Image Compositing Using In-Focus Degree Information
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating composited images by positional alignment suffer from image quality deterioration in overlapping regions, particularly affecting the detection of small defects like cracking.
Innovation Solution
An image processing apparatus and method that obtain in-focus degree information for overlapping regions between images and select the image with better focus to composite, thereby reducing image quality deterioration and enhancing defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If blend processing is performed on the overlapping region for positional alignment, then the joint between images becomes smooth and noticeable, but image quality deteriorates causing blurring in the overlapping region
Solution Approach 1:
The patent applies local quality by differentiating the processing method for overlapping regions versus non-overlapping regions. Instead of applying uniform blend processing to the entire image, the system identifies overlapping regions and applies selective processing only where needed, preserving sharpness in critical areas while maintaining smooth transitions in overlapping zones.
Solution Approach 2:
The patent segments the image into overlapping regions and non-overlapping regions, applying different processing strategies to each segment. The overlapping regions are processed to maintain continuity, while non-overlapping regions retain their original sharpness, thus resolving the contradiction between smooth joints and image quality preservation.
2Measurement precision
If blend processing is performed on the overlapping region, then positional alignment is achieved, but detection of small defects becomes difficult due to blurring
Solution Approach 1:
The patent applies local quality by differentiating the processing method for overlapping regions versus non-overlapping regions. Instead of applying uniform blend processing to the entire image, the system identifies overlapping regions and applies selective processing only where needed, preserving sharpness in critical areas while maintaining smooth transitions in overlapping zones.
Solution Approach 2:
The patent segments the image into overlapping regions and non-overlapping regions, applying different processing strategies to each segment. The overlapping regions are processed to maintain continuity, while non-overlapping regions retain their original sharpness, thus resolving the contradiction between smooth joints and image quality preservation.
3Measurement precision
If high-resolution images are used to detect small defects, then detection capability improves, but the complexity of processing multiple images increases
Solution Approach 1:
The patent merges multiple high-resolution images into a single composited image by leveraging their overlapping regions. Instead of processing each image separately, the system combines them strategically, using the overlap information to align and composite the images efficiently, thus reducing processing complexity while maintaining high-resolution detection capability.
Solution Approach 2:
The patent performs preliminary actions by capturing multiple images with intentional overlaps before the actual processing. This preliminary capture strategy enables subsequent automated alignment and compositing, reducing the complexity of processing by pre-organizing the image data structure through controlled overlapping during capture.
Data Source
AI summary
An image processing apparatus comprises one or more processors, wherein the one or more processors function as: a first obtaining unit configured to obtain a plurality of images having an overlapping region which is a region that is captured by overlapping at least a portion of a capturing target, a second obtaining unit configured to obtain in-focus degree information in the overlapping region of each of the plurality of images, wherein the in-focus degree information indicates a degree of focus for each predetermined region of the image, and a compositing unit configured to, based on the in-focus degree information of each of the plurality of images, generate a composited image for which the plurality of images are composited.


