US 11,995,897 B2
Device for detecting the type and the level of clothing of a passenger of a motor vehicle utilizing a neural network
Daniel Neveu, Le Mesnil Saint Denis (FR); and Josselin Gour, Le Mesnil Saint Denis (FR)
Assigned to VALEO SYSTEMES THERMIQUES, Le Mesnil-Saint-Denis (FR)
Appl. No. 16/622,022
Filed by Valeo Systemes Thermiques, Le Mesnil-Saint-Denis (FR)
PCT Filed Jun. 1, 2018, PCT No. PCT/FR2018/051270
§ 371(c)(1), (2) Date Apr. 7, 2021,
PCT Pub. No. WO2018/229384, PCT Pub. Date Dec. 20, 2018.
Claims priority of application No. 1755507 (FR), filed on Jun. 16, 2017.
Prior Publication US 2021/0232834 A1, Jul. 29, 2021
Int. Cl. G06V 20/00 (2022.01); G06V 10/143 (2022.01); G06V 20/59 (2022.01); H04N 5/33 (2023.01)
CPC G06V 20/597 (2022.01) [G06V 10/143 (2022.01); G06V 20/59 (2022.01); H04N 5/33 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A device for identifying a clothing family index, an index representative of the type and of the level of clothing of a passenger of a motor vehicle, comprising:
at least one camera arranged to pick up at least one image of a part of the clothing of the passenger; and
a processing unit arranged to:
determine at least one clothing family index by using a recognition algorithm based on a model of neural network type, trained over a library of representative images picked up by the camera and previously categorized by a human operator,
determine a rate of confidence on the recognition of the at least one clothing family,
when the rate of confidence is above a predetermined threshold of 90%, validate the clothing index without intervention from the passenger, and
inform the passenger of the clothing family or families selected to manage his or her comfort.