US 11,989,369 B1
Neural network-based touch input classification
Nadav Shlomo Ben-Amram, Herzliya (IL); Adam Hakim, Tel Aviv (IL); Roy-Gan Maiberger, Kiryat Ono (IL); Anatoly Tsvetov, Kfar Yona (IL); Yoel Yehezkel Einhoren, Petah Tiqwa (IL); and Etai Zajonts, Tel Aviv (IL)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Mar. 30, 2023, as Appl. No. 18/193,188.
Int. Cl. G06F 3/041 (2006.01); G06F 3/0354 (2013.01); G06F 3/044 (2006.01); G06N 3/04 (2023.01)
CPC G06F 3/04166 (2019.05) [G06F 3/03545 (2013.01); G06F 3/0416 (2013.01); G06F 3/0418 (2013.01); G06F 3/044 (2013.01); G06N 3/04 (2013.01); G06F 2203/04108 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A touch detection device, comprising:
an array of antennas configured to measure touch input and output a touch matrix of pixels having touch values corresponding to the touch input measured at each antenna of the array of antennas; and
a neural network having an input layer including a plurality of nodes, each node configured to receive a touch value corresponding to a different pixel of the touch matrix, the neural network configured to output classified touch data corresponding to the measured touch input based at least on the touch matrix.