Impact of Input Data on Intelligence Partitioning Decisions for IoT Smart Camera Nodes

Round 1

Reviewer 1 Report

The authors proposed an analysis about the influence of the input image size and its content within the conventional image processing systems affect the decision on where tasks should be implemented, with respect to node energy and latency.

Major Comments:

The study is very interesting, and the proposed results show that the "optimal partition” strongly depends on many factors: such as the size of the images, the amount of data, the type of communication and the type of processing system. However, it would be interesting to illustrate a practical example, defining working hypotheses and arriving at a selection of the optimal partition. In this case is possible to validate what has been presented.

 

Minor Comments:

1. In line 18, the authors should be writing the first letters of “wireless vision sensor networks” in the upper case. The same suggestion is for line 105: “Field-Programmable Gate Array” instead of “field-programmable gate array”.
2. In line 33, it is better writing “relationship” instead of “relation”.
3. In line 67, it is better structured the sentence in “Then, we study and compare…” instead of “We then study and compare…”
4. Authors should be careful to be consistent in how they refer to equations in the text. For an example, in lines 257, 470 and 539 is used this form “Equations (6) and (7),” instead in lines 296, 383, 432-433 is used “Equations 6 and 7”.
5. In line 418, “Figure 5” is repeated in the sentence. In fact, there is already a reference to the figure at the start of the sentence.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors carry out a very comprehensive study on the processing of images captured by cameras. They analyze the problems that exist with the processing of these images for the specific case of partitioning decisions for
IoT smart camera nodes.

Their study is based on a deep experimental study that supports the conclusions obtained in the work.

Authors must take care of the presentation of the graphics since many of them are too small and cannot be seen clearly.

For the rest, and once this fact has been corrected, the article can be considered for publication in this journal due to the relevance of the subject discussed and the current interest that exists in this subject in the scientific community.

Author Response

Dear Reviewer,

Thank you for giving us the opportunity to submit a revised version of the article titled “Impact of input data on intelligence partitioning decisions for IoT smart camera nodes” to the journal MDPI Electronics. We appreciate the time and effort that you have put in providing valuable feedback on our manuscript.

The suggestion provided have been addressed by making the article figures more readable.