Ophthalmologic Apparatus Fundus Detection Edge Template Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ophthalmologic techniques face challenges in accurately identifying and reliably detecting specific sites of interest in the eye fundus, such as the optic nerve head, particularly in low-quality fundus images obtained through near infrared imaging, leading to inconsistencies in detection accuracy across different racial groups.
Innovation Solution
An ophthalmologic apparatus employing a combination of image processing methods, including edge detection and template matching, along with OCT imaging, to enhance the detection of interested sites in eye fundus images, utilizing multiple processors and optical systems to improve accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single image processing method is used to detect interested sites in fundus images, then the detection process is simple and fast, but the detection reliability and accuracy are insufficient, especially in low-quality images and across different racial groups
Solution Approach 1:
The patent segments the image processing task into multiple distinct methods: edge detection method, template matching method, and method using optical coherence tomography. Each method processes the fundus image independently to detect interested sites, and their results are integrated to improve overall detection reliability while maintaining processing efficiency through parallel execution of multiple algorithms
Solution Approach 2:
The patent merges results from multiple independent image processing methods (edge detection, template matching, OCT-based detection) to produce a final detection outcome. By combining the strengths of different methods, the system achieves higher detection reliability and accuracy across diverse populations while managing computational complexity through efficient result integration
2Measurement precision
If conventional image processing methods are used, then the processing speed is fast, but the detection accuracy varies significantly across different racial groups due to differences in image representation
Solution Approach 1:
The patent implements a multi-functional detection system that universally applies multiple image processing methods (edge detection, template matching, OCT analysis) to fundus images from different racial groups. This universal approach ensures consistent detection accuracy across diverse populations by not relying on a single method that may be biased toward specific racial characteristics
Solution Approach 2:
The patent applies different image processing methods with specialized characteristics to different detection needs: edge detection for structural boundaries, template matching for pattern recognition, and OCT for depth information. This local quality approach ensures that each method's unique strengths are utilized appropriately, improving overall accuracy across different racial groups with varying image characteristics
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An ophthalmologic apparatus of an embodiment example includes a front image acquiring device, a first search processor, and a second search processor. The front image acquiring device is configured to acquire a front image of a fundus of a subject's eye. The first search processor is configured to search for an interested region corresponding to an interested site of the fundus based on a brightness variation in the front image. The second search processor is configured to search for the interested region by template matching between the front image and a template image in the event that the interested region has not been detected by the first search processor.