Round 1

Reviewer 1 Report

Overall Decision: Major revision

 

This manuscript presents a research study focused on estimating maize crop productivity using remote sensing imagery and machine learning algorithms. The study aims to develop a predictive model that accurately estimates maize yield based on spectral data acquired from satellite imagery. The researchers utilized various vegetation indices and machine learning techniques to predict maize yield across different phenological stages of the crop.

 

General Considerations:

 

1. The paper covers important topics in precision agriculture, specifically the utilization of remote sensing data and machine learning algorithms to estimate maize yields. The authors employed multispectral images and remote sensing indices to monitor changes during various stages of maize growth, facilitating yield prediction. They also utilized Support Vector Machine and other machine learning methods to construct predictive models, testing them in real-world scenarios.

 

2. The paper underscores the potential of remote sensing and machine learning in modern agriculture, enhancing the efficiency and predictability of crop production. While I cannot delve into depth, overall, this article offers valuable insights for both scientific research and practical applications in the agricultural field.

 

3. The document contains 52 references, with 15 (28.85%) published in the last 5 years, 28 (53.85%) in the last 5-10 years, and 9 (17.31%) older than 10 years.

 

4. Technique Concerns: The paper mentions the use of multispectral images and machine learning algorithms for prediction but lacks an explicit description of feature selection and extraction. Detailing these processes is crucial to the model's success.

 

Title, Abstract, and Keywords:

5. The main objectives, methods, and results of the study could be more clearly emphasized in the abstract. Ensure that the summary concisely conveys key information, enabling readers to quickly grasp the study's main points.

 

Chapter 1: Introduction:

6. Enhance the introduction with a more comprehensive literature review and background, highlighting the study's significance and objectives. Describe the application and potential of precision agriculture and remote sensing techniques in crop yield forecasting more clearly (Rachis detection and three-dimensional localization of cut off point for vision-based banana robot; Computers and Electronics in Agriculture. Detection and counting of banana bunches by integrating deep learning and classic image-processing algorithms; Computers and Electronics in Agriculture. Optimization strategies of fruit detection to overcome the challenge of unstructured background in field orchard environment: a review; Precision Agriculture. Fruit detection and positioning technology for a Camellia oleifera C. Abel orchard based on improved YOLOv4-tiny model and binocular stereo vision; Expert Systems with Applications.).

 

Chapter 2: The Method:

7. In the methods section, provide a detailed description of data collection, processing, and pre-processing steps, including handling missing data and performing data tagging and cleaning. Elaborate on feature selection and extraction methods and explain the rationale behind choosing specific machine learning algorithms.

 

Chapter 3: Experiments and Results:

8. Alongside charts and tables, clarify the physical meaning of key results to enhance reader understanding. Perform an in-depth analysis of model performance at different time points or on various land plots to convey model stability and reliability.

 

Chapter 4: Conclusions:

9. Expand the discussion section to explain the implications of results and compare them with existing literature. Explore the model's applicability in diverse geographical and agricultural contexts, discussing opportunities for further enhancement and optimization.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Defining the ideal phenological stage for estimation of corn 2 yield from multispectral images

 

Summary:

This study explores the use of image-based spectral models to estimate maize yield, specifically focusing on the vegetative and reproductive phenological phases when the crop undergoes changes due to various stresses. Spectral models are valuable tools for crop management, but determining the optimal timing to obtain these images remains challenging. The research hypothesizes the possibility of estimating corn productivity using multispectral images, considering the best time to detect differences caused by different phenological stages. Multispectral bands and vegetation indices from the Planet satellite serve as predictor variables for the models. The reproductive phenological phase was found to be optimal for spectral models, achieving the best root mean square error percentage (9.17%) and second-best mean absolute percentage error (7.07%) for productivity estimates. This study demonstrates the feasibility of predicting corn yield before harvest using multispectral satellite images.

 

Paper Comments:

 

Abstract:

·         Please clearly mention what you have done in this paper.

Introduction:

·         Write down something about your work and algorithm use in your paper and followed by paper organization.

·                     Improve the  introcution with some remote sensing paper like : https://doi.org/10.3390/futuretransp3010012

Materials and Methods:

·         Legends color in figure 2 need to be highlight

·         Write something about Pearson model you mentioned in subsection 2.9 and how it is more efficient than other models if possible test his capability and shows it results.

·         Please mentioned flow chart or algorithm of given model you used.

Conclusion:

·         Highlight background of your work and key results that what you achieved and future work.

Result section:

·         Is not well organized and its make confusion.

Other Comments:

 

·         Write 2 to 3 more relevant keywords.

·         Remove empty spaces.

·         Improve quality of figure 3 and figure 5.

·         Figure 7 are not visible.

·         How we should know about figure 8 when printed your paper in black and white that which legend belong to graph inside figure and it legends are too small.

·         Same for figure 9.

·         You didn’t conclude your paper well.

 

Final Recommendation:

·         Your paper is technically sound but still need modification to look better. ·     Remove all empty spaces and improve quality of paper.

 

· 

Defining the ideal phenological stage for estimation of corn 2 yield from multispectral images

 

Summary:

This study explores the use of image-based spectral models to estimate maize yield, specifically focusing on the vegetative and reproductive phenological phases when the crop undergoes changes due to various stresses. Spectral models are valuable tools for crop management, but determining the optimal timing to obtain these images remains challenging. The research hypothesizes the possibility of estimating corn productivity using multispectral images, considering the best time to detect differences caused by different phenological stages. Multispectral bands and vegetation indices from the Planet satellite serve as predictor variables for the models. The reproductive phenological phase was found to be optimal for spectral models, achieving the best root mean square error percentage (9.17%) and second-best mean absolute percentage error (7.07%) for productivity estimates. This study demonstrates the feasibility of predicting corn yield before harvest using multispectral satellite images.

 

Paper Comments:

 

Abstract:

·         Please clearly mention what you have done in this paper.

Introduction:

·         Write down something about your work and algorithm use in your paper and followed by paper organization.

·                     Improve the  introcution with some remote sensing paper like : https://doi.org/10.3390/futuretransp3010012

Materials and Methods:

·         Legends color in figure 2 need to be highlight

·         Write something about Pearson model you mentioned in subsection 2.9 and how it is more efficient than other models if possible test his capability and shows it results.

·         Please mentioned flow chart or algorithm of given model you used.

Conclusion:

·         Highlight background of your work and key results that what you achieved and future work.

Result section:

·         Is not well organized and its make confusion.

Other Comments:

 

·         Write 2 to 3 more relevant keywords.

·         Remove empty spaces.

·         Improve quality of figure 3 and figure 5.

·         Figure 7 are not visible.

·         How we should know about figure 8 when printed your paper in black and white that which legend belong to graph inside figure and it legends are too small.

·         Same for figure 9.

·         You didn’t conclude your paper well.

 

Final Recommendation:

·         Your paper is technically sound but still need modification to look better. ·     Remove all empty spaces and improve quality of paper.

 

· 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I did not see any point-by-point response from the authors. The authors mixed the replies from reviewer 1 and reviewer 2 together and avoided addressing my comments. This is very strange and absurd.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

 

1.survay on pervious paper are less please extend it with some new paper  

2.about reserch contribution i ask perviously to mention with numbering order 

3. beside please clearly mention the research gap

4.the table 1 include some formula which at least the short defenition of paramter should be table caption 

5. the conclotion part should extend and  have the future work plan.i think the bullet form is not propoer

6. in fig 6 the E.f is not specified in caption 

 

 

1.survay on pervious paper are less please extend it with some new paper  

2.about reserch contribution i ask perviously to mention with numbering order 

3. beside please clearly mention the research gap

4.the table 1 include some formula which at least the short defenition of paramter should be table caption 

5. the conclotion part should extend and  have the future work plan.i think the bullet form is not propoer

6. in fig 6 the E.f is not specified in caption 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

accept

Author Response

Thanks for the corrections.

Reviewer 2 Report

Authors are responds all concern but i think this better to have one more line about Parmeter  for table 1 

Authors are responds all concern but i think this better to have one more line about Parmeter  for table 1 

Author Response

Please see the attachment.

Author Response File: Author Response.docx