Stereoscopic Image Processing via Standard Viewpoint Edge Enhancement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods for generating high-definition stereoscopic images from a single camera system face challenges in avoiding Moire fringes and spurious resolution due to insufficient pixel density, particularly when applying filtering processes like edge enhancement and noise elimination on parallax images.
Innovation Solution
An image processing method that uses a pixel array with non-parallax, left parallax, and right parallax pixels to generate temporary parallax images and a standard view point image, followed by edge enhancement or noise elimination on the standard image to produce high-resolution stereoscopic images, avoiding individual filtering of parallax images to prevent Moire fringes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If filtering processes (edge enhancement, noise elimination) are applied individually to each parallax image, then image quality is improved, but Moire fringes and spurious resolution occur due to insufficient sampling density
Solution Approach 1:
The patent applies filtering processes (edge enhancement, noise elimination) to the standard viewpoint image before generating parallax images. By performing the filtering action in advance on the standard image, the benefits of filtering are extended to the final parallax images without directly processing them, thus avoiding Moire fringes while maintaining image quality
Solution Approach 2:
The standard viewpoint image serves as an intermediary between the captured raw image and the final parallax images. The filtering operations are performed on this intermediate standard image, which then serves as the basis for generating the final parallax images, thereby mediating the harmful effects of direct filtering on parallax images
2Measurement precision
If higher resolution parallax images are generated using generated parallax images, then resolution is improved, but spurious resolution occurs due to insufficient density of sampling
Solution Approach 1:
The patent performs filtering and processing operations on the standard viewpoint image before it is used to generate the final parallax images. This preliminary processing ensures that the standard image has the necessary quality and resolution characteristics before being used as the basis for generating higher resolution parallax images, preventing spurious resolution
Solution Approach 2:
The patent uses the standard viewpoint image as a reference or template that is then used to generate the final parallax images. By copying and adapting the processed standard image to create the parallax images, the resolution and quality characteristics are transferred while avoiding the sampling issues that would arise from directly processing the original parallax images
Data Source
AI summary
The present invention includes the steps of: generating a temporary left parallax image using the pixel value of the left parallax pixels of the first image for each pixel; generating a temporary right parallax image using the pixel value of the right parallax pixels of the first image for each pixel; generating a standard view point image using the pixel values of at least the non-parallax pixels of the first image for each pixel; generating an edge enhanced standard view point image, by carrying out an edge enhancement process on the standard view point image; and generating a left direction view point image and a right direction view point image based on the edge enhanced standard view point image and the temporary left parallax image and the temporary right parallax image for each pixel.