Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
SYSTEM FOR CLASSIFYING THE USAGE OF A HANDHELD CONSUMER DEVICE
Document Type and Number:
WIPO Patent Application WO/2019/159130
Kind Code:
A3
Abstract:
The present disclosure is concerned with a system for classifying the usage of a handheld consumer device, the system having a movable handheld consumer device, the consumer device having a sensor unit for determining usage data at successive time instants, the sensor unit being arranged for providing a temporally successive sequence of at least one usage data during a usage session, and an analyzing device being arranged for classifying the usage of the consumer device with respect to at least one set of at least two usage classes relating to different usage properties, the analysis device being arranged for assembling a temporally successive sequence of input tuples of usage data relating to a predetermined time period of the usage session, each of the input tuples having at least one element representing the usage data at the respective time instant, and for inputting the sequence of input tuples into at least one artificial neural network that is arranged to output at least one output tuple that comprises a number of elements in accordance with the number of usage classes, each element of the output tuple representing a prediction value that the usage of the consumer device at the given time instant relates to a respective usage class.

Inventors:
SHERMAN FAIZ (US)
MAO XIAOLE (US)
Application Number:
PCT/IB2019/051250
Publication Date:
October 17, 2019
Filing Date:
February 15, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BRAUN GMBH (DE)
International Classes:
G06K9/00; G06K9/62
Foreign References:
CN107203753A2017-09-26
Other References:
SHUANGQUAN WANG ET AL: "Human Activity Recognition with User-Free Accelerometers in the Sensor Networks", NEURAL NETWORKS AND BRAIN, 2005. ICNN&B '05. INTERNATIONAL CONFERENCE ON BEIJING, CHINA 13-15 OCT. 2005, PISCATAWAY, NJ, USA,IEEE, vol. 2, 13 October 2005 (2005-10-13), pages 1212 - 1217, XP010906397, ISBN: 978-0-7803-9422-3, DOI: 10.1109/ICNNB.2005.1614831
Download PDF: