Tropical Wood Species Recognition: A Dataset of Macroscopic Images
, Holman Raul Gaitan Mesa 2 and Rubiel Vargas-Cañas 1,*
Round 1
Reviewer 1 Report
1. There are some grammatical mistakes in the manuscript, please consider correcting them.The language of the whole paper needs to be improved.
2.The format of the references should be improved.
3.Pictures of the same tree species in different locations are also suggested to be added in the part of Data Description.
4. If possible, some RGB data of wood could be added in the manuscript.
5. The author suggests that the photos should be taken within one ring of a tree in the further study.
Author Response
We would like to thank all reviewers for the time dedicated to the revision of our manuscript, and all their comments and suggestions. They have helped us to improve in great measure our document. We appreciate and present our gratitude to the reviewers for the good reception of our manuscript and all of their positive suggestions.
Point 1. There are some grammatical mistakes in the manuscript, please consider correcting them. The language of the whole paper needs to be improved.
Response 1: We have double-checked the manuscript in order to correct typos and grammatical errors.
Point 2. The format of the references should be improved.
Response 2: References have been corrected according to the MDPI guidelines.
Point 3. Pictures of the same tree species in different locations are also suggested to be added in the part of Data Description.
Response 3: Section Data Description has been updated to explain that pictures are from different parts of the log and that the database is being populated with pictures of the same tree species coming from different locations, i.e., amazon rain forest, pacific rain forest.
Point 4. If possible, some RGB data of wood could be added in the manuscript.
Response 4: we have included in the Data Description section a figure with representative RGB picture of each one of the species.
Point 5. The author suggests that the photos should be taken within one ring of a tree in the further study.
Response 5: We have updated the paper to clarify that images are taken from the tree cross-section due to it displays more of the features of interest.
Reviewer 2 Report
The proposed manuscript is well written and provides important information about the dataset of macroscopic photos of selected trees. From my point of view, the dataset is important, especially for its ability to provide training data for intelligent systems to automatically determine selected species usable in forest/nature protection. I hope the dataset will continue to expand. I like the manuscript. Great job!
I found some typos in the text - rows 43, 56 and 109.
Regards,
Author Response
We would like to thank all reviewers for the time dedicated to the revision of our manuscript, and for all their comments and suggestions. They have helped us to improve in great measure our document. We appreciate and present our gratitude to the reviewers for the good reception of our manuscript and all of their positive suggestions.
Point 1. The proposed manuscript is well written and provides important information about the dataset of macroscopic photos of selected trees. From my point of view, the dataset is important, especially for its ability to provide training data for intelligent systems to automatically determine selected species usable in forest/nature protection. I hope the dataset will continue to expand. I like the manuscript. Great job!
Response 1: We highly appreciate the reviewer’s comment which encourages us to carry out our research. Moreover, we will continue expanding the dataset with images from other Colombian locations.
Point 2. I found some typos in the text - rows 43, 56 and 109.
Response 2: We have double-checked the manuscript in order to correct typos and grammatical errors.
Reviewer 3 Report
Incorrect formatting of the text from the first paragraph.
Line 56: superscript in the unit.
There is no precise aim of the work. Please write what the purpose of the work was.
Line 67, table 1: Why were only common Colombian names chosen and no English names added?
Line 122: moisture content, not humidity.
Line 132: parenchyma, not parenquima.
Why were these 11 wood species chosen? What were the criteria for selecting species?
The authors have prepared a very extensive library of photos of selected types of wood. However, the images should be chosen appropriately to show more distinctive features. Most pictures show almost identical areas within one species that strongly differ in colour, which may suggest variable lighting. Moreover, the plane has not been appropriately prepared. You can see the unevenness created during the cutting process. Images taken with a microscope, even a stereoscopic one, would be more advantageous.
Author Response
We would like to thank all reviewers for the time dedicated to the revision of our manuscript, and for all their comments and suggestions. They have helped us to improve in great measure our document. We appreciate and present our gratitude to the reviewers for the good reception of our manuscript and all of their positive suggestions.
Point 1. Incorrect formatting of the text from the first paragraph.
Point 2. Line 56: superscript in the unit.
Responses 1 and 2: We have double-checked the manuscript in order to correct typos and grammatical errors.
Point 3. There is no precise aim of the work. Please write what the purpose of the work was.
Response 3: The aim of the work is to provide a set of digital macroscopic images of eleven tropical forest species, which can be used as support at checkpoints, to carry out studies and research based on macroscopic analysis of cross-sectional images of tree species such as dendrology, forestry, as well as algorithms of artificial intelligence. The abstract of the paper, as well as the introduction, were updated to clarify it.
Point 4. Line 67, table 1: Why were only common Colombian names chosen and no English names added?
Response 4: There are three key names, which are useful to recognize the species within a region (Colombia) and around the world, the scientific name, the common name and the trading name. However, the common and trading names have several variations within the countries. In this paper, the three key names were included in the data description.
Point 5. Line 122: moisture content, not humidity.
Point 6. Line 132: parenchyma, not parenquima.
Responses 5 and 6: We have double-checked the manuscript in order to correct typos and grammatical errors.
Point 7. Why were these 11 wood species chosen? What were the criteria for selecting species?
Response 7: The selection criteria were established by the experts of the environmental division of the state public agency, based on access to the species, species more commercialized or rated as high risk for illegal extraction in the country. This was explained in the method.
Point 8. The authors have prepared a very extensive library of photos of selected types of wood. However, the images should be chosen appropriately to show more distinctive features. Most pictures show almost identical areas within one species that strongly differ in colour, which may suggest variable lighting. Moreover, the plane has not been appropriately prepared. You can see the unevenness created during the cutting process. Images taken with a microscope, even a stereoscopic one, would be more advantageous.
Response 8: We agree with the reviewer that plane preparation and microscopic image acquisition would be more advantageous. Nevertheless, our aim is to provide the environmental authorities with technological tools to support decision-making at checkpoints where illegal timber trade should take place, therefore, machine learning algorithms have to be trained with images as close as possible to the real environment, i.e., rural checkpoints or wood shipments.
Round 2
Reviewer 1 Report
Accept in present form
Reviewer 3 Report
After introducing the corrections, the text is consistent and very interesting. He makes a great contribution to the science of wood.
