CMOS Image Sensor Circuit Dynamic Range Linearity Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CMOS image sensors face challenges in achieving wide dynamic range (WDR) with low linearity, high signal-to-noise ratio (SNR), high fill factor, and low output data rate, particularly due to limitations in self-reset pixel designs and digital noise coupling.
Innovation Solution
A CMOS image sensor circuit with a self-reset pixel approach, partial-quantization architecture, and interpolation-based digital calibration method to improve linearity and reduce fixed-pattern noise, along with activity-triggered data rate reduction using temporal difference circuits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If self-reset pixel design is used to expand dynamic range, then dynamic range is improved, but linearity deteriorates due to nonlinear well capacity
Solution Approach 1:
The patent applies preliminary action by performing digital calibration before actual imaging to characterize and store the nonlinear well capacity of each pixel. The calibration process pre-measures the relationship between integrated voltage and photocurrent for each pixel, storing this characterization data for later use during normal operation to correct linearity errors.
Solution Approach 2:
The patent implements feedback by using the stored calibration data to continuously correct the nonlinear well capacity effects during image capture. The system reads the stored calibration characteristics and applies compensatory calculations to the measured signals, creating a feedback loop that maintains high linearity across the extended dynamic range.
2Illumination intensity
If digital circuits are integrated in pixel for self-reset functionality, then dynamic range is improved, but digital noise couples into analog pixel reducing signal-to-noise ratio
Solution Approach 1:
The patent applies segmentation by completely separating digital and analog circuits into different domains. Digital circuits (comparators, counters, calibration logic) are placed in digital pixels while analog circuits (photodetectors, floating diffusion nodes) remain in analog pixels. This spatial segmentation prevents digital noise from coupling into the sensitive analog imaging path while preserving self-reset functionality.
Solution Approach 2:
The patent uses an intermediary approach by introducing dedicated digital buffer circuits and isolation structures between digital and analog domains. These intermediary elements act as noise barriers, allowing digital control signals to pass while blocking digital switching noise from reaching the analog photodetector nodes.
3Measurement precision
If continuous full-frame imaging is performed, then image quality is maintained, but output data rate increases
Solution Approach 1:
The patent implements periodic action by using activity-triggered imaging where the imaging system alternates between low-data-rate differential mode and high-quality WDR mode based on scene activity detection. The system periodically checks for activity and only activates full WDR capture when needed, reducing overall data rate while maintaining image quality during important events.
Solution Approach 2:
The patent applies dynamics by making the imaging mode adaptive and changeable based on real-time conditions. The system dynamically switches between different operational modes (differential mode, WDR mode, calibration mode) depending on scene activity, allowing optimal balance between data rate and image quality under varying conditions.
4Illumination intensity
If multiple sampling with different integration times is used, then dynamic range is improved, but device complexity increases
Solution Approach 1:
The patent applies parameter changes by utilizing the natural variation of well capacity with integrated voltage as a built-in mechanism for dynamic range extension. Instead of adding complex multi-capacitor structures or multiple photodetectors per pixel, the system leverages the inherent parameter change of well capacity to achieve WDR with a single photodetector and simple readout circuitry.
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
The solution achieves a dynamic range of 95.3 dB with improved linearity and SNR, a high fill factor of 27%, and reduced output data rate, effectively addressing the limitations of existing WDR CMOS image sensors.
Implementation Method 1
a photodiode for generating a photocurrent in response to incident light
Data Source
AI summary
Described herein is a circuit and related method for improving the dynamic range and the linearity characteristic of a CMOS image sensor. In various embodiments of the CMOS image sensor, a current sampler, a comparator, and a 1-bit memory are incorporated in each pixel circuit. In the image sensor, pixels are arranged in columns and a column slice is used to read the digital and analog singles from each column. In addition, a calibration circuit is incorporated in the sensor circuit for providing calibration current, which is used to generate calibration parameter. The image sensor operates in three non-overlapping modes: the difference mode, the WDR mode, and the calibration mode. The image sensor is switched among the three modes by control signals, which are provided to the image sensor by various control circuits. The image sensor normally operates in the difference mode and switches to the WDR mode when the difference between consecutive frames is over a threshold. The calibration mode allows the image sensor generate calibration parameters which are used to improve the linearity of the sensor through a interpolation method.