False Color Correction in RGBW Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods fail to effectively correct false colors in high brightness regions of small areas, leading to image quality deterioration, particularly in images captured with RGBW array cameras, as they cannot handle brightness false colors effectively without compromising resolution.
Innovation Solution
An image processing device and method that detects false colors in high brightness regions by converting RGBW array images to RGB array images using a data conversion processing unit, which includes a false color detection unit, a low-band signal calculation unit, and a pixel interpolation unit. The units apply different low-pass filter coefficients based on the presence of false colors to calculate and interpolate low-band signals, ensuring proportional relations between W and RGB pixel signals, thereby correcting false colors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If an optical low-pass filter is used to decrease high-frequency components and correct false colors, then false color correction is improved, but resolution decreases and image quality deteriorates
Solution Approach 1:
The image processing is segmented into multiple stages: false color detection, selective low-pass filtering based on detection results, and interpolation processing. This segmentation allows different processing methods to be applied to different regions, correcting false colors while preserving resolution in non-affected areas.
Solution Approach 2:
The low-pass filter coefficients are adjusted locally based on false color detection results. In regions where false colors are detected, stronger low-pass filtering is applied, while in other regions, weaker filtering or no filtering is applied. This local adaptation corrects false colors without unnecessarily degrading resolution in the entire image.
2Adaptability or versatility
If conventional purple fringing processing is applied, then lens aberration false colors are corrected, but brightness false colors in small high brightness regions cannot be handled
Solution Approach 1:
The invention changes the processing parameters (low-pass filter coefficients) based on the detected false color type and location. Different coefficient sets are applied depending on whether the region is a small high brightness region or a conventional edge region, enabling effective correction of brightness false colors while maintaining compatibility with conventional false color processing.
Data Source
AI summary
There are provided a device and method that correct a false color occurring in a locally highlighted region in an image. A false color pixel is detected in data conversion processing of generating an RGB array image from an RGBW array image, low-band signals corresponding to respective RGBW colors that are different according to whether a pixel is a false color pixel, and the RGBW array is converted by interpolation processing to which the calculated low-band signals are applied to generate the RGB array image. The interpolation processing is performed using the low-band signals on an assumption that a W low-band signal mW, and RGB respective low-band signals mR, mG, and mB have a proportional relation in a local region. When a pixel of interest is a false color pixel, the low-band signal is calculated by an application of a low-pass filter having a coefficient in which a contribute rate of pixel values in the vicinity of the pixel of interest is made relatively lower than that of separated pixels.