US 10,764,519 B2Granted

System and method for high dynamic range digital double sampling

AssigneeGvbb Holdings S.A.R.L.

InventorsJeroen Rotte, Rik Visser, Juul Josephus Johannes Van Den Heijkant

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system and method is provided for performing high dynamic range digital double sampling. More particularly, a CMOS image sensor is provided that includes a pixel array with each pixel sampling both dark and bright values for digital double sampling. After the sampled signals are digitized, a mean dark value is determined and each dark value is further fed to a lookup table that generates an output value taking into account whether the pixel has been saturated. In over exposed conditions, the lookup table will generate a negative value output to eliminate image artifacts. All three values are fed to adder logic circuit that subtracts the mean dark value and the lookup table output from the bright value. This resulting output is fed to a video viewer.

Claims24

Dependent claims

An apparatus for correcting oversaturated pixels of an image sensor, the apparatus comprising:

a pixel sampler configured to sample a digital dark value and a digital bright value of a pixel in a pixel array of a CMOS image sensor during capture of an image, with the digital bright value corresponding to a sampled output of a first sub-pixel of the pixel in the pixel array;
a dark value average calculator configured to calculate a mean dark value of the pixel based on the sampled digital dark value and one or more previously stored dark values of the pixel;
a digital dark value adjuster configured to subtract the calculated mean dark value of the pixel from the sampled digital dark value to generate an adjusted dark value;
a lookup table configured to identify an illumination correction output value based on the generated and adjusted dark value;
an image correction signal generator configured to generate an image correction signal by subtracting the calculated mean dark value and the illumination correction output value from the sampled digital bright value;
an image generator configured to display the image captured by the CMOS image sensor on a display device without black sun artifacts based on the image correction signal;
wherein the image correction signal generator includes a digital logic circuit with inputs configured to receive the sampled digital bright value, the calculated mean dark value and the illumination correction output value, respectively;
wherein the illumination correction output value is configured as an optimal artifact correction signal configured to remove at least one artifact from the sampled digital bright value of the pixel during the capture of the image;
wherein the lookup table identifies a negative illumination correction output value for the sampled digital dark value if the pixel is in an overexposed illumination state during the capture of the image.

6 more elements

An apparatus for correcting oversaturated pixels of an image sensor, the apparatus comprising:

a pixel sampler configured to sample a digital dark value and a digital bright value of a pixel in a pixel array of a CMOS image sensor during image capture;
a dark value average calculator configured to calculate a mean dark value of the pixel based on the sampled digital dark value and one or more previously stored dark values of the pixel;
a digital dark value adjuster configured to subtract the calculated mean dark value of the pixel from the sampled digital dark value to generate an adjusted dark value;
a lookup table configured to identify an illumination correction output value based on the generated and adjusted dark value;
an image correction signal generator configured to generate an image correction signal by subtracting the calculated mean dark value and the illumination correction output value from the sampled digital bright value;
an image generator configured to display a captured image on a display device without black sun artifacts based on the image correction signal;
wherein the illumination correction output value is configured as an optimal artifact correction signal configured to remove at least one artifact from the sampled digital bright value of the pixel during the image capture.

4 more elements

Dependencies are inferred from the claim text.

Facts

Application: 16/571,660Filed 16 Sep 2019 · continuation
Priority: 8 Sep 2016
Published: US 2020/0014866 A19 Jan 2020
Granted: 1 Sep 2020
Term ends: Not estimatedThe source records don’t support an estimate.

Classification6 codes

H04N 25/65
Noise processing, e.g. detecting, correcting, reducing or removing noise: applied to reset noise, e.g. KTC noise related to CMOS structures by…
H04N 25/627
Detection or reduction of noise due to excess charges produced by the exposure, e.g. smear, blooming, ghost image, crosstalk or leakage between…
H04N 25/76
SSIS architectures; Circuits associated therewith: addressed sensors, e.g. MOS or CMOS sensors
H04N 25/78
Addressed sensors, e.g. MOS or CMOS sensors: readout circuits for addressed sensors, e.g. output amplifiers or A/D converters

Family

Application62/384,970Us provisional application · filed 8 Sep 2016
Application15/693,034Parent of this continuation · filed 31 Aug 2017
US 2020/0014866 A1Pre-grant publication · published 9 Jan 2020
US 10,419,697 B2Parent of this continuation
US 2018/0070032 A1Published application
US 10,764,519 B2This patent

From the US filing history. Foreign family members aren’t listed.