Next Article in Journal
A Single-Sensor Approach to Quantify Gait in Patients with Hereditary Spastic Paraplegia
Previous Article in Journal
A Novel Optimized iBeacon Localization Algorithm Modeling
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sensors
Volume 23
Issue 14
10.3390/s23146562
Review Report
sensors-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

Alignment Integration Network for Salient Object Detection and Its Application for Optical Remote Sensing Images

Sensors 2023, 23(14), 6562; https://doi.org/10.3390/s23146562
by Xiaoning Zhang 1,2, Yi Yu 1,*, Yuqing Wang 1, Xiaolin Chen 1 and Chenglong Wang 1
Reviewer 1:
Yi-Jen Mon
Reviewer 2:
Cheng Peng
Sensors 2023, 23(14), 6562; https://doi.org/10.3390/s23146562
Submission received: 27 June 2023 / Revised: 18 July 2023 / Accepted: 19 July 2023 / Published: 20 July 2023
(This article belongs to the Section Intelligent Sensors)

Round 1

Reviewer 1 Report

1. AI is already a proper term for artificial intelligence, so it is better to change the AI in AINet to ALNet. All alignment integration in this article can be represented by AL abbreviation, because the focus is on alignment rather than artificial intelligence.

2. All OURS in the text are changed to ALNet.

3. Do you use tensorflow? Use model.save to access the .h5 file, and then list the complete network architecture diagram. How many neurons are needed for each layer? How many epochs are calculated? What is the learning rate? What optimizer is used?

4. Please write down the full name of FCN.

5. Why is there no need for dropout and dense layers? How to correct weighting in theory? How to generate the .hdf5 file generated by the experimental results?

Author Response

Please see the attachment. 

Author Response File: Author Response.pdf

Reviewer 2 Report

The paper introduces the Alignment Integration Network (AINet) to address the misalignment issue in multi-level convolutional feature combination. The key contributions, as highlighted, include the Strip Attention Module (SAM), the Boundary Enhancement Module (BEM), and an attention-weighted loss function. However, this paper could be improved considering the following points:
1.The introduction could benefit from being more compact. Avoid complex terminologies or in-depth details in this section, which could be saved for the method and results sections.
2. The authors should also provide a discussion on the time complexity and the cost of the proposed method.
3. What is the choice of hyperparameters β and λ in the final loss function? Were the parameters based on previous research or heuristic methods?

There are some typos and grammar issues that should carefully checked.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

All my concerned comments have been explained and revised.

Back to TopTop