Infrared Image Offset Correction via Scale Factor Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Infrared image sensors face challenges in performing offset correction without mechanical shutters, which are cumbersome and time-consuming, and existing image processing techniques are complex and inadequate.
Innovation Solution
A method involving a processing device that calculates scale factors to generate column and pixel offset values based on reference images, applying these values to input images to produce corrected images without the need for mechanical shutters, using equations that incorporate high-pass filters and gradient minimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a mechanical shutter is used for offset correction, then calibration can be performed to correct spatial non-uniformity, but the device weight increases, cost increases, and the shutter is fragile
Solution Approach 1:
The patent replaces the mechanical shutter system with a purely software-based image processing algorithm. The method uses a pixel array to capture images and applies computational techniques (including gradient calculation, column spread analysis, and 2D dispersion correction) to achieve offset correction without any moving mechanical parts, thereby eliminating the associated weight, cost, and fragility issues
Solution Approach 2:
The patent uses reference images captured by the same pixel array to create correction data. By capturing reference images of a uniform scene and processing them to extract offset information, the system creates a digital copy of the non-uniformity pattern that can then be used to correct subsequent images, replacing the need for physical shutter-based calibration
2Measurement precision
If a mechanical shutter is used for calibration, then offset correction can be achieved, but image capture is interrupted during calibration periods
Solution Approach 1:
The patent enables continuous image capture without interruption for calibration. The method processes images continuously using software-based offset correction algorithms, eliminating the need to stop image capture for mechanical shutter calibration. This maintains uninterrupted observation while achieving the necessary correction
Solution Approach 2:
The patent performs offset correction calculations on reference images in advance to generate correction parameters. These pre-computed correction values are then applied to subsequent images in real-time, allowing continuous capture without interruption while maintaining correction accuracy
3Device complexity
If existing image processing techniques are used for offset correction, then mechanical shutters can be avoided, but the techniques are complex and do not adequately correct the image
Solution Approach 1:
The patent divides the offset correction process into distinct segments: column spread correction and 2D dispersion correction. By separating these correction tasks and applying specialized algorithms to each, the method achieves comprehensive correction without requiring overly complex integrated systems, maintaining simplicity while improving accuracy
Solution Approach 2:
The patent transforms the correction approach by changing key parameters: using gradient calculations instead of simple pixel comparisons, applying column spread analysis to capture spatial variations, and using 2D dispersion matrices to model non-uniformity. These parameter changes enable accurate correction with relatively simple processing
Data Source
AI summary
The invention concerns a method of image processing involving: receiving, by a processing device, an input image (IB) captured by a pixel array sensitive to infrared radiation; determining, based on the input image and on a column component vector (VCOL), a first scale factor (α) by estimating a level of the column spread present in the input image; generating column offset values (α.VCOL(y)) based on the product of the first scale factor with the values of the vector; determining, based on the input image and on a 2D dispersion matrix (IDISP), a second scale factor (β) by estimating a level of the 2D dispersion present in the input image; generating pixel offset values (β.IDISP(x,y)) based on the product of the second scale factor with the values of the matrix; and generating a corrected image (IC′) by applying the column and pixel offset values.