Event-Camera Ophthalmic Apparatus for Real-Time Face-Movement Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ophthalmologic apparatuses struggle to quickly and accurately check for face movement during alignment and ocular information acquisition, leading to potential misalignment, low precision in eye characteristic measurements, and image blurring.
Innovation Solution
An ophthalmologic apparatus equipped with an event camera that captures anterior ocular segment images and outputs event data with luminance changes, allowing real-time monitoring of face movement by extracting inner and outer canthus points to determine face position and movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional pupil image detection and face movement checking methods are used, then the system can determine face support status, but it cannot check face movement in real time during alignment and ocular information acquisition
Solution Approach 1:
The patent replaces conventional frame-based image processing with event camera technology that detects luminance changes asynchronously. This substitution enables real-time face movement detection during alignment without adding time overhead, as the event camera outputs data only when changes occur rather than continuously capturing full frames
Solution Approach 2:
The patent introduces an event camera as an intermediary device between the face support system and the control system. This intermediary captures luminance changes at the cornea surface and converts them into event data that indicates face movement, enabling real-time monitoring without disrupting the alignment process
2Measurement precision
If the examinee's face moves during alignment, then the positional relationship adjustment takes longer or fails, but conventional systems cannot detect this movement
Solution Approach 1:
The patent implements a feedback mechanism where the event camera continuously monitors luminance changes at the cornea surface and provides real-time movement information to the control system. This feedback loop enables immediate detection of face movement during alignment, allowing the system to adjust or terminate the alignment process before measurement precision deteriorates
Solution Approach 2:
The patent performs preliminary detection of face movement using the event camera before alignment is completed. By detecting luminance changes and determining face movement in advance, the system can prevent misalignment from occurring, thereby ensuring both alignment success and subsequent measurement precision
3Manufacturing precision
If the examinee's face moves during image capture, then image quality deteriorates with blur or flare, but conventional systems cannot prevent this
Solution Approach 1:
The patent uses the event camera to provide real-time feedback on face movement during image capture. When luminance changes indicate face movement, the system can immediately terminate or pause the capture process, preventing blur or flare while minimizing loss of capture time by only interrupting when necessary
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables rapid detection of face movement during alignment and ocular information acquisition, ensuring precise eye characteristic measurements and clear image capture by promptly adjusting the optical system and face support.
Implementation Method 1
an event camera that is installed at a position to capture an anterior ocular segment of the subject eye, the event camera being configured to output only event data, in which a luminance change has exceeded a set threshold from pixel data of an image frame
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An ophthalmologic apparatus (A) includes an optical system (50) that is configured to acquire ocular information of a subject eye (E) of an examinee; a face support (30, 33) that is configured to support a face of the examinee; a controller (60) that is configured to control at least the optical system (50) and the face support (30, 33). The ophthalmologic apparatus (A) further includes an event camera (80) that is installed at a position to capture an anterior ocular segment of the subject eye (E), the event camera (80) being configured to output only event data, in which a luminance change has exceeded a set threshold from pixel data of an image frame of the anterior ocular segment of the subject eye (E), and the event data being output together with information on coordinates and time as output information. The controller (60) includes a face movement monitoring processing portion (64) that is configured to extract inner canthus point (IC) and outer canthus point (OC) of the subject eye (E) by using the output information from the event camera (80) and to monitor a movement of the face of the examinee based on the extracted inner and outer canthus points (IC, OC).