# US 12,770,433 B2 · Method and device for detecting changes in eye status > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,770,433 B2 - Document ID: US12770433B2 - Status: Granted - Kind: B2 - Application number: 18/745,237 - Filed: 2024-06-17 - Published: 2025-09-11 - Granted: 2026-10-20 - Earliest claimed: 2024-03-11 - Assignees: Wistron Corporation - Applicants: Wistron Corp. - Inventors: Chih Hao Chiu - Web page: https://patentgrep.com/patent/US12770433B2 ## CPC - G06V 40/18: Eye characteristics, e.g. of the iris - G06V 20/597: {Recognising the driver's state or behaviour, e.g. attention or drowsiness} - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method for detecting changes in eye status is provided. The method includes receiving multiple consecutive facial images generated by a photography device photographing a person's face, wherein each facial image includes an eye region. The method includes obtaining a partial image including the eye region from each of the facial images. The method includes obtaining an eye contour area based on the partial image. The method includes obtaining standard deviations of eye contour heights based on the eye contour area. The method includes determining a change in the eye status of eyes within a time interval based on the standard deviations of all eye contour heights within the time interval, wherein the eye status includes an eye-open status and an eye-closed status. ## Claims ### Claim 1 A method for detecting changes in eye status, comprising: receiving multiple consecutive facial images generated by a photography device photographing a person's face, wherein each facial image includes an eye region; obtaining a partial image including the eye region from each of the facial images; obtaining an eye contour area based on the partial image; obtaining standard deviations of eye contour heights based on the eye contour area; using a smoothing window to sequentially perform a smoothing process on the standard deviations of eye contour heights corresponding to the multiple consecutive facial images to obtain average values of the standard deviations of eye contour heights; and determining a change in the eye status of eyes within a time interval based on the standard deviations of all eye contour heights within the time interval, wherein the determining the change in the eye status of eyes within the time interval further comprises: determining that the change in the eye status of eyes within the time interval is from the eye-closed status to an eye-open status when the average values of the standard deviations of eye contour heights in the time interval continue to increase; and determining that the change in the eye status of eyes within the time interval is from the eye-open status to an eye-closed status when the average values of the standard deviations of eye contour heights in the time interval continue to decrease. ### Claim 2 (depends on 1) The method for detecting changes in eye status as claimed in claim 1, wherein the step of obtaining an eye contour area based on the partial image further comprises: performing a grayscale conversion on the partial image to obtain a grayscale image; performing a binarization process on the grayscale image to obtain a binary image; and performing a dilation process and an erosion process on the binary image to obtain the eye contour area. ### Claim 3 (depends on 1) The method for detecting changes in eye status as claimed in claim 1, wherein the step of obtaining standard deviations of eye contour heights based on the eye contour area further comprises: obtaining an array of eye contour height differences according to the eye contour area; retrieving a portion of eye contour height difference data from the array of eye contour height difference; and calculating the standard deviations of the eye contour heights based on the portion of eye contour height difference data. ### Claim 4 (depends on 3) The method for detecting changes in eye status as claimed in claim 3, wherein the portion of the eye contour height difference data is composed of the first N % of the array of eye contour height differences and the last N % of the array of eye contour height differences. ### Claim 5 (depends on 1) The method for detecting changes in eye status as claimed in claim 1, wherein: the step of determining that the change in the eye status of eyes within the time interval is from the eye-closed status to the eye-open status further requires that a current average value of a standard deviation of a current eye contour height at a current time point is greater than an eye-open threshold; and the step of determining that the change in the eye status of eyes within the time interval is from the eye-open status to the eye-closed status further requires that the current average value of the standard deviation of the current eye contour height at the current time point is less than an eye-closed threshold. ### Claim 6 (depends on 5) The method for detecting changes in eye status as claimed in claim 5, wherein the eye-open threshold and the eye-closed threshold are obtained according to the following steps: obtaining an array of the average values of the standard deviations of eye contour heights according to a preset time interval; retrieving a portion of average values of the standard deviations of eye contour heights data from the array of the average values of the standard deviations of eye contour heights; averaging the portion of the average values of the standard deviations of eye contour heights data to obtain a dynamic average; and setting the dynamic average multiplied by a first coefficient as the eye-open threshold, and setting the dynamic average multiplied by a second coefficient as the eye-closed threshold. ### Claim 7 (depends on 6) The method for detecting changes in eye status as claimed in claim 6, wherein a sum of the first coefficient and the second coefficient is 1 and the first coefficient is greater than the second coefficient. ### Claim 8 (depends on 1) The method for detecting changes in eye status as claimed in claim 1, wherein the standard deviations of the eye contour heights comprise a standard deviation of a right eye contour height or/and a standard deviation of a left eye contour height. ### Claim 9 (depends on 1) The method for detecting changes in eye status as claimed in claim 1, wherein when the eyes are in the eye-closed status for more than a preset time interval, a warning message is sent to notify the person. ### Claim 10 A device for detecting changes in eye status, comprising: one or more processors; and one or more computer storage media for storing one or more computer-readable instructions, wherein the processor is configured to drive the computer storage media to execute the following tasks: receiving multiple consecutive facial images generated by a photography device photographing a person's face, wherein each facial image includes an eye region; obtaining a partial image including the eye region from each of the facial images; obtaining an eye contour area based on the partial image; obtaining standard deviations of eye contour heights based on the eye contour area; using a smoothing window to sequentially perform a smoothing process on the standard deviations of eye contour heights corresponding to the multiple consecutive facial images to obtain average values of the standard deviations of eye contour heights; and determining a change in the eye status of eyes within a time interval based on the standard deviations of all eye contour heights within the time interval, wherein, in determining the change in the eye status of eyes within the time interval, the processor further executes the following tasks: determining that the change in the eye status of eyes within the time interval is from the eye-closed status to an eye-open status when the average values of the standard deviations of eye contour heights in the time interval continue to increase; and determining that the change in the eye status of eyes within the time interval is from the eye-open status to an eye-closed status when the average values of the standard deviations of eye contour heights in the time interval continue to decrease. ### Claim 11 (depends on 10) The device for detecting changes in eye status as claimed in claim 10, wherein the step of obtaining an eye contour area based on the partial image executed by the processor further comprises: performing a grayscale conversion on the partial image to obtain a grayscale image; performing a binarization process on the grayscale image to obtain a binary image; and performing a dilation process and an erosion process on the binary image to obtain the eye contour area. ### Claim 12 (depends on 10) The device for detecting changes in eye status as claimed in claim 10, wherein the step of obtaining standard deviations of eye contour heights based on the eye contour area executed by the processor further comprises: obtaining an array of eye contour height differences according to the eye contour area; retrieving a portion of eye contour height difference data from the array of eye contour height difference; and calculating the standard deviations of the eye contour heights based on the portion of eye contour height difference data. ### Claim 13 (depends on 12) The device for detecting changes in eye status as claimed in claim 12, wherein the portion of the eye contour height difference data is composed of the first N % of the array of eye contour height differences and the last N % of the array of eye contour height differences. ### Claim 14 (depends on 10) The device for detecting changes in eye status as claimed in claim 10, wherein the tasks executed by the processor further comprise; the step of determining that the change in the eye status of eyes within the time interval is from the eye-closed status to the eye-open status further requires that a current average value of a standard deviation of a current eye contour height at a current time point is greater than an eye-open threshold; and the step of determining that the change in the eye status of eyes within the time interval is from the eye-open status to the eye-closed status further requires that the current average value of the standard deviation of the current eye contour height at the current time point is less than an eye-closed threshold. ### Claim 15 (depends on 14) The device for detecting changes in eye status as claimed in claim 14, wherein the eye-open threshold and the eye-closed threshold are obtained according to the following steps executed by the processor: obtaining an array of the average values of the standard deviations of eye contour heights according to a preset time interval; retrieving a portion of average values of the standard deviations of eye contour heights data from the array of the average values of the standard deviations of eye contour heights; averaging the portion of the average values of the standard deviations of eye contour heights data to obtain a dynamic average; and setting the dynamic average multiplied by a first coefficient as the eye-open threshold, and setting the dynamic average multiplied by a second coefficient as the eye-closed threshold. ### Claim 16 (depends on 15) The device for detecting changes in eye status as claimed in claim 15, wherein a sum of the first coefficient and the second coefficient is 1 and the first coefficient is greater than the second coefficient. ### Claim 17 (depends on 10) The device for detecting changes in eye status as claimed in claim 10, wherein the standard deviations of the eye contour heights comprise a standard deviation of a right eye contour height or/and a standard deviation of a left eye contour height. ### Claim 18 (depends on 10) The device for detecting changes in eye status as claimed in claim 10, wherein when the eyes are in the eye-closed status for more than a preset time interval, a warning message is sent to notify the person.