Next Article in Journal
Dual-Band, Dual-Output Power Amplifier Using Simplified Three-Port, Frequency-Dividing Matching Network
Next Article in Special Issue
Scene Classification, Data Cleaning, and Comment Summarization for Large-Scale Location Databases
Previous Article in Journal
Persistent Postural-Perceptual Dizziness Interventions—An Embodied Insight on the Use Virtual Reality for Technologists
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Electronics
Volume 11
Issue 1
10.3390/electronics11010143
Review Report
electronics-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Facial Skincare Products’ Recommendation with Computer Vision Technologies

Electronics 2022, 11(1), 143; https://doi.org/10.3390/electronics11010143
by Ting-Yu Lin 1, Hung-Tse Chan 2, Chih-Hsien Hsia 2,* and Chin-Feng Lai 1,*
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Electronics 2022, 11(1), 143; https://doi.org/10.3390/electronics11010143
Submission received: 21 October 2021 / Revised: 28 December 2021 / Accepted: 29 December 2021 / Published: 3 January 2022
(This article belongs to the Special Issue Applied Deep Learning and Multimedia Electronics)

Round 1

Reviewer 1 Report

This paper propose an application system which uses Histogram equalization, Gabor wavelet filter, SVM, and DeepLab-v3+ model to determine the skin type and detect acne. Based on the detection results, proper product is selected from MySQL database and recommended to the user. Payment system is also taken into consideration. The overall representation is good. System flowchart, experimental results, and writing are well done. I have no further suggestion.  

Author Response

Thank you.

Reviewer 2 Report

This work proposes a new business system for facial skincare product recommendations.  The comments of the manuscript are listed as follows:

Would you please double-check the formulas in the manuscript and make sure that all the mathematical symbols are explained in this manuscript? 

The title of this manuscript is not suitable for the content. Facial skincare products recommendation should be the focus on the recommendation algorithm. However, this manuscript is more like a realization of a skincare product vending machine. 

How many training and testing samples are used in the skin type classification experiments? The authors should give more detail of the dataset.

Table 3 lists some of the skincare products. How does the author classify these skincare products? Is it based on prior knowledge?

Figure 4 is the flowchart of the skincare products recommendation system.  The authors apply the binary-classification SVM twice to get the skin type classification results (Oily, Neutral and Dry). Why not consider some multi-classification methods instead?

In the introduction section, the authors introduce the impact of acne on the skin. However, there is no experiment on acne identification in the experimental section. The authors should give the acne detection results （also mentioned in subsection 2.2.3. ）

How to recommend a suitable skincare product is what the manuscript concerns most. However, the recommending process is only related to the skin type and has nothing to do with the severity of acne. The authors need to re-adjust the content and logical structure of the manuscript.

Author Response

Reply letter as attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

This manuscript has been improved.

Back to TopTop