Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Aug 9;19(16):3481.
doi: 10.3390/s19163481.

Feature Extraction Methods Proposed for Speech Recognition Are Effective on Road Condition Monitoring Using Smartphone Inertial Sensors

Frederico Soares Cabral  1 , Hidekazu Fukai  2 , Satoshi Tamura  3
Affiliations

Feature Extraction Methods Proposed for Speech Recognition Are Effective on Road Condition Monitoring Using Smartphone Inertial Sensors

Frederico Soares Cabral et al. Sensors (Basel). .

Abstract

The objective of our project is to develop an automatic survey system for road condition monitoring using smartphone devices. One of the main tasks of our project is the classification of paved and unpaved roads. Assuming recordings will be archived by using various types of vehicle suspension system and speeds in practice, hence, we use the multiple sensors found in smartphones and state-of-the-art machine learning techniques for signal processing. Despite usually not being paid much attention, the results of the classification are dependent on the feature extraction step. Therefore, we have to carefully choose not only the classification method but also the feature extraction method and their parameters. Simple statistics-based features are most commonly used to extract road surface information from acceleration data. In this study, we evaluated the mel-frequency cepstral coefficient (MFCC) and perceptual linear prediction coefficients (PLP) as a feature extraction step to improve the accuracy for paved and unpaved road classification. Although both MFCC and PLP have been developed in the human speech recognition field, we found that modified MFCC and PLP can be used to improve the commonly used statistical method.

Keywords: deep neural network; feature extraction; paved and unpaved classification; road condition monitoring; signal processing; smartphone inertial sensors.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflicts of interest.

Figures

Figure 1
Figure 1
Overall system architecture for paved and unpaved road classification.
Figure 2
Figure 2
The map visualization process.
Figure 3
Figure 3
An architecture for statistical feature extraction.
Figure 4
Figure 4
Comparison of inertial sensor and axis.
Figure 5
Figure 5
Comparison of statistical feature extraction.
Figure 6
Figure 6
The general Mel Frequency Cepstral Coefficient (MFCC) computation steps.
Figure 7
Figure 7
Accuracy score based on pre-emphasis.
Figure 8
Figure 8
A comparison when removing the energy band logarithm.
Figure 9
Figure 9
A comparison of the number of MFCC coefficients.
Figure 10
Figure 10
A comparison when removing the energy framed signals.
Figure 11
Figure 11
Accuracy score based on MFCC with deltas coefficient.
Figure 12
Figure 12
A proposed step for MFCC feature extraction.
Figure 13
Figure 13
The general Perceptual Linear Predictive (PLP) computation steps.
Figure 14
Figure 14
A comparison when removing the equalization in PLP computation.
Figure 15
Figure 15
A comparison when removing the compression in PLP computation.
Figure 16
Figure 16
A comparison of the added RASTA filter.
Figure 17
Figure 17
A comparison of the number of PLP coefficients.
Figure 18
Figure 18
Accuracy score based on PLP with deltas coefficient.
Figure 19
Figure 19
A proposed step for PLP feature extraction.
Figure 20
Figure 20
Result of maps visualization: (a) paved roads and (b) unpaved roads.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Timor-Leste Strategic Plan 2011–2030. [(accessed on 11 December 2018)]; Available online: http://timor-leste.gov.tl/wp-content/uploads/2011/07/Timor-Leste-Strateg....
    1. Building a Sustainable Rural Road Network in Timor-Leste. [(accessed on 11 December 2018)]; Available online: https://dfat.gov.au/about-us/publications/Documents/roads-for-developmen....
    1. Roads for Development (R4D) [(accessed on 11 December 2018)]; Available online: https://dfat.gov.au/about-us/publications/Documents/roads-for-developmen....
    1. Beyond the Surface. [(accessed on 11 December 2018)]; Available online: http://www.roadscanners.com.
    1. The Statistics Portal. [(accessed on 11 December 2018)]; Available online: https://www.statista.com.

MeSH terms