Review Reports

Round 1

Reviewer 1 Report

The paper tries to address CBIR problem by fusing features extracted by two CNNs, namely, ResNet50 and VGG16.

The paper is well-written and well-organized. However, I have some concerns:

1. Please clarify the role of the K-NN more clearly.

2. For each datasets that you have used, please tell us what is the retrieval radius? How many similar images have you retrieved?

3. Please proved some examples of some samples irrelevant to your query.

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors have proposed an effective technique for Content-Based Image Retrieval using Feature Fusion. They introduce a novel hybrid deep learning and machine learning-based CBIR system that uses a transfer learning technique and is implemented using two pre-trained deep learning models: ResNet50, VGG16, and KNN. They have performed extensive experimentation to show the effectiveness of the proposed system. However, I suggest addressing the following comments to improve the quality.

1.       Discuss the shortcomings and challenges of the existing systems in the introduction section.

2.       Add some more recent works of 2022 and 2023 in related work.

3.       Figure 1 is not readable, therefore improve its text and dpi.

4.       Also, improve Figure 2 as well. The text is too blurry to view.

5.  I suggest mentioning the layers architecture in tabular form of employed DL methods.

6.     I am confused about using only one metric for performance evaluation. Please explain why you’re not using any other metric for comparison.

7.       Improve the overall grammatical structure of the manuscript.

Comments for author File: Comments.docx

Author Response

Please see the attachment.

Author Response File: Author Response.docx