IP Library Granted Patent US 12,684,261
Granted Patent B2
US 12,684,261 · App. 18/416,543 · Granted Jul 14, 2026

Defective pixel correction device and method of operating the same

Inventors: Chang Hoon Choi (Suwon-si, KR); Hyunyup Kwak (Suwon-si, KR); Jaeuk Ahn (Suwon-si, KR); Yongmi Lee (Suwon-si, KR); Seungwon Choi (Suwon-si, KR)
Assignee: SAMSUNG ELECTRONICS CO., LTD.
H04N25/683
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 12,684,261
App. No.
18/416,543
Filed
Jan 18, 2024
Granted
Jul 14, 2026
Kind
B2
Art Unit
2638
USPC
348/246
Abstract

A defective pixel correction device is provided. The defective pixel correction device includes: a pixel grading circuit configured to output a grade map including a plurality of grades indicating defect levels and respectively corresponding to a plurality of pixels included in an input image; a pixel selection circuit configured to select one or more candidate pixels from among the plurality of pixels, based on whether the plurality of grades correspond to a correction level, in the grade map; and a correction circuit configured to correct the one or more candidate pixels.

Claims (46)

1 . A defective pixel correction device comprising:

a pixel grading circuit configured to output a grade map comprising a plurality of grades indicating defect levels and respectively corresponding to a plurality of pixels included in an input image, wherein the plurality of pixels comprises red pixels, green pixels and blue pixels arranged in a regular pattern;

a pixel selection circuit configured to select one or more candidate pixels from among the plurality of pixels, based on whether the plurality of grades correspond to a correction level, in the grade map; and

a correction circuit configured to:

calculate one or more correction values based on the one or more candidate pixels and a plurality of neighboring pixels included in an image patch, wherein the image patch includes the one or more candidate pixels;

correct the one or more candidate pixels based on the one or more correction values; and

replicate the image patch for each of the one or more candidate pixels.

2 . The defective pixel correction device of claim 1 , wherein the pixel grading circuit is further configured to assign a grade to a target pixel of the plurality of pixels based on an average value or a median value of a plurality of neighboring pixels included in an image patch being compared with a pixel value of the target pixel.

3 . The defective pixel correction device of claim 2 , wherein the pixel grading circuit is further configured to:

assign a first grade as the grade of the target pixel based on satisfaction of a first condition in which a difference between the pixel value and the average value or the median value is less than a first threshold value;

assign a third grade as the grade of the target pixel based on satisfaction of a second condition in which the difference between the pixel value and the average value or the median value is greater than a second threshold value; and

assign a second grade as the grade of the target pixel based on the target pixel satisfying neither the first condition nor the second condition.

4 . The defective pixel correction device of claim 3 , wherein the correction level comprises at least one of the second grade and the third grade.

5 . The defective pixel correction device of claim 1 , wherein the pixel selection circuit is further configured to set at least one of the plurality of grades to the correction level based on a ratio of one of the plurality of grades in the grade map to another one of the plurality of grades in the grade map.

6 . The defective pixel correction device of claim 1 , wherein the correction circuit is further configured to calculate the one or more correction values for the one or more candidate pixels and the correcting for the one or more candidate pixels in parallel.

7 . The defective pixel correction device of claim 1 , further comprising a replication circuit configured to select one of a correction value and a reference value for each of the one or more candidate pixels.

8 . The defective pixel correction device of claim 7 , wherein the replication circuit is further configured to:

compare a first difference between one or more pixel values of the one or more candidate pixels before being corrected and the correction value, and a second difference between the one or more pixel values and the reference value; and

select the correction value based on the first difference being greater than the second difference, and select the reference value based on the first difference not being greater than the second difference.

9 . The defective pixel correction device of claim 1 , wherein the regular pattern is a Bayer pattern.

10 . A method of operating a defective pixel correction device, the method comprising:

assigning a plurality of grades, among grades indicating defect levels, to a plurality of pixels included in an input image, respectively, wherein the plurality of pixels comprises red pixels, green pixels and blue pixels arranged in a regular pattern;

outputting a grade map comprising the plurality of grades;

selecting one or more candidate pixels from among the plurality of pixels, based on whether the plurality of grades correspond to a correction level, in the grade map;

correcting the one or more candidate pixels;

replicating an image patch for each of the one or more candidate pixels; and

calculating one or more correction values based on the one or more candidate pixels and a plurality of neighboring pixels included in the image patch.

11 . The method of claim 10 , further comprising:

setting at least one of the plurality of grades to the correction level;

determining which of the plurality of grades correspond to the correction level; and

selecting the one or more candidate pixels based on which of the plurality of grades correspond to the correction level.

12 . The method of claim 10 , further comprising:

receiving a reference value for the one or more candidate pixels;

calculating a first difference between one or more pixel values of the one or more candidate pixels before being corrected and correction value of the one or more correction values, and a second difference between the one or more pixel values of the one or more candidate pixels before being corrected and the reference value; and

selecting the correction value based on the first difference being greater than the second difference, and selecting the reference value based on the first difference not being greater than the second difference.

13 . The method of claim 10 , wherein the calculating the one or more correction values and the correcting are performed for the one or more candidate pixels in parallel.

14 . The method of claim 11 , wherein the correction level is set based on a ratio of each of the plurality of grades in the grade map.

15 . An electronic device comprising:

an image sensor comprising red pixels, green pixels and blue pixels arranged in a regular pattern;

a pixel grading circuit configured to output a grade map comprising a plurality of pixels to which a grade, among a plurality of grades indicating defect levels, is assigned, based on the grade being assigned, wherein the plurality of pixels are included in an input image output from the image sensor;

a pixel selection circuit configured to select one or more candidate pixels, having the grade corresponding to a correction level, in the grade map; and

a correction circuit configured to:

calculate one or more correction values based on the one or more candidate pixels and a plurality of neighboring pixels included in an image patch, wherein the image patch includes the one or more candidate pixels;

correct the one or more candidate pixels based on the one or more correction values; and

replicate the image patch for each of the one or more candidate pixels.

16 . The electronic device of claim 15 , wherein the pixel selection circuit is further configured to set at least one of the plurality of grades to the correction level based on a ratio of one of the plurality of grades in the grade map to another one of the plurality of grades in the grade map.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 19, 2024
From: CHOI, CHANG HOON; KWAK, HYUNYUP; AHN, JAEUK; LEE, YONGMI; CHOI, SEUNGWON
To: SAMSUNG ELECTRONICS CO., LTD.
Reel/Frame 066185/0750 →
Priority Claims (2)
KR 10-2023-0040778 · Mar 28, 2023 · national
KR 10-2023-0090685 · Jul 12, 2023 · national
Continuity (1)
Related Publication 20240334084A1 · Oct 3, 2024
References Cited (27)
US 7280142B2 · Toyoda · 2007 [cited by examiner]
US 7920186B2 · Forutanpour · 2011 [cited by applicant]
US 8358319B2 · Cote et al. · 2013 [cited by applicant]
US 8971659B2 · Forutanpour · 2015 [cited by applicant]
US 9131196B2 · Lim et al. · 2015 [cited by applicant]
US 10158815B2 · Aflalo et al. · 2018 [cited by applicant]
US 10666884B1 · Campbell · 2020 [cited by examiner]
US 20020176623A1 · Steinberg · 2002 [cited by applicant]
US 20060164692A1 · Fujiwara et al. · 2006 [cited by applicant]
US 20100073526A1 · Watanabe · 2010 [cited by examiner]
US 20100086179A1 · Verboven et al. · 2010 [cited by applicant]
US 20100321537A1 · Zamfir et al. · 2010 [cited by applicant]
US 20110002545A1 · Steinberg et al. · 2011 [cited by applicant]
US 20110032393A1 · Yamaguchi · 2011 [cited by examiner]
US 20110149123A1 · Matsuoka · 2011 [cited by examiner]
US 20120133804A1 · Kim · 2012 [cited by examiner]
US 20120182452A1 · Yasuma · 2012 [cited by examiner]
US 20130329098A1 · Lim · 2013 [cited by examiner]
US 20140092280A1 · Tozawa · 2014 [cited by applicant]
US 20150288935A1 · Shinozaki · 2015 [cited by examiner]
US 20240292124A1 · Takaiwa · 2024 [cited by examiner]
Yap-Peng Tan et al., “A Robust Sequential Approach for the Detection of Defective Pixels in an Image Sensor,” 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing Proceedings, pp. 2239-2242. [cited by applicant]
Anthony A. Tanbakuchi et al., “Adaptive Pixel Defect Correction,” Sensors and Camera Systems for Scientific, Industrial, and Digital Photography Applications IV, Proceedings of SPIE-IS&T Electronic Imaging, SPIE vol. 50… [cited by applicant]
Edward Chang et al., “Kernel-size selection for defect pixel identification and correction”, Digital Photography III, SPIE-IS&T Electronic Imaging, SPIE vol. 6502, pp. 178-187 (Total 10 pages). [cited by applicant]
Noha El-Yamany et al., “Robust Defect Pixel Detection and Correction for Bayer Imaging Systems,” IS&T International Symposium on Electronic Imaging 2017 Digital Photography and Mobile Imaging XIIII, pp. 46-51. [cited by applicant]
Liu Yongji et al., “A Design of Dynamic Defective Pixel Correction for Image Sensor,” 2020 IEEE International Conference on Artificial Intelligence and Information Systems (ICAIIS), pp. 713-716. [cited by applicant]
Jeehoon An et al., “Adaptive Detection and Concealment Algorigthm of Defective Pixel,” 2007 IEEE Workshop on Signal Processing Systems, pp. 651-656. [cited by applicant]