Peripheral Image Stitching with Boundary Regions for 3D Overlap
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for generating combined images by superimposing multiple images on a projection surface often result in issues such as double parking lot lines, leading to distorted projections.
Innovation Solution
An image processing device that includes a determination unit to identify boundary regions where adjacent images overlap, using position information from detection points and self-position information to generate a combined image, and a projection conversion unit to adjust the projection surface based on detected points, ensuring accurate image alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple images are superimposed on a projection surface to generate a combined image, then the coverage area and information completeness are improved, but image alignment accuracy deteriorates causing double parking lot lines
Solution Approach 1:
The projection surface is divided into multiple regions, each corresponding to a different peripheral image. The determination unit segments the overlapping regions between adjacent images and identifies boundary regions where only one image should be displayed. This segmentation approach allows the system to maintain comprehensive coverage while preventing alignment errors by treating different regions with different processing rules.
Solution Approach 2:
The determination unit performs preliminary identification of boundary regions before the final image combination is displayed. By using position information and self-position information to pre-determine where images overlap and where boundaries should be placed, the system prevents the double line problem from occurring in the first place, rather than attempting to correct it afterward.
2Productivity
If peripheral images are combined without boundary region determination, then the processing speed is improved, but image quality deteriorates due to double imaging
Solution Approach 1:
The determination unit applies boundary region identification selectively only to overlapping areas between adjacent peripheral images, rather than processing the entire image set uniformly. This partial action approach maintains high processing speed for non-overlapping regions while applying precise boundary control only where needed, thus achieving both speed and accuracy.
3Manufacturing precision
If the projection surface is adjusted based on detection points, then the image alignment accuracy is improved, but the device complexity increases
Solution Approach 1:
The determination unit utilizes self-position information that is already available from the moving body's navigation system, rather than requiring additional external reference systems. By leveraging existing position data and detection points from the environment recognition system, the unit adjusts the projection surface autonomously without adding significant hardware complexity, thus achieving high alignment accuracy through software-based self-correction.
Data Source
AI summary
In an aspect, an image processing device includes a hardware processor configured to function as a determination unit and an image generation unit. The determination unit determines a boundary region in a region where spatially adjacent peripheral images of a plurality of peripheral images of a moving body overlap with each other, using position information including a plurality of detection points around the moving body and self-position information of the moving body. The image generation unit generates a combined image, using the spatially adjacent peripheral images, using the boundary region.