Ophthalmic HDR Image Generation Using Blink-Similarity Screening
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ophthalmic apparatuses face disturbances in generating high dynamic-range images due to blinks during imaging, as existing methods assume identical imaging conditions, leading to inconsistencies when actual images differ.
Innovation Solution
The apparatus includes an imaging means to capture eye images at varying exposure conditions, a storage means to store images with time stamps, and a high dynamic-range generation means that updates images only when similarity is confirmed, using threshold values to handle blinks and image changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If images are combined to generate high dynamic-range images using conventional methods, then the dynamic range is improved, but image disturbances occur when blinks are included in the captured images
Solution Approach 1:
The system performs preliminary actions by storing multiple images captured at different exposure conditions before generating the high dynamic-range image. It also preliminarily identifies and flags images containing blinks or abnormalities, preventing them from being combined into the final image, thus avoiding disturbances while maintaining high dynamic range.
Solution Approach 2:
The system incorporates feedback mechanisms by evaluating the quality and appropriateness of each captured image before inclusion in the high dynamic-range composite. It uses algorithms to detect blinks and abnormalities, providing feedback to exclude problematic images, thereby ensuring reliable image quality while preserving the extended dynamic range achieved through multiple exposure conditions.
2Manufacturing precision
If multiple images are processed to create high dynamic-range images, then the image quality is improved, but the processing time increases
Solution Approach 1:
The system performs preliminary sorting and quality assessment of captured images before the actual high dynamic-range processing. By pre-identifying suitable images and excluding those with blinks or abnormalities early in the workflow, it reduces the number of images requiring intensive processing, thereby maintaining high image quality while reducing overall processing time.
Solution Approach 2:
The system extracts and removes problematic images (those containing blinks or abnormalities) from the processing queue before combining the remaining valid images. This extraction of unsuitable data reduces the computational burden and processing time while preserving the quality of the final high dynamic-range image through selective combination of only appropriate images.
3Productivity
If all captured images are used for high dynamic-range generation, then the data utilization is maximized, but disturbances occur from blink images
Solution Approach 1:
The system extracts and separates images containing blinks or abnormalities from the total set of captured images. By removing these problematic images before combining the rest, it maintains high data utilization from valid images while ensuring image consistency and avoiding disturbances in the final high dynamic-range output.
Solution Approach 2:
The system applies local quality assessment to each individual image in the sequence, evaluating specific characteristics such as blink detection and image stability. This localized quality control allows selective inclusion of only appropriate images in the high dynamic-range composite, maximizing the use of valid data while maintaining overall image consistency and avoiding disturbances from problematic frames.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An imaging means (10) for periodically taking test subject eye images for each different exposure condition, a storage means (20) for storing the eye images with capturing time information as imaging data, and the selected eye image from the imaging data as base image, and an arithmetic unit (50) for storing a copy of the base image in the storage means (20) as an output image, performing HDR processing on the output image to generate an HDR image, and determining identicalness of the eye image to the base image are provided, wherein the arithmetic unit (50) performs updating the output image by superimposing the eye image on the output image to store the output image in the storage means (20) on the eye images in the same exposure condition to generate the HDR image when the eye image and the base image are determined to be the same chronologically.