# How Eye Trackers Detect Eyelid Movements and Blinks > This patent describes a method for eye tracking systems to precisely detect eyelid movements and blinks by comparing a current eye image to a reference image, then analyzing the differences in rows of pixels. - **Patent:** US 20230004223 - **Original title:** Eye tracking system - **Owner:** Tobii AB - **Status:** Active - **Times cited:** 1 - **Field:** consumer_electronics, software, telecommunications, automotive, gaming, ai_ml ## What it does This eye tracking system works by taking two pictures of a user's eye: a 'reference image' and a 'current image' (Claim 1). It then finds the differences between these two images, creating a 'differential image' where only what has changed is visible. This differential image is like a grid of pixels, each showing how much it changed (Claim 1). The system then adds up or combines the change values in each horizontal row of pixels to get 'row values' (Claim 1, Claim 6). Finally, it uses these row values to figure out 'eyelid data,' such as the exact position of an eyelid edge (Claim 7) or whether the eye is blinking (Claim 9). For example, if you blink, the system would see a big change in certain rows of the differential image, indicating a blink. ## What it does NOT cover - Does not cover eye tracking systems that detect eyelid movement without first creating a differential image by comparing a current image to a reference image. - Does not cover systems that determine eyelid data solely from analyzing individual pixels or columns, rather than combining differential intensity values in corresponding rows (Claim 1). - Does not cover eye tracking methods that rely only on infrared reflections or pupil dilation without image differencing for eyelid detection. - Does not cover systems that track the entire eye's movement without specifically processing images to extract eyelid location or blink data. - Does not cover methods for determining eyelid data that do not involve a two-dimensional pixel array with differential intensity values. ## The clever bit The clever part is using a 'differential image' created by subtracting a reference image from a current one, then combining pixel intensity differences specifically along rows. This highlights only the eyelid movement, making it easier and potentially faster to detect changes like blinks or eyelid position without needing complex analysis of the entire eye image. ## Real-world examples 1. Virtual reality headsets with eye tracking 2. Augmented reality glasses 3. Driver monitoring systems in cars 4. Assistive technologies for people with disabilities 5. Gaming peripherals with eye control 6. Market research tools for attention analysis 7. Clinical diagnostic tools for eye conditions ## Why it matters Accurate eyelid tracking is crucial for many applications, from understanding user attention in virtual reality to diagnosing neurological conditions. By precisely identifying eyelid positions and blinks, this technology can improve the reliability of eye-tracking data, making it more useful for both research and commercial products. This precision helps prevent false readings and provides richer information about a user's state. ## Frequently asked questions ### What does How Eye Trackers Detect Eyelid Movements and Blinks cover? This patent describes a method for eye tracking systems to precisely detect eyelid movements and blinks by comparing a current eye image to a reference image, then analyzing the differences in rows of pixels. ### Who owns patent US 20230004223? This patent is owned by Tobii AB. ### When does this patent expire? This patent is expected to expire on June 30, 2042, when the invention enters the public domain. ### What is patent US 20230004223 cited by? This patent has been cited by 1 later patents that build on its ideas. ### What problem does this patent solve? Accurate eyelid tracking is crucial for many applications, from understanding user attention in virtual reality to diagnosing neurological conditions. By precisely identifying eyelid positions and blinks, this technology can improve the reliability of eye-tracking data, making it more useful for both research and commercial products. This precision helps prevent false readings and provides richer information about a user's state. ### What does this patent NOT cover? Does not cover eye tracking systems that detect eyelid movement without first creating a differential image by comparing a current image to a reference image. **Full plain-English explainer:** https://patentbrief.org/patent/us/20230004223/eye-tracking-system **Original patent:** https://patents.google.com/patent/US20230004223 --- _Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._ ## Related patents Semantically similar inventions in the PatentBrief corpus: - [How Eye Tracking Systems Adapt When They Can't See Eye Glints](https://patentbrief.org/patent/us/10878237/systems-and-methods-for-performing-eye-gaze-tracking) — This patent describes an eye-tracking system that intelligently switches between two methods: one using bright reflections on the eye when available, and another using head position and facial features when those reflections are absent. - [How Computers Learn to Track Your Eyes Better](https://patentbrief.org/patent/us/9727135/gaze-calibration) — This patent describes a method for eye-tracking systems to get more accurate by first asking you to look at specific points, and then secretly improving its accuracy as you use the computer normally. - [How Computers Find Your Pupil Even With Glare](https://patentbrief.org/patent/us/10692210/recording-medium-storing-computer-program-for-pupil-detection-information-proces) — This patent describes a computer method to accurately find the outline of a person's eye pupil by using radial search lines and handling bright reflections differently from clear areas. - [Using Different Light Colors to Detect Touches on Screens](https://patentbrief.org/patent/us/20110234542/methods-and-systems-utilizing-multiple-wavelengths-for-position-detection) — This 2011 patent describes a system that uses two different colors of light to accurately detect where an object, like a finger, touches a screen or surface. - [System for Controlling a Computer Using Hand Images](https://patentbrief.org/patent/us/9317130/visual-feedback-by-identifying-anatomical-features-of-a-hand) — This 2016 patent describes a system that uses a camera under a transparent surface to track a user's hand and display corresponding cursors on a screen, allowing hand gestures to control a computer interface.