Surround View Image Stitching With De-Processed Color Harmonization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing surround view systems (SVS) suffer from color mismatches in stitched images due to different camera sensors capturing light under varying conditions and individual image processing, leading to noticeable artifacts at seams, which can distract drivers or autonomous systems.
Innovation Solution
Transfer color statistics from a ground projection of a reference frame to a target frame, rather than computing over entire images, to harmonize colors across multiple camera sensors, using statistical moments or properties of color channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If individual image processing is applied to each camera frame independently, then processing efficiency is maintained, but color mismatches occur at seams in stitched images
Solution Approach 1:
The patent segments the image processing pipeline by separating color statistics computation from the main processing flow. It extracts color statistics from reference frames independently, then applies these statistics to target frames, allowing parallel processing while ensuring color consistency at seams.
Solution Approach 2:
The patent performs preliminary computation of color statistics from reference frames before processing target frames. By pre-computing color statistics (mean, standard deviation) from reference frames, the system prepares color harmonization parameters in advance, enabling efficient application to multiple target frames without re-computation.
2Manufacturing precision
If color statistics are computed over entire images, then comprehensive color harmonization is achieved, but computational demands and latency increase
Solution Approach 1:
The patent extracts only the essential color statistics (mean and standard deviation of color channels) from reference frames, rather than processing entire images. This extraction approach captures the necessary color information while significantly reducing computational complexity and processing time.
Solution Approach 2:
The patent applies color harmonization selectively at seam regions rather than uniformly across entire images. By focusing color adjustment operations primarily at overlapping seam areas where mismatches are most noticeable, the system achieves effective color harmonization with reduced computational overhead.
Data Source
AI summary
In various examples, color harmonization is applied to images of an environment in a reference light space. For example, different cameras on an ego-object may use independent capturing algorithms to generate processed images of the environment representing a common time slice using different capture configuration parameters. The processed images may be transformed into deprocessed images by inverting one or more stages of image processing to transform the processed images into a reference light space of linear light, and color harmonization may be applied to the deprocessed images in the reference light space. After applying color harmonization, corresponding image processing may be reapplied to the harmonized images using corresponding capture configuration parameters, the resulting processed harmonized images may be stitched into a stitched image, and a visualization of the stitched image may be presented (e.g., on a monitor visible to an occupant or operator of the ego-object).