Image Summarization via Loss Area Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image summarization methods fail to effectively delete images from large sequences without missing critical areas, particularly in medical applications where lesions or abnormalities might be overlooked due to inadequate consideration of object relationships between deleted and retained images.
Innovation Solution
An image processing device and method that selects a reference image and a determination target image, calculates a loss area based on deformation information, and determines whether the target image can be deleted by assessing the coverage area and unnecessary areas, ensuring that critical information is preserved.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If image summarization process deletes images from sequence, then number of images is reduced and checking time is decreased, but risk of missing critical areas such as lesions or abnormalities increases
Solution Approach 1:
The patent performs preliminary actions by calculating coverage areas and loss areas before deleting images. It computes deformation information between adjacent images, identifies coverage areas where objects are captured in both current and next images, and calculates loss areas representing regions that would be unobserved if deletion occurs. This preliminary analysis enables informed deletion decisions that reduce checking time while maintaining reliability by preserving images that cover critical areas.
Solution Approach 2:
The patent implements feedback mechanisms by using coverage area calculations and loss area assessments to determine whether to delete images. The system continuously evaluates the impact of potential deletions on observation coverage, using this feedback to make intelligent decisions about which images to retain. This feedback loop ensures that critical areas are not missed while still achieving significant reduction in the number of images to be checked.
2Quantity of substance
If image summarization process deletes images from sequence, then data volume is reduced, but information completeness about captured objects deteriorates
Solution Approach 1:
The patent performs preliminary calculations of coverage areas and loss areas before deleting images. By computing deformation information between adjacent images and identifying regions that would be unobserved upon deletion, the system can make informed decisions about which images to retain. This preliminary analysis reduces data volume while preserving information completeness by ensuring that critical object areas are not lost through deletion.
Solution Approach 2:
The patent uses feedback from coverage area calculations and loss area assessments to determine image deletion decisions. The system continuously evaluates whether deleting an image would result in loss of important object information, using this feedback to maintain information completeness while reducing overall data volume. The feedback mechanism ensures that only images with redundant coverage are deleted.
3Speed
If simple image summarization is applied without considering deformation information, then processing speed is improved, but accuracy in determining deletion suitability deteriorates
Solution Approach 1:
The patent performs preliminary calculations of coverage areas and loss areas using deformation information between adjacent images. By computing how objects deform and move between frames, the system accurately determines which images can be safely deleted. This preliminary analysis maintains high processing speed while improving accuracy in deletion suitability determination compared to simple summarization methods.
Solution Approach 2:
The patent implements feedback mechanisms that use coverage area and loss area calculations to continuously refine deletion decisions. The system evaluates deformation information and uses this feedback to accurately assess whether deleting an image would result in loss of critical object areas. This feedback-driven approach maintains processing speed while significantly improving the accuracy of deletion suitability determination.
Data Source
AI summary
An image processing device includes an image sequence acquisition section that acquires an image sequence that includes a plurality of constituent images, and a processing section that performs an image summarization process that deletes some of the plurality of constituent images included in the image sequence to generate a summary image sequence, the processing section selecting a reference image and a determination target image from the plurality of constituent images, detecting an unnecessary area from the determination target image, calculating a loss area based on deformation information about the reference image and the determination target image, and the unnecessary area, the loss area being an area that is lost when the determination target image is deleted, and determining whether or not the determination target image can be deleted based on the calculated loss area.