Next Article in Journal
Effectiveness of Virtual Reality-Based Early Postoperative Rehabilitation after Total Knee Arthroplasty: A Systematic Review with Meta-Analysis of Randomized Controlled Trials
Next Article in Special Issue
Research on an Improved Non-Destructive Detection Method for the Soluble Solids Content in Bunch-Harvested Grapes Based on Deep Learning and Hyperspectral Imaging
Previous Article in Journal
A Neural Topic Modeling Study Integrating SBERT and Data Augmentation
Previous Article in Special Issue
A Novel Generative Adversarial Network Model Based on GC-MS Analysis for the Classification of Taif Rose
 
 
Article
Peer-Review Record

Convolutional Neural Network-Based Automated System for Dog Tracking and Emotion Recognition in Video Surveillance

Appl. Sci. 2023, 13(7), 4596; https://doi.org/10.3390/app13074596
by Huan-Yu Chen 1, Chuen-Horng Lin 1,*, Jyun-Wei Lai 1 and Yung-Kuan Chan 2
Reviewer 1:
Reviewer 2:
Reviewer 3:
Appl. Sci. 2023, 13(7), 4596; https://doi.org/10.3390/app13074596
Submission received: 6 February 2023 / Revised: 26 March 2023 / Accepted: 3 April 2023 / Published: 5 April 2023
(This article belongs to the Special Issue Applications of Machine Learning in Agriculture)

Round 1

Reviewer 1 Report

The authors presented an article regarding a CNN-based automated system for dog tracking and emotion. This topic is very interesting, mainly for behavioural and ethological studies. However, the manuscript has several problems. The introduction is extensive and exhausting to read. Also, the Methodology is excessively detailed, without reason to do it. Finally, I did not find the Results and Discussion section, without any literature background and discussion. The structure is awkward and confusing. The citations are out of MDPI's standards. So, I won't be able to approve this manuscript with this presentation. 

Author Response

The referee’s comments are very much appreciated. According to the referee’s advice, the contents of this paper have been modified. To respond to the comments, please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper proposes a novel system that utilizes a multi-convolutional neural network (CNN) approach for the detection, tracking, and recognition of dogs' emotions in surveillance videos. The paper provides a clear and concise description of the proposed system's components and methods, including the use of YOLOv3 for dog detection and a deep association metric model (DeepDogTrack) for real-time tracking.

 

One notable contribution of this paper is the categorization of dogs' emotional behaviors into three distinct types, which were identified based on the input of veterinary experts and dog breeders. This categorization could potentially lead to a better understanding of dog behavior and assist in the development of more effective training and handling strategies.

 

Overall, this paper presents an interesting and innovative approach to detecting, tracking, and recognizing the emotions of dogs in surveillance videos. However, it would be helpful if the paper included more details about the dataset used for training and testing the proposed system and provided more information about the accuracy and performance of the system in recognizing dogs' emotions in various real-world settings.

Author Response

The referee’s comments are very much appreciated. According to the referee’s advice, the contents of this paper have been modified. To respond to the comments, please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

1. In Figure 10, the content " Linearly interpolation" written on a box is having spelling mistake.

2. Why the proposed model uses a Mask R-CNN model to remove backgrounds from the image set before the dog tracking and emotion recognition are processed by DeepDogTrack and the LDFDMN model, respectively? Is it doing any segmentation or removal of background?

3. Kindly list the emotion type used for classification.( it is sated in Abstract but make a Table representation)

4. What are features are identified for a dog?

Author Response

The referee’s comments are very much appreciated. According to the referee’s advice, the contents of this paper have been modified. To respond to the comments, please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thank you to following my suggestions. 

Back to TopTop