Panoramic Starry Sky Imaging via Multi-Exposure Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for capturing panoramic starry sky images face challenges due to the diurnal motion of astronomical bodies, which causes stars to appear as tracks of light in long exposures, and the limited maximum change in shooting direction restricts the capture of wider-angle views.
Innovation Solution
An image capturing device that captures a first image at a longer exposure, followed by shorter exposure images in different regions, with alignment processing using the shorter exposure images to combine the images correctly, allowing for the generation of a high-quality panoramic image despite star movement between shots.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If long exposure is used to capture starry sky images, then the image quality and brightness are improved, but the stars appear as tracks of light due to diurnal motion
Solution Approach 1:
The patent divides the panoramic image capture into multiple short-exposure unit images taken at different directions, rather than using a single long exposure. This segmentation allows each image to capture stars as points rather than tracks, while the panoramic combination process reconstructs the complete starry sky view.
Solution Approach 2:
The patent performs preliminary alignment processing using short-exposure images to determine the relative positions and movements of stars before combining the long-exposure images. This preliminary action establishes the correct positioning relationship, allowing the final combination to account for stellar motion and produce accurate star positions in the panoramic image.
2Area of stationary object
If multiple images are shot from different directions to create a panoramic view, then the angle of view is widened, but the positioning accuracy fails due to star movement between shots
Solution Approach 1:
The patent performs preliminary alignment processing using short-exposure images taken between the long-exposure shots. This preliminary action calculates the stellar movement and establishes the correct positioning relationship before combining the panoramic images, ensuring that stars appear at accurate positions even though they have moved during the capture sequence.
Solution Approach 2:
The patent uses short-exposure images as an intermediary to bridge the positioning gap between long-exposure panoramic images. These intermediary images capture the star positions at specific moments and provide the reference data needed to correctly align and combine the panoramic images, overcoming the positioning failure caused by stellar motion.
3Measurement precision
If optical shake correction means are used to correct positional skew, then the positioning accuracy is improved, but the maximum change in shooting direction is limited
Solution Approach 1:
The patent segments the panoramic capture into multiple unit images taken at different directions with short exposures, rather than relying on a single optical shake correction system. This allows the system to capture images across a wider range of directions without being constrained by the maximum correction capability of optical shake correction means.
Solution Approach 2:
The patent transitions from correcting position within a single image plane (optical shake correction) to correcting position across multiple image dimensions through computational alignment. By using alignment processing on multiple short-exposure images taken at different directions, the system achieves positioning accuracy without limiting the shooting direction range.
Data Source
AI summary
An image capturing device includes image sensor, capturing a first image in a first image capturing region, capturing a second image in the first image capturing region at a shorter exposure time than that of the first image, after capturing the first image, capturing a third image in a second image capturing region at a shorter exposure time than that of the first image, after capturing the second image; and capturing a fourth image in the second image capturing region at a longer exposure time than those of the second and third images, after capturing the third image; and combining unit configured to carry out alignment processing between the images using the second image and the third image, and combine the first image and the fourth image on the basis of a result of the alignment processing.