US 12,435,985 B2
Multi-modal cognitive mechanism for road section recognition
Si Tong Zhao, Xian (CN); Jing Wen Xu, Shanghai (CN); Zhong Fang Yuan, Xi'an (CN); Ya Dong Li, Beijing (CN); Hai Bo Zou, Beijing (CN); and Xuan Yin Xia, Beijing (CN)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by International Business Machines Corporation, Armonk, NY (US)
Filed on Nov. 30, 2021, as Appl. No. 17/456,895.
Prior Publication US 2023/0168093 A1, Jun. 1, 2023
Int. Cl. G01C 21/34 (2006.01); G06F 18/214 (2023.01); G06N 20/00 (2019.01); G06V 20/56 (2022.01)
CPC G01C 21/3461 (2013.01) [G01C 21/3492 (2013.01); G06F 18/2148 (2023.01); G06N 20/00 (2019.01); G06V 20/56 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving, by one or more processors, an audio signal from a road test;
processing, by the one or more processors, the audio signal to generate an acoustic spectrum density distribution map to identify a respective at least one road section switching point in a first mode;
processing, by the one or more processors, a spectrogram of sound waves of the audio signal to identify the respective at least one road section switching point in a second mode;
using, by the one or more processors, a machine learning model to predict an expected sound at each frame of the audio signal, to calculate a similarity between the expected sound and an actual sound, and to identify the respective at least one road switching point when the similarity is lower than a pre-set similarity threshold in a third mode; and
combining, by the one or more processors, results of the first mode, the second mode, and the third mode to obtain a final set of road section switching points.