a plurality of indices using the respectively selected plurality of processed pixels.

14. The method according to claim 13 , wherein each processed pixel of the pluralities of processed pixels comprises a respective array index with respect to the array of pixels.

15. The method according to claim 14 , further comprising:
constructing the processed image by arranging each processed pixel of the pluralities of processed pixels according to the respective array index.

Assignees

• Melexis Technologies Nv

Inventors

• Reusen Wouter
• Hill Adrian
• Buydens Luc

Classifications

• H04N25/44 Primary

  by partially reading an SSIS array · CPC title

• H04N23/80 Primary

  Camera processing pipelines; Components thereof · CPC title

• G06T5/70 Primary

  Denoising; Smoothing · CPC title

• G06T5/73 Primary

  Deblurring; Sharpening · CPC title

• G06T2207/10048

  Infrared image · CPC title