Multispectral Image Sensor Demosaicing with Adjacent Filter Arrangement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Multiband cameras with five or more color filters arranged in front of a solid-state image sensing device face challenges in demosaicing, leading to deterioration of image definition and resolution, especially when color filters with adjacent peak wavelengths are not adjacent, resulting in false colors and reduced spatial frequency accuracy.
Innovation Solution
An image processing apparatus and color image sensor that uses five or more color filters, where high-definition and low-definition filters with adjacent peak wavelengths are alternately arranged, performing interpolation using the results from high-definition filter interpolation to enhance color information for both types of filters, thereby maintaining image definition during demosaicing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If five or more kinds of color filters are arranged in front of the image pickup device to improve spectral information reproduction, then the number of color filter types increases, but definition deteriorates due to false colors in demosaicing
Solution Approach 1:
The patent segments the color filters into two categories: high-definition color filters (first color filters) and low-definition color filters (second color filters). This segmentation allows different regions of the sensor to have different levels of detail, with high-definition filters providing accurate color information in critical areas and low-definition filters filling in remaining areas, thereby maintaining overall image definition while supporting multiple spectral bands.
Solution Approach 2:
The patent applies local quality by alternating arrangements of high-definition and low-definition color filters in specific regions. High-definition color filters are strategically placed in areas where accurate color reproduction is most important, while low-definition filters are positioned in other regions. This creates spatially varying quality levels that optimize both spectral information reproduction and image definition.
2Adaptability or versatility
If color filters with adjacent peak wavelengths are not adjacent to each other, then spectral coverage is improved, but spatial frequency accuracy deteriorates due to false colors
Solution Approach 1:
The patent segments color filters into high-definition and low-definition groups, where high-definition color filters are specifically designed to have adjacent peak wavelengths positioned next to each other. This segmentation ensures that in regions with high-definition filters, spatial frequency accuracy is maintained while spectral coverage is expanded through the low-definition filters in other regions.
Solution Approach 2:
The patent implements local quality by creating different local arrangements: regions with high-definition color filters have adjacent peak wavelengths for accurate spatial frequency reproduction, while regions with low-definition filters have broader spectral coverage. This local differentiation allows the system to optimize both spectral coverage and spatial frequency accuracy in different areas simultaneously.
3Device complexity
If three kinds of color filters (R, G, B) are used for demosaicing, then the method is simple, but definition deteriorates when five or more color filters are used
Solution Approach 1:
The patent segments the demosaicing process into two stages: first processing high-definition color filters, then processing low-definition color filters using the results from the first stage. This segmentation allows the system to maintain the simplicity of traditional three-color-filter demosaicing methods for high-definition regions while extending capability to handle five-or-more color filters in low-definition regions, thereby preserving image definition.
Solution Approach 2:
The patent applies preliminary action by first performing demosaicing on high-definition color filters to generate accurate color information, then using these results as a foundation for processing low-definition color filters. This preliminary processing of high-definition regions provides a reference that guides the demosaicing of low-definition regions, preventing definition deterioration even when five or more color filter types are used.
Data Source
AI summary
An image processing apparatus includes a unit to obtain mosaic image data by shooting a subject with use of a color image sensor in which five or more kinds of color filters are arranged in front of an image pickup device; a first interpolator to perform interpolation to color information of a high-definition color filter, in regard to the obtained mosaic image data; and a second interpolator to perform interpolation to color information of a low-definition color filter of which a peak wavelength in spectral transmittance is adjacent to that of the high-definition color filter, by using a result of the interpolation by the first interpolator, wherein the mosaic image data is image data which has one color information for each pixel, and the five or more kinds of color filters include the plural high-definition color filters and the plural low-definition color filters, thereby performing demosaicing without deteriorating definition.