Pixel-Wise Dark Current Modeling for Shutterless Uncooled Cameras
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating dark current patterns in shutterless uncooled cameras are not robust and generic, particularly when temperature and exposure time cannot be controlled, leading to inaccurate corrections in applications like video acquisition and single image acquisition without mechanical shutters.
Innovation Solution
A method that characterizes each pixel of a 2D sensor independently based on exposure time and temperature, using a pseudo-empirical model with spline interpolation to estimate dark current patterns without on-site measurements, dividing the pattern into rate and offset terms, and employing a simplified temperature ramp for efficient implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dark reference image is taken just before the image of interest using a mechanical shutter, then the dark current correction accuracy is improved, but the device complexity and acquisition time increase due to the mechanical shutter requirement
Solution Approach 1:
The patent extracts and removes the mechanical shutter component from the system by using a shutterless acquisition mode. Instead of physically blocking the sensor, the method characterizes dark current through software-based estimation using laboratory-obtained images and pixel-wise modeling, thereby eliminating the complexity and reliability issues associated with mechanical shutters while maintaining correction accuracy
Solution Approach 2:
The patent replaces the mechanical shutter system with a computational approach. Dark current patterns are estimated using pixel-wise models fitted from laboratory data, combining elements of calibration-based methods with scene-based processing. This substitution eliminates moving parts while achieving comparable or superior correction performance through mathematical modeling and interpolation techniques
2Ease of manufacture
If existing methods fit laboratory data with temperature-dependent equations, then the method can be implemented without on-site measurements, but the robustness and generality are reduced
Solution Approach 1:
The patent segments the dark current characterization into two independent components: a temperature-dependent rate term and an exposure time-dependent offset term. Each component is modeled separately using pixel-wise approaches, allowing the method to capture complex non-uniformity patterns without requiring oversimplified global equations. This segmentation enables both ease of implementation and high robustness
Solution Approach 2:
The patent transforms the dark current model from a single temperature-dependent equation into a dual-parameter model separating rate and offset components. By fitting these parameters independently from laboratory images and using interpolation for temperature extrapolation, the method achieves both simplicity of implementation and accuracy across varying conditions, improving robustness while maintaining ease of use
3Measurement precision
If the dark current pattern is characterized for each pixel as a function of temperature and exposure time, then the estimation accuracy for uncontrolled conditions is improved, but the data acquisition and processing complexity increases
Solution Approach 1:
The patent performs comprehensive pixel-wise characterization of dark current as a function of temperature and exposure time during an initial laboratory calibration phase. This preliminary action creates a library of fitted parameters for each pixel that can be later interpolated for any operating condition, eliminating the need for complex real-time measurements while maintaining high estimation accuracy in field applications
Solution Approach 2:
The patent extends the dark current model from a single-dimensional approach to a two-dimensional parameter space by independently modeling both temperature dependence (rate term) and exposure time dependence (offset term). This dimensional separation allows the method to accurately predict dark current under any combination of uncontrolled temperature and variable exposure time using simple interpolation, reducing real-time processing complexity
Data Source
Figure 1A~1C
Figure 1D~2
Figure 3~4
AI summary
A dark current pattern estimation method of a 2D sensor for a shutterless uncooled camera, wherein the 2D sensor comprises a matrix of pixels (x, y), and being the dark current pattern divided into rate and offset. The method comprises the steps of setting a new temperature, waiting for the new temperature to stabilize, acquiring averaged images at at least two texp, alternatively and continuously, obtaining from each averaged image a dark current offset and a dark current rate for each pixel, interpolating the dark current offsets obtained for each pixel, interpolating the dark current rate obtained for each pixel, and obtaining the dark current pattern in terms of temperature, exposure time and pixel position.