# US 10,369,926 B2 · Driver state sensing system, driver state sensing method, and vehicle including the same > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,369,926 B2 - Document ID: US10369926B2 - Status: Granted - Kind: B2 - Application number: 15/680,161 - Filed: 2017-08-17 - Published: 2018-10-25 - Granted: 2019-08-06 - Earliest claimed: 2017-04-25 - Assignees: Mando-Hella Electronics Corporation - Applicants: Mando-Hella Electronics Corporation - Inventors: Shinwook Kim - Web page: https://patentgrep.com/patent/US10369926B2 ## CPC - B60Q 9/00: Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00 {, e.g. haptic signalling} - A61B 5/0077: {Devices for viewing the surface of the body, e.g. camera, magnifying lens} - A61B 5/1128: {using image analysis (A61B5/1127 takes precedence)} - A61B 5/18: for vehicle drivers {or machine operators} - G06V 20/597: {Recognising the driver's state or behaviour, e.g. attention or drowsiness} - G08B 21/06: indicating a condition of sleep, e.g. anti-dozing alarms - A61B 5/163: {by tracking eye movement, gaze, or pupil change} - B60Q: ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G08B: SIGNALLING SYSTEMS, e.g. PERSONAL CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS ## Abstract Disclosed are a driver state sensing system, a driver state sensing method, and a vehicle including the same. The driver state sensing system includes an image photographing unit composed of a single channel to photograph images, a controller that extracts at least one shape point from a driver image sensed in the images and compares a motion vector calculated based on the amount of change of the shape point with at least one drowsiness vector stored in advance to determine whether a driver is in a drowsiness state, and a warning unit to warn when it is determined that the driver is in a drowsiness state. ## Claims ### Claim 1 A driver state sensing system comprising: an image photographing unit comprising a channel to photograph images; a controller that extracts at least two shape points from driver images sensed in the images and compares a motion vector calculated based on an amount of change of the shape points with at least one drowsiness vector stored in advance to determine whether a driver is in a drowsiness state; and a warning unit to warn when it is determined that the driver is in a drowsiness state. ### Claim 2 (depends on 1) The driver state sensing system according to claim 1, wherein the controller calculates the motion vector based on a first shape point of a first frame captured at a first time and a second shape point of a second frame captured a second time which is different from the first time. ### Claim 3 (depends on 2) The driver state sensing system according to claim 2, wherein the controller comprises a frame buffer to store the first frame of the first time and the second frame of the second time. ### Claim 4 (depends on 1) The driver state sensing system according to claim 1, wherein the controller determines that a driver is in the drowsiness state and learns the calculated motion vector if difference between the calculated motion vector and the at least one drowsiness vector stored in advance has an error smaller than a preset first threshold value. ### Claim 5 A driver state sensing method comprising: photographing images based on a camera sensor; extracting at least two shape points from driver images sensed in the images; calculating a motion vector based on an amount of change of the shape points; comparing the calculated motion vector with at least one drowsiness vector stored in advance to determine whether a driver is in a drowsiness state; and warning when it is determined that the driver is in the drowsiness state. ### Claim 6 (depends on 5) The driver state sensing method according to claim 5, wherein the calculating the motion vector comprises: storing a first shape point of a first frame captured at a first time; storing a second shape point of a second frame captured at a second time which is different from the first time; and calculating the motion vector based on the first shape point and the second shape point. ### Claim 7 (depends on 5) The driver state sensing method according to claim 5, wherein the comparing the calculated motion vector with the at least one drowsiness vector comprises: determining that the driver is in the drowsiness state if difference between the calculated motion vector and the at least one drowsiness vector stored in advance has an error smaller than a preset first threshold value. ### Claim 8 (depends on 7) The driver state sensing method according to claim 7, further comprising learning the calculated motion vector. ### Claim 9 A vehicle comprising a driver state sensing system, wherein the driver state sensing system comprises: an image photographing unit comprising a channel to photograph images; a controller that extracts at least two shape points from driver images sensed in the images and compares a motion vector calculated based on an amount of change of the shape points with at least one drowsiness vector stored in advance to determine whether a driver is in a drowsiness state; and a warning unit to warn when it is determined that the driver is in the drowsiness state. ### Claim 10 (depends on 9) The vehicle according to claim 9, wherein the image photographing unit comprises a camera sensor, and the camera sensor is installed at a position of at least one of a front pillar, a rearview mirror, and a dashboard of the vehicle to photograph the driver. ### Claim 11 (depends on 10) The vehicle according to claim 10, wherein the controller calculates the motion vector based on a first shape point of a first frame captured at a first time and a second shape point of a second frame captured at a second time which is different from the first time. ### Claim 12 (depends on 11) The vehicle according to claim 11, wherein the controller determines that the driver is in the drowsiness state and learns the calculated motion vector if the difference between the calculated motion vector and the at least one drowsiness vector stored in advance has an error smaller than a preset first threshold value.