Ophthalmological Image Processing Apparatus Reducing Projection Artifacts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ophthalmological images captured by imaging devices like OCT and fundus cameras often suffer from artifacts at the retinal deep layer due to fluctuations or shadows from the retinal shallow layer, which existing technologies fail to effectively address.
Innovation Solution
An ophthalmological image processing apparatus and program that utilize a first front image from a shallower depth region to correct a second front image from a deeper region, by dividing the images into local areas, detecting blood vessel positions, and calculating weights based on similarity degrees to minimize artifacts, thereby reducing projection artifacts in the deeper layer images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ophthalmological imaging is used to capture deep layer images, then the deep layer structures can be visualized, but projection artifacts from shallow layer blood vessels appear and degrade image quality
Solution Approach 1:
The imaging data is segmented into multiple depth regions, with a first depth region (shallower) and a second depth region (deeper). By separating the analysis into distinct depth segments, the patent can independently process and correct artifacts in the deeper layer using information from the shallower layer, thereby improving deep layer image quality while eliminating projection artifacts.
Solution Approach 2:
The patent uses the shallower depth region image as an intermediary to correct the deeper depth region image. The blood vessel information from the first depth region serves as a mediator to identify and remove corresponding projection artifacts in the second depth region, enabling artifact reduction without losing deep layer structural information.
2Measurement precision
If image correction is applied to remove artifacts, then artifact reduction is achieved, but risk of over-correction or introducing new artifacts increases
Solution Approach 1:
The patent applies correction with locally optimized weights to each pixel or region in the deeper layer image based on corresponding features in the shallower layer. This localized approach ensures that correction is applied appropriately only where projection artifacts exist, preventing over-correction in regions without artifacts and maintaining reliability of the correction process.
Solution Approach 2:
The patent modifies image parameters (pixel intensities) in the deeper layer image by a controlled amount based on the shallower layer information. By adjusting the correction magnitude using optimized weights and combining it with the original deep layer image, the patent achieves artifact reduction while maintaining image integrity and avoiding over-correction.
Data Source
AI summary
An ophthalmological image processing apparatus includes an image obtaining portion for obtaining ophthalmological image data, which is captured by an imaging unit, of a subject eye, and an image processing portion for processing the ophthalmological image data. The image processing portion uses a first front image of the subject eye in a first depth region to correct a second front image of the subject eye captured in a second depth region which is different from the first depth region. The first front image and the second front image are constructed based on the ophthalmological image data. The ophthalmological image processing apparatus can reduce an artifact which is generated in an ophthalmological image.