Next Article in Journal
On-Channel Integrated Optofluidic Pressure Sensor with Optically Boosted Sensitivity
Previous Article in Journal
Zinc/Aluminium–Quinclorac Layered Nanocomposite Modified Multi-Walled Carbon Nanotube Paste Electrode for Electrochemical Determination of Bisphenol A
Open AccessArticle
Peer-Review Record

Driver Drowsiness Detection Based on Steering Wheel Data Applying Adaptive Neuro-Fuzzy Feature Selection

Sensors 2019, 19(4), 943; https://doi.org/10.3390/s19040943
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Sensors 2019, 19(4), 943; https://doi.org/10.3390/s19040943
Received: 15 January 2019 / Revised: 11 February 2019 / Accepted: 18 February 2019 / Published: 22 February 2019
(This article belongs to the Section Intelligent Sensors)

Round 1

Reviewer 1 Report

This paper presents interesting and important research on driver drowsiness detection based on data driven methods. The contributions of this work should be summarized in separate items to highlight the novelty of this work. The following questions and comments are expected to be addressed in the revision. - Related literature are recommended to be discussed and compared: Recursive total principle component regression based fault detection and its application to vehicular cyber-physical systems (vehicular CPS safety control framework); A Data-Driven Fuzzy Information Granulation Approach for Freight Volume Forecasting (PSO based algorithm); Recent results on key performance indicator oriented fault detection using the DB-KIT toolbox (abnormality detection); Recent advances in key-performance-indicator oriented prognosis and diagnosis with a Matlab toolbox: DB-KIT (abnormality detection). - How can Fig. 2 show the performance of road curve removal? The compared signals are before and after removal. Indeed, the average steering angle is close to zero deg. But how is the performance evaluated? - What is 'random matrices' in Table 3? - Please increase font size in FIg. 6. It is unreadable in the printed version. - As shown in Table 4, FP and FN still has notable amounts. Is the performance shown in Table 4 applicable to practical driver drowsiness detection purpose?

Author Response

Dear reviewer,

Many thanks for your insightful comments on our manuscript. All of them have been considered to improve the manuscript and responded point by point in the attached file.

 

Best regards,

 

Authors of the manuscript 


Author Response File: Author Response.pdf

Reviewer 2 Report

Section 2.1, the average value of every three seconds is subtracted from the raw signal. You have to specify the corresponding equation to perform this and explain why three seconds and no other value.
Section 2.2, why you use three seconds and one and a half second for the signal windows?
Section 2.3, the explanation of xi and yi are not clear, furthermore, the sentence where they are described is not finalized.
Figure 7 and 8, number on labels must be improved, they are difficult to read.
Section 4, authors said that results are compared with the results of [23], which is a different subject paper, it is about the method, not a experiment of driving. Furthermore, in any part of the paper appear a comparison with other method, only in table 4 their results and in Figure 6 the classification with the different methods.
To sum up, the paper is well organized and could be interesting if it compares the results obtained with results of other studies in drive drowsiness.

Author Response

Dear reviewer,

Many thanks for your insightful comments on our manuscript. All of them have been considered to improve the manuscript and responded point by point in the attached file.

 

Best regards,

 

Authors of the manuscript 


Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The paper can be accepted in its present form.

Reviewer 2 Report

Authors have performed all the review suggestions


Back to TopTop