Smart Disease Detection System for Citrus Fruits Using Deep Learning with Edge Computing

Round 1

Reviewer 1 Report

In this study  examines and explores four different diseases of citrus fruits using CNN deep learning 8 models to be adopted as edge computing solution. 

These are the commnets to be address 

1. The novelty of the study missing

2. Abbreviations must be written before the references section.

3. Abstract to be rewritten with data set description. 

4. In CNN, What is the size of each image, and pooling value, and stride value not described in detail.

5.Implication of  the study must be develop one paragraph in the study

6. The following references to be consider, 

Statistical growth prediction analysis of rice crop with pixel-based mapping technique, Monika Mangla, Vaishali Mehta, Sachi Nandan Mohanty, Nonita Sharma, Anusha Preetham,(2022), International Journal Artificial Intelligence and Soft Computing (Inderscience) Vol.7, Issue 3.208-227, ISSN:1755-4969

Author Response

In this study  examines and explores four different diseases of citrus fruits using CNN deep learning 8 models to be adopted as edge computing solution. 

 These are the commnets to be address 

1. The novelty of the study missing

Reply:  The proposed model utilizes the fusion of deep learning models CNN and LSTM with edge computing. The proposed model employs an enhanced feature-extraction mechanism, with a down-sampling approach, and then a feature-fusion subsystem to ensure significant recognition on edge computing devices with retaining citrus fruit disease detection accuracy.

2. Abbreviations must be written before the references section.

Reply: Abbreviations are added at the end of the article just before the bibliography.

3. Abstract to be rewritten with data set description. 

Reply: Abstract is revised.

4. In CNN, What is the size of each image, and pooling value, and stride value not described in detail.

Reply: Explained in the last paragraph of the the section 3.2.

5.Implication of  the study must be develop one paragraph in the study

Reply: Implication of the study is added just before the conclusion

6. The following references to be consider, 

Statistical growth prediction analysis of rice crop with pixel-based mapping technique, Monika Mangla, Vaishali Mehta, Sachi Nandan Mohanty, Nonita Sharma, Anusha Preetham,(2022), International Journal Artificial Intelligence and Soft Computing (Inderscience) Vol.7, Issue 3.208-227, ISSN:1755-4969

 Reply: Added as the reference number 1.

Reviewer 2 Report

1. The abstract is not clear how the proposed method is constructed.

2. The introduction does not show the motivation clearly for the proposed method. The authors could provide a more detailed explanation of the background of this study, particularly why the research problem is important. The introduction should clearly explain the fundamental limitations of earlier work that are relevant to the current paper. Authors must provide a formal definition of the problem. Before describing the specifics, the authors should provide an overview of their solution. An innovative solution is offered; however, it is important to explain the design considerations properly. It is important to clearly explain what is new and what is not in the proposed solution. If some elements are identical, they should be attributed properly, and discrepancies should be indicated. The solution is stated; however, further examples are needed. Some experimentation should be included to demonstrate that the proposed solution can be employed in real-world scenarios.

3. The related work is improper form, does not show the pros and cons of the previous work, with short survey on the field.

4. Some latest papers from top conferences and journals of 2020-2022 need to be referenced and compared.

5. Many method improvements of the authors are the integration of a large number of existing methods. Are they simple system integration or have new features? How can the authors prove the effectiveness of the integration methods?

6. The computational complexity or execution time of the proposed scheme should be given.

7.: rewrite the conclusion and consider the following comments:

-  Highlight your analysis and reflect only the important points for the whole paper.

-  Mention the benefits.

-  Mention the implication at the last of this section.

Author Response

1. The abstract is not clear how the proposed method is constructed.

Reply: The abstract is re-written and organized.

2. The introduction does not show the motivation clearly for the proposed method. The authors could provide a more detailed explanation of the background of this study, particularly why the research problem is important. The introduction should clearly explain the fundamental limitations of earlier work that are relevant to the current paper. Authors must provide a formal definition of the problem. Before describing the specifics, the authors should provide an overview of their solution. An innovative solution is offered; however, it is important to explain the design considerations properly. It is important to clearly explain what is new and what is not in the proposed solution. If some elements are identical, they should be attributed properly, and discrepancies should be indicated. The solution is stated; however, further examples are needed. Some experimentation should be included to demonstrate that the proposed solution can be employed in real-world scenarios.

Introduction is re-Witten by considering all the suggestion.

3. The related work is improper form, does not show the pros and cons of the previous work, with short survey on the field.

Reply:  Related work is re-constructed as per your suggestions

4. Some latest papers from top conferences and journals of 2020-2022 need to be referenced and compared.

Reply: Comparison table is added in the introduction section.

4. Many method improvements of the authors are the integration of a large number of existing methods. Are they simple system integration or have new features? How can the authors prove the effectiveness of the integration methods?

Reply: This research presents efficient fruit disease identification and classification model using deep learning-based CNN and LSTM techniques with edge computing.  The key contribution of the research includes.

Citrus fruits dataset with various diseases is gathered from various online sources and submitted towards the edge computing devices for categorization. Edge computing devices which have a limited number of resources are unable to run complex DNN-based applications.  

This research utilizes a light processing-based CNN-LSTM model with edge computing for Citrus fruit disease classification. We have also quantized and pruned the edge network. So that, the proposed model can able to run on limited resources.

The proposed model utilizes three phases. Phase 1 includes data collection, and data pre-processing, and phase 2 includes re-scaling of images, data augmentation, and image segmentation by K-mean clustering and canny edge detection, also applying CNN-LSTM for model training. Phase 3 is the final process which includes testing of the images and classification. In this phase, experimental analysis is performed using Magnitude-Based Pruning, and in the second step Magnitude-Based Pruning with Post Quantization.

6. The computational complexity or execution time of the proposed scheme should be given.

Reply: In our research work we are mainly focusing on the size of the model and we have evaluated various other parameters like precision, recall, f-score and accuracy. In future work we will definitely compute the time complexity of the proposed work.

7.: rewrite the conclusion and consider the following comments:

-  Highlight your analysis and reflect only the important points for the whole paper.

-  Mention the benefits.

-  Mention the implication at the last of this section.

Reply: Conclusion is re-written by considering all the suggestion of the reviewers

Round 2

Reviewer 1 Report

These are the comments to be addressed

In Table 6, the % symbol repeats many places, It may write like Accuracy (%).

Refer to the following manuscript for the more technical aspect

➢        Agricultural Recommendation System for Crops Using Different Machine Learning Regression Methods, Mamta Ganayak, Sachi Nandan Mohanty, Alok Kumar Jagadev, International Journal of Agricultural and Environmental Information Systems, (2021) Vol. 12, No.1, 1-20, ISSN: 1947-3192, DoI: 10.4018/IJAEIS

The implication of the study must be developed one paragraph before the reference section. 

Author Response

please check the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript can be accepted

Author Response

Please check the attached file.

Author Response File: Author Response.docx