Experimental Dataset for Fiber Optic Specklegram Sensing Under Thermal Conditions and Use in a Deep Learning Interrogation Scheme

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

My understanding of the journal Data is that data-descriptor papers should keep to data descriptors and not have a section like section 4 which reports on an ML analysis. The ML analysis would be better peer reviewed in a stand-alone paper dedicated to the ML analysis.

Additionally, many basic descriptors of the experiment are missing from the manuscript, and their absence would lower the reproducibility and applicability of the submitted datasets.

1) the description of the multi-mode fiber lacks numerical aperture and cross-sectional refractive-index profile shape and parameters

2) the description of the CCD camera lacks bit depth, exposure/capture time, dynamic range, a measure of its noise

3) the description of the fiber lacks the total length of the fiber and the position of the temperature zone along the fiber

4) the description of the laser lacks the spectral bandwidth of the laser and the kind of laser

Author Response

REVIEWER 1:

My understanding of the journal Data is that data-descriptor papers should keep to data descriptors and not have a section like section 4 which reports on an ML analysis. The ML analysis would be better peer reviewed in a stand-alone paper dedicated to the ML analysis.

We fully understand the suggestion regarding the scope of data-descriptor papers. However, we believe that Section 4 offers critical insights by demonstrating an actual application of the dataset. Rather than presenting the ML analysis as an exhaustive study, our aim is to showcase the dataset’s practical utility—specifically, its potential for temperature sensing using fiber optic specklegram sensors. This applied example illustrates how the dataset can be leveraged in real-world scenarios, thereby enhancing its value to the community.

Additionally, many basic descriptors of the experiment are missing from the manuscript, and their absence would lower the reproducibility and applicability of the submitted datasets.

1) the description of the multi-mode fiber lacks numerical aperture and cross-sectional refractive-index profile shape and parameters

Thank you for your feedback. We appreciate your comment regarding the need to include additional details about the multi-mode fiber. In response to your suggestion, we have updated Section 2.2 (line 84) to provide a more comprehensive description of the fiber parameters. The updated information now includes:

• Fiber Refractive-Index Profile: Step-index
• Fiber Core Diameter: 62.5 μm
• Fiber Cladding Diameter: 125 μm
• Fiber Numerical Aperture (NA): 0.14

2) the description of the CCD camera lacks bit depth, exposure/capture time, dynamic range, a measure of its noise

Thank you for your valuable feedback. We have updated the description of the CCD camera in Method to include the requested details (line 115):

(CCD camera with 1280 x 1024 pixel resolution).The RGB camera used in this study features an 8-bit depth per channel. The exposure time was set to approximately 50 ms, optimized to maximize the dynamic range of around 48 dB while avoiding pixel saturation. Noise levels in the acquired images were found to be not significant, and ongoing computational validation, including data augmentation with random noise addition, confirms the robustness of the results.

 

 

3) the description of the fiber lacks the total length of the fiber and the position of the temperature zone along the fiber

We appreciate the reviewer’s observation regarding the description of the optical fiber. To address the noted omission, we clarify in the manuscript (line 102):

The dataset was generated using an automated system that consists of a multimode optical fiber and a temperature-controlled heating system with PID (Proportional-Integral-Derivative) control. This system regulates the temperature along a section of the optical fiber, ranging from 25 °C to 200 °C. The total length of the multimode optical fiber used in our experimental setup is 70 cm. It is essential for automating the acquisition of experimental specklegrams and allows for perturbation over the length of the sensing zone, which in this case is 2 cm, positioned 5 cm from the distal end of the fiber.

4) the description of the laser lacks the spectral bandwidth of the laser and the kind of laser

The description of the laser has been included in the manuscript (line 108):

A Helium-Neon (HeNe) laser at 633 nm with a power of less than 2.5 mW and a spectral bandwidth of 2 pm was used as the light source for the setup.

                                                               

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This article is very interesting because it shows an example of how to use speckle images in a sensor using AI. It also describes the entire process from obtaining the images, the process for training the neural network and testing the system as a temperature sensor. This gives the possibility to other researchers to use the methods developed in other research.

 

If the code developed in python could also be published, it would be a great help for other researchers.

 

Identify the laser source in the text because the images obtained depend on it

 

In Figure 3 (a) identify the optical components used to condition the laser light. Also the model of the optical connector and optical fiber used Figure 3(b) indicate the model and focal length of the lens used This information can be of great importance when reproducing the experimental scheme to develop a FSS

 

Figure 4

change the units in Axis X, Y: (ºC)

 

The representation of the data in Figure 4 could be improved if 2 graphs were used instead of one. One graph represents the Test and the Training and another graph represents the Test and the Validation

Author Response

REVIEWER 2:

This article is very interesting because it shows an example of how to use speckle images in a sensor using AI. It also describes the entire process from obtaining the images, the process for training the neural network and testing the system as a temperature sensor. This gives the possibility to other researchers to use the methods developed in other research.

We thank the reviewer for their valuable comments and appreciation of our work.

If the code developed in python could also be published, it would be a great help for other researchers.

We thank the reviewer for the valuable suggestion to publish the developed code. We are pleased to inform you that the code for automating the experimental setup and training the MNet-reg neural network has been included in the repository, accessible at the following link: https://osf.io/8nxvk/files/osfstorage, under the folder named "CODE." This ensures that the code is openly available for the research community to access, review, and utilize for further studies or replication of our work.

 Identify the laser source in the text because the images obtained depend on it

The description of the laser has been included in the manuscript (line 108):

A Helium-Neon (HeNe) laser at 633 nm with a power of less than 2.5 mW and a spectral bandwidth of 2 pm was used as the light source for the setup.

In Figure 3 (a) identify the optical components used to condition the laser light. Also the model of the optical connector and optical fiber used Figure 3(b) indicate the model and focal length of the lens used. This information can be of great importance when reproducing the experimental scheme to develop a FSS

Thank you for your valuable feedback. In response to your questions regarding Figure 3, we have made the following adjustments and clarifications:

Figure 3(b) has been updated to more accurately represent the experimental setup.

A detailed description of the setup has been added to the text for clarity (line 142).

In Figure 3(a), the optical components used to condition the laser light include a 633 nm helium-neon laser, a polarizer to ensure linear polarization, and a lens for focusing or collimating the light into the optical fiber during free-space propagation. The optical connector used is an FC/PC connector, and the optical fiber is a multimode fiber with a 62.5 μm core diameter, 125 μm cladding diameter, and a 2 cm sensing zone length. In Figure 3(b), the lens used in the image-forming system is a 20x Olympus microscope objective, which typically has a focal length of approximately 9–10 mm. This objective lens is responsible for forming the speckle pattern image captured by the camera.

Figure 3(b)

 

Figure 4. change the units in Axis X, Y: (ºC)

The units of the axis have been adjusted to ºC in Figures 4a, 4b (line 216)

 The representation of the data in Figure 4 could be improved if 2 graphs were used instead of one. One graph represents the Test and the Training, and another graph represents the Test and the Validation

In response to the reviewer’s request, we have separated the data into two figures: Figure 4(a) now displays the Training vs. Test results, while Figure 4(b) presents Validation vs. Test results. This reduces visual clutter and allows readers to see each comparison more distinctly.

                                                                4(a)                                                   4(b)

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have improved their description of the experimental system by adding many new data descriptors. Doing so has improved the reproducibility and applicability of the submitted dataset. As for Sections 1, 2, and 3, I rate their "significance of content," "quality of presentation," and "scientific soundness" as "high." However, I maintain my opinion from the initial review that section 4 is not appropriate, and should be spun off and submitted to be evaluated and stand on its own as an expanded paper.