Multi-Camera Image Fusion for Seamless Zoom and Noise Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems in electronic devices face challenges with high resource demands, limited image detail, and inconsistent field of view, particularly when combining images from wide-angle and telephoto cameras, leading to noisy and jarring transitions.
Innovation Solution
A method and system for aligning and fusing images from multiple cameras using a diffusion kernel and adaptive bandwidth averaging filter, guided by reference image structure, to enhance image quality and maintain seamless transitions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If images from wide-angle and telephoto cameras are combined, then image detail and field of view are improved, but noise and transition quality deteriorate
Solution Approach 1:
The patent combines images from wide-angle and telephoto cameras by aligning feature points between the two images and blending them using diffusion kernels. This merging approach allows the system to leverage the broad field of view from the wide-angle camera while incorporating the detailed telephoto regions, thereby improving overall image detail without simply concatenating the images which would introduce visible seams and noise.
Solution Approach 2:
The patent applies local quality by using adaptive diffusion kernels that adjust their parameters based on local image characteristics. The diffusion process is controlled to preserve edges and features in high-contrast regions while applying stronger smoothing in low-contrast areas. This local adaptation ensures that noise reduction does not blur important image details, and that the transition between wide-angle and telephoto regions is seamless.
2Reliability
If images from multiple cameras are processed, then image quality is improved, but processing resources increase
Solution Approach 1:
The patent segments the image processing task into distinct stages: feature point detection, alignment transformation, diffusion kernel application, and final blending. By dividing the processing workflow, the system can optimize each stage independently and process different regions of the image with appropriate computational effort, reducing overall resource consumption compared to uniform full-image processing.
Solution Approach 2:
The patent applies partial action by focusing computational resources only on regions where blending is necessary. The diffusion kernel is applied selectively based on the overlap between wide-angle and telephoto fields of view, rather than processing the entire image from both cameras. This partial processing approach maintains image quality while significantly reducing the computational burden.
3Object-affected harmful factors
If adaptive bandwidth filtering is applied, then noise is reduced, but processing complexity increases
Solution Approach 1:
The patent uses parameter changes by making the diffusion kernel bandwidth adaptive rather than fixed. The bandwidth parameter is adjusted dynamically based on local image contrast and gradient information. In regions with high contrast (edges, boundaries), the bandwidth is reduced to preserve detail, while in low-contrast regions, the bandwidth is increased for stronger noise reduction. This parameter adaptation achieves effective noise filtering without requiring complex multi-scale processing.
Data Source
AI summary
A method performed by an electronic device is described. The method includes obtaining a first image from a first camera, the first camera having a first focal length and a first field of view. The method also includes obtaining a second image from a second camera, the second camera having a second focal length and a second field of view disposed within the first field of view. The method further includes aligning at least a portion of the first image and at least a portion of the second image to produce aligned images. The method additionally includes fusing the aligned images based on a diffusion kernel to produce a fused image. The diffusion kernel indicates a threshold level over a gray level range. The method also includes outputting the fused image. The method may be performed for each of a plurality of frames of a video feed.