Ophthalmologic Apparatus Automatic Eye Alignment via Facial Feature Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ophthalmologic apparatuses require manual intervention and examiner movement to align the subject's eye when it does not appear in the anterior ocular segment image during imaging start mode, especially in remote operations, due to difficulties in visual confirmation and alignment control.
Innovation Solution
An ophthalmologic apparatus with an alignment controller that automatically calculates and adjusts the predicted position of the subject's eye based on image recognition of facial features, allowing for automatic alignment without manual intervention, even when the eye is not initially in the anterior ocular segment image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual alignment adjustment is performed when the subject eye does not appear in the anterior ocular segment image, then the subject eye can be positioned correctly, but the examiner must move to the subject's position and spend additional time on preparation work
Solution Approach 1:
The system performs preliminary detection of the subject eye position using the anterior ocular segment camera before the main measurement. When the eye is not detected, the system automatically executes preliminary alignment actions (moving the optical head or adjusting the chin rest position) based on facial feature recognition, eliminating the need for manual intervention and examiner movement.
Solution Approach 2:
The system performs self-alignment by automatically detecting facial features, calculating the predicted eye position, and adjusting its own components (optical head or chin rest) without external human intervention. The alignment controller enables the system to service itself, particularly valuable in remote operations where the examiner cannot visually confirm or manually adjust.
2Measurement precision
If the examiner performs visual confirmation and manual operation for alignment, then accurate positioning can be achieved, but this becomes difficult or impossible in remote operations
Solution Approach 1:
The system replaces the mechanical/visual alignment process (examiner visually confirming eye position and manually adjusting components) with an automated image processing and control system. The anterior ocular segment camera captures facial images, the image processing unit detects features and calculates eye position, and the alignment controller automatically adjusts positioning, enabling remote operation without sacrificing accuracy.
Solution Approach 2:
The system introduces an intermediary automated alignment control system between the examiner and the physical alignment process. This intermediary uses image recognition algorithms and automated control to bridge the gap between remote examiner location and the physical positioning requirements, maintaining alignment accuracy without requiring the examiner's physical presence.
3Ease of operation
If automatic alignment control is implemented without requiring the subject eye to appear in the anterior ocular segment image, then remote operations become feasible, but the system must rely on image recognition of facial features to predict eye position
Solution Approach 1:
The system performs preliminary detection of facial features (eyes, nose, mouth, chin) using the anterior ocular segment camera before main measurement. Based on the detected facial landmarks, the system calculates the predicted eye position and uses this information to automatically adjust alignment, enabling remote operation while maintaining detection accuracy through multi-step image processing.
Solution Approach 2:
The system uses feedback from the anterior ocular segment camera images to continuously monitor and adjust alignment. The image processing unit analyzes facial features, calculates predicted eye position, and feeds this information back to the alignment controller, which adjusts the optical head or chin rest position accordingly. This closed-loop feedback system maintains accuracy even when the eye is not initially visible in the image.
Data Source
AI summary
An ophthalmologic apparatus includes a main body including a measurement optical system that measures eye characteristics of a subject eye; an anterior ocular segment camera provided in the main body to acquire an anterior ocular segment image by imaging an anterior ocular segment; and a controller that includes an alignment controller configured to control to adjust a relative positional relationship between the subject eye and the main body based on the anterior ocular segment image. The alignment controller includes an imaging start mode controller that is configured, when it is determined that the subject eye does not appear in the anterior ocular segment image in an imaging start mode by the anterior ocular segment camera, to calculate a predicted position of the subject eye based on image recognition of a face part in the anterior ocular segment image and to control movement toward the calculated predicted position.