Linear-Logarithmic Image Sensor Pixel Calibration via Floating Diffusion Voltage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional CMOS image sensors require complex and computationally expensive image signal processing algorithms for calibration, which can be time-consuming and introduce significant fixed pattern noise, especially when transitioning between linear and logarithmic responses.

Innovation Solution

On-the-fly calibration methods that measure voltages at the pixel's floating diffusion region using sub-threshold currents, allowing for photocurrent-independent calibration without the need for large frame memories or extensive ISP algorithms, by depleting the photodiode and reading voltages through a totem pole arrangement of transistors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional calibration techniques using external frame memory and complex ISP algorithms are used, then calibration accuracy can be achieved, but device complexity and computational cost increase significantly

Engineering Contradiction:
Improvecalibration accuracyVSAvoidcomplexity of calibration system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the calibration function from external frame memory and complex ISP algorithms, implementing it directly within the pixel circuit itself. By integrating the calibration functionality into the pixel's transfer transistor and floating diffusion region, the system eliminates the need for large external memory structures and complex post-processing algorithms, thereby reducing device complexity while maintaining calibration accuracy.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The pixel circuit performs its own calibration using its inherent components (transfer transistor, floating diffusion region) without requiring external calibration hardware. The calibration process utilizes the pixel's own sub-threshold current characteristics and voltage measurements, enabling the system to self-calibrate and reducing dependency on external resources.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If conventional calibration techniques with full frame capture and external frame memory are used, then calibration can be performed, but loss of time occurs due to computational expense

Engineering Contradiction:
Improvecalibration accuracyVSAvoidcalibration time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs calibration measurements during the pixel readout process itself, rather than as a separate post-processing step. By measuring voltages at the floating diffusion region during normal operation and using these measurements to determine calibration parameters, the system eliminates the time-consuming separate calibration phase while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The calibration process is merged with the pixel readout operation. The same circuit paths and measurement mechanisms used for normal image capture are utilized for calibration, combining two functions into one unified process and eliminating the time loss associated with separate calibration operations.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If conventional calibration methods are used, then calibration can be achieved, but fixed pattern noise increases significantly

Engineering Contradiction:
Improvesensor performanceVSAvoidfixed pattern noise
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent replaces complex mechanical/electronic calibration systems (external frame memory, multi-step ISP algorithms) with an electrical measurement approach using sub-threshold currents and voltage readings. This substitution simplifies the calibration mechanism and reduces the introduction of fixed pattern noise that occurs with complex processing chains.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach reduces fixed pattern noise and simplifies the calibration process, enabling efficient and accurate calibration of linear-logarithmic image sensors, independent of photocurrent, with calibration measurements that can be performed adjacent to pixel readout, thus improving image sensor performance.

Implementation Method 1

a photoelectric conversion characteristic defined by a linear characteristic area where an electrical signal is outputted after being linearly converted in relation to the amount of incident light, and a logarithmic characteristic area where the electrical signal is outputted after being logarithmically converted in relation to the amount of incident light

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Implementation Method 2

measuring a voltage at a pixel floating diffusion (FD) region after at least one sub-threshold current is enabled to flow from a photodiode (PD) to the FD region via a transfer transistor

Methodology Applied
Scientific EffectSub-threshold conduction:

Data Source

PatentUS9843750B2Methods of calibrating linear-logarithmic image sensors
Publication Date: 2017.12.12 SAMSUNG ELECTRONICS CO LTD
  • US9843750B2 patent drawing
  • US9843750B2 patent drawing
  • US9843750B2 patent drawing

AI summary

Methods of calibrating a linear-logarithmic image sensor pixel include performing a reset of the pixel in advance of establishing a leakage current between a photodiode and a floating diffusion region of the pixel. A first voltage of the floating diffusion region is then read through a source follower and selection transistor, after the leakage is terminated. A step is then performed to transfer charge between the photodiode and the floating diffusion region of the pixel so that a voltage of a cathode of the photodiode is increased. Thereafter, a second voltage of the floating diffusion region is read. The first and second read voltages are then used to perform a calibration operation. These steps may be repeated to establish another leakage current of different duration/magnitude and yield third and fourth read voltages, which support further calibration.