Digital Signal Processing Reference
In-Depth Information
Table 1.5 Accuracy of distraction detection
Threshold
Maneuver
Measure
0.2
0.1
0 (Binary)
LKS
Count
72/96
62/96
84/96
76/96
95/96
76/96
Acc (%)
75
64
87
79
98
79
LKC
Count
65/113
64/113
82/113
79/113
95/113
79/113
Acc (%)
57
56
72
69
84
69
The system offers a low-cost, driver-dependent, and reliable distraction detection
submodule. Future work will focus on generic distraction detection using sums within
the same feature space.
1.5 Conclusions
In this study, the impact of cognitive load on drivers was analyzed using the
UTDrive database that comprises real-world driving recordings. In particular,
driver's speech signal and CAN-Bus signals were studied and subsequently utilized
in the design of autonomous speech and CAN-Bus domain neutral/stress (distraction)
classifiers. The speech-based neutral/stress classification reached an accuracy
of 88.2% in the driver-/maneuver-independent open test set task. The distraction
detector exploiting CAN-Bus signals was evaluated in a driver-/maneuver-dependent
closed test set task, providing 98% and 84% distraction detection accuracy in lane
keeping segments and curve negotiation segments, respectively. The results suggest
that future fusion of speech and CAN-Bus-based classifiers could yield a robust
continuous stress (distraction) assessment framework.
References
1. Neiberg D, Elenius K, Karlsson I, Laskowski K (2006) Emotion recognition in spontaneous
speech using GMMs. In: Proceedings of ICSLP'06, Pittsburgh, pp 809-812
2. Lee CM, Narayanan SS (2005) Toward detecting emotions in spoken dialogs. IEEE Trans
Speech Audio Proc 13:293-303
3. Ijima Y, Tachibana M, Nose T, Kobayashi T (2009) Emotional speech recognition based on
style estimation and adaptation with multiple-regression HMM. In: Proceedings of IEEE
ICASSP'09, Taipei, pp 4157-4160
4. Callejas Z, Lopez-Cozar R (2008) Influence of contextual information in emotion annotation
for spoken dialogue systems. Speech Comm 50(5):416-433
5. Hansen JHL (1996) Analysis and compensation of speech under stress and noise for environ-
mental robustness in speech recognition. Speech Comm 20, 151-173
6. Cummings K, Clements M (1990) Analysis of glottal waveforms across stress styles.
In: Proceedings of IEEE ICASSP'90, Albuquerque, vol 1, pp 369-372
Search WWH ::
Custom Search