US 12,307,791 B2
Method for determining a drowsiness level of a motor vehicle driver
Alain Giralt, Toulouse (FR); and Martin Petrov, Toulouse (FR)
Assigned to Continental Automotive France, Toulouse (FR); and Continental Automotive GmbH, Hannover (DE)
Appl. No. 17/296,783
Filed by Continental Automotive France, Toulouse (FR); and Continental Automotive GmbH, Hannover (DE)
PCT Filed Dec. 13, 2019, PCT No. PCT/EP2019/085147
§ 371(c)(1), (2) Date May 25, 2021,
PCT Pub. No. WO2020/120760, PCT Pub. Date Jun. 18, 2020.
Claims priority of application No. 1872824 (FR), filed on Dec. 13, 2018.
Prior Publication US 2022/0027646 A1, Jan. 27, 2022
Int. Cl. G06V 20/59 (2022.01); B60W 40/08 (2012.01); B60W 50/14 (2020.01); B60W 60/00 (2020.01); G06F 18/214 (2023.01); G08B 21/06 (2006.01)
CPC G06V 20/597 (2022.01) [B60W 40/08 (2013.01); G06F 18/214 (2023.01); G08B 21/06 (2013.01); B60W 2040/0827 (2013.01); B60W 50/14 (2013.01); B60W 60/0051 (2020.02); B60W 60/0059 (2020.02); B60W 2420/403 (2013.01); B60W 2540/221 (2020.02); B60W 2540/223 (2020.02); B60W 2540/229 (2020.02)] 18 Claims
OG exemplary drawing
 
1. A method for determining a level of drowsiness of a driver of a motor vehicle, on the basis of a predetermined image analysis algorithm, said vehicle comprising a camera and a computer, said computer implementing said predetermined algorithm based on a set comprising at least one parameter relating to an attitude of the driver, the method, implemented by the computer, comprising:
a) a learning phase performed for a predetermined duration, comprising the steps of:
i) the camera generating a sequence of consecutive images of the driver,
ii) determining at least one characteristic point of the images of the generated sequence of images,
iii) during the learning phase, running the predetermined algorithm on the at least one characteristic point of the images of the generated sequence of images in a plurality of implementations performed simultaneously by the computer in order to perform a plurality of diagnostics, each implementation using a different set of at least one predetermined parameter and a different predetermined range of values for the at least one predetermined parameter, so as to determine a plurality of values for each parameter of said set,
iv) determining a different degree of relevance for each implementation performed, the degrees of relevance being different from one another, a highest degree of relevance being assigned to the set of the at least one predetermined parameter and the different predetermined range of values for the at least one predetermined parameter for which the determined plurality of values for each parameter of said set vary the least,
b) once the learning phase has ended, a phase of monitoring the state of the driver, comprising the steps of:
i) the camera generating a sequence of images of the driver,
ii) running the predetermined algorithm on said sequence of images generated in at least one implementation performed by the computer based on at least the set of the at least one predetermined parameter and the different predetermined range of values for the at least one predetermined parameter having the highest degree of relevance, so as to determine a plurality of values for each parameter of said set,
iii) determining a level of drowsiness of the driver based on the plurality of values determined for each parameter of said set and of at least one predetermined threshold relating to said at least one parameter.