Wide-Field Image Stitching With Weighted Similarity Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image acquisition methods fail to accurately combine multiple picked-up images due to abnormal positional deviations caused by container edges with high contrast, leading to inaccurate compositing of images.
Innovation Solution
An image acquisition apparatus and method that utilizes a similarity map to specify relative positional deviations, identifies specific image pairs with directivity, and adjusts weights to minimize the impact of unreliable deviations, enabling precise image combination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple picked-up images are acquired with overlapping regions to combine into one wide field image, then the field of view is expanded, but the combination accuracy deteriorates due to abnormal positional deviations caused by container edges with high contrast
Solution Approach 1:
The patent applies local quality by differentiating the treatment of image pairs based on their local characteristics. Image pairs are classified into first image pairs (without container edges in overlapping regions) and second image pairs (with container edges). Different weight values are assigned to these pairs during combination position determination, allowing the system to maintain high accuracy for reliable pairs while reducing the impact of unreliable pairs caused by container edges.
Solution Approach 2:
The patent changes the parameter of weight values assigned to different image pairs during the combination position determination process. By adjusting weight values based on the presence of container edges in overlapping regions, the system optimizes the combination accuracy. First image pairs receive higher weights while second image pairs receive lower weights, effectively managing the trade-off between field of view expansion and combination accuracy.
2Device complexity
If all image pairs are treated equally in combination position determination, then the process is simple, but the combination accuracy deteriorates due to the influence of abnormal positional deviations from container edges
Solution Approach 1:
The patent implements local quality by assigning different weights to different image pairs based on their local characteristics (presence of container edges in overlapping regions). This differentiated treatment allows the system to maintain simplicity in the overall process while improving accuracy by preventing unreliable image pairs from unduly influencing the combination positions.
Solution Approach 2:
The patent applies preliminary action by determining the weight values for image pairs before performing the combination position determination. The system preliminarily identifies which image pairs have container edges in their overlapping regions and pre-assignes appropriate weights, ensuring that abnormal positional deviations are accounted for before the main combination process begins.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables high-accuracy image combination by reducing the influence of unreliable positional deviations, resulting in a stable and accurate composite image.
Implementation Method 1
generating a similarity map indicating a distribution of degree of similarity by template matching on the overlapping region of the each image pair
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In an image acquisition apparatus, acquired are picked-up images 71 representing divided regions, respectively, which are obtained by dividing a predetermined region on a target object, and in the picked-up images 71, each image pair 72 representing adjacent divided regions has an overlapping region 73. The amount of relative positional deviation in each image pair 72 is specified by generating a similarity map 74 indicating a distribution of the degree of similarity. Combination positions of the picked-up images 71 are determined on the basis of the amounts of positional deviation in image pairs 72 included in the picked-up images 71 while it is assumed that an image pair 72 whose similarity map 74 has a directivity is a specific image pair 72 and a weight of the specific image pair 72 is set to be lower than those of other image pairs 72.