GrapesNet: Indian RGB & RGB-D Vineyard Image Datasets

Tracking #: 744-1724

Authors:

	Name	ORCID
	Dhanashree Barbole	https://orcid.org/0000-0002-3969-1043
	Parul Jadhav	https://orcid.org/0000-0002-5083-2900

Responsible editor:

Tobias Kuhn

Submission Type:

Resource Paper

Abstract:

In most of the countries, grapes are considered as a cash crop. Currently huge research is going on in development of automated grape harvesting systems. Speedy and reliable grape bunch detection is prime need for various deep learning based automated systems which deals with object detection and object segmentation tasks. But currently very few datasets are available on grape bunches in vineyard, because of which there is restriction to the research in this area. In comparison to the vineyard in outside countries, Indian vineyard structure is more complex, so it becomes hard to work in real-time. To overcome these problems and to make vineyard dataset for suitable for Indian vineyard scenarios, this paper proposed four different datasets on grape bunches in vineyard. For creating all datasets in GrapesNet, natural environmental conditions have been considered. GrapesNet includes total 11000+ images of grape bunches. Proposed datasets can be used for prime tasks like grape bunch detection, grape bunch segmentation, and grape bunch weight estimation etc. of future generation automated vineyard harvesting technologies.

Manuscript:

ds-paper-744.docx

Data repository URLs:

Direct URL to data: https://data.mendeley.com/datasets/mhzmzd5cwx/1

Date of Submission:

Monday, January 23, 2023

Date of Decision:

Friday, July 21, 2023

Nanopublication URLs:

Decision:

Reject

Solicited Reviews:

Review #1 submitted on 07/Feb/2023

By Shradha Verma ORCID logo

https://orcid.org/0000-0002-7053-1654

Review Details

Reviewer has chosen not to be Anonymous

Overall Impression: Good
Suggested Decision: Undecided
Technical Quality of the paper: Average
Presentation: Average
Reviewer`s confidence: Medium
Significance: Moderate significance
Background: Reasonable
Novelty: Unable to judge
Data availability: All used and produced data (if any) are FAIR and openly available in established data repositories
Length of the manuscript: The length of this manuscript is about right

Summary of paper in a few sentences:

The paper presents four datasets constituting grape bunches captured in a vineyard, along with the details of equipment used, research methodology, augmentation techniques, image capturing modes (RGB, RGB-D), and dataset description.

Reasons to accept:

Data collection is extremely crucial for developing efficient prediction models, requiring huge datasets. The dataset captured by the authors is diverse and might be useful for deep learning-based applications.

Reasons to reject:

The language is grammatically incorrect in several places.
The time 10:00 - 17:00 covers late morning, afternoon, and evening hours. The authors emphasize on sunlight, hence, the image count could also indicate the same.
The RGB-D dataset in folders 'Dataset-3' and 'Dataset 4', has not been explained in the manuscript.

Nanopublication comments:

Further comments:

Review #2 submitted on 11/Apr/2023

Review Details

Reviewer has chosen to be Anonymous

Overall Impression: Average
Suggested Decision: Undecided
Technical Quality of the paper: Average
Presentation: Weak
Reviewer`s confidence: High
Significance: Moderate significance
Background: Reasonable
Novelty: Unable to judge
Data availability: All used and produced data (if any) are FAIR and openly available in established data repositories
Length of the manuscript: The authors need to elaborate more on certain aspects and the manuscript should therefore be extended (if the general length limit is already reached, I urge the editor to allow for an exception)

Summary of paper in a few sentences:

The paper proposes the GrapesNet dataset, which could be utilized for the development of automatic grape bunch detection and segmentation systems. Four sub-datasets are created; one grape bunch per image, multiple grape bunches per image, Dataset-3 with images taken under natural lighting conditions, and Dataset-4 that was created with artificial environment of the fixed environmental conditions and constant artificial light.

Reasons to accept:

Overall, I found the article to be a valuable contribution to the development of automatic grape bunch detection and segmentation systems, through the creation of the GrapesNet dataset.

Reasons to reject:

* Not justification for creating the one grape bunch per image and multiple grape bunches per image datasets separately
* It would also be helpful to see some evaluations, possibly initial results of a baseline model trained and evaluated on the dataset. The evaluation would help prove your dataset was designed and built well
* The organization of the contents of the article should be improved, and grammatical errors corrected to enhance readability

Nanopublication comments:

Further comments:

2 Comments

Meta-Review by Editor

Submitted by Tobias Kuhn on Tue, 04/11/2023 - 05:27

The reviewers agree that this is a potentially useful resource worthy of publication, but there are a few things that should be better explained and the presentation of the manuscript needs to be improved. I would therefore like to ask the authors to submit an improved version taking these points into account.

Tobias Kuhn (https://orcid.org/0000-0002-1267-0234)

Withdrawn by the authors

Submitted by Tobias Kuhn on Fri, 07/21/2023 - 02:25

Upon request by the authors, this submission was withdrawn and is thereby marked as Rejected.

Data Science