Segmented Sphere Projection for Spherical Camera Image Stitching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The Equirectangular Projection method causes significant distortion in the arctic and antarctic parts of spherical images relative to the equator part, leading to suboptimal panoramic image stitching.
Innovation Solution
The Segmented Sphere Projection method divides images into arctic, equator, and antarctic parts, using a similar-edge technique and feather blending for stitching, transforming them into semicircular and then spherical shapes to reduce distortion, and combining them with Equirectangular Projection for the equator parts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If Equirectangular Projection method is used to transform spherical image into rectangular shape, then the image can be processed and stitched, but the arctic part and the antartic part have great distortion relative to the equator part
Solution Approach 1:
The patent divides the spherical image into three separate projection zones: arctic zone (north pole region), equatorial zone (equator region), and antarctic zone (south pole region). Each zone is projected independently using appropriate projection methods, allowing differential handling of distortion in different regions while maintaining overall image coherence.
Solution Approach 2:
Different projection methods are applied to different regions of the spherical image. The equatorial region uses equirectangular projection, while the arctic and antarctic regions use alternative projections (such as stereographic or orthographic) that minimize polar distortion. This local optimization ensures each region is represented with minimal distortion appropriate to its location.
2Manufacturing precision
If the spherical image is divided into three parts and stitched using similar-edge method, then the distortion is reduced, but the stitching process becomes more complex
Solution Approach 1:
The patent segments the spherical panorama into three distinct projection zones that are processed and stitched separately. This segmentation allows each zone to be optimized independently with appropriate projection methods, reducing overall distortion while the modular stitching approach manages complexity through systematic processing of defined regions.
Solution Approach 2:
The patent introduces intermediate processing steps including overlay generation, feature point detection, and homography calculation as mediators between the divided image parts. These intermediary processes facilitate accurate alignment and blending of the three zones while maintaining a structured workflow that manages the complexity of the stitching operation.
3Stability of the object's composition
If feather blending method is used for overlapping area at stitching, then the transition between parts is smooth, but the processing time increases
Solution Approach 1:
The patent applies feather blending selectively only in the overlapping transition zones between adjacent projection areas, rather than processing the entire image. This partial application of the blending operation achieves smooth transitions at critical seams while minimizing the overall processing time by limiting the computational effort to only where it is most needed.
Data Source
AI summary
The present invention provides a method for Spherical Camera Image Stitching. By using two fisheye lens to catch two fisheye images and then being developed into three pairs of flat figures based on Segmented Sphere Projection (SSP) method. Thereafter each corresponding pair is stitched based on a similar-edge method, and then three pairs are combined to form a panoramic image. At the end, the combined panoramic image is projected to a 3-D ball sphere space.