Thankyou for response.

The units of the x-axis in figures 9,10 and 11, are sequenced 'scanning steps' of the system. Step 1 represents the first 3-D data point logged for this accuracy test.
Step 2, represents the 2nd data point and so on.

Therefore, for any one of these scanning steps, we can observe the corresponding X,Y and Z accuracies.

Your feedback is valuable and much appreciated and we will provide more clarification in future work.

Regards

Ben Horan

Hello authors: If you wish, you could make changes to your paper to address the comment posted by Maria before the paper is printed. However, you would need to be quick. Thank you. DTP.

If you wish to amend your paper as Prof.Pham said, you can send the revised version to me or antony

Thank you

Thankyou for the information LiuH.

I have responded to you by email in more detail

Kind Regards

Ben

Dear Ben,

I have a few comments and questions to your paper as follows

Comments and Question No.1 :

In your abstract, you mention that the scanning system has an angular resolution of 1.8 degrees, and the results obtained demonstrate the system to have an accuracy of approximately +/- 2 cm at a scanning distances of 1.0 m.

This accuracy and resolution is very bad for industrial applications as well as art modeling. Compared to the commercial scanners with Laser Triangulation techniques: the accuracy is +/- 25 to 50 Micron (0.025 to 0.5 mm); and commercial scanners with Time-Of-Flight techniques, the accuracy is +/- (5 to 15 mm) with the range up to 1000 m for Cultural Heritage Documentation and City Modelling applications.

My questions are as follows:

Question 1.1: “What kind of applications that the proposed scanning system is aimed at?”

Question 1.2: “ How to increase the accuracy of the system TECHNICALLY and THEORETICALLY?. For example in Triangulation techniques, the error of the Z measurement is directly proportional to the square of Z, but inversely proportional to the focal length and the baseline length of the scanner. Therefore, increasing the baseline length can produce a better accuracy in the measurement.

Comments and Question No.2 :

The accuracy of the scanner depends a lot on the calibration techniques. My question here is

“How do you implement the calibration for the proposed system? Do you use a plane or sphere as the reference objects for calibration process?

Comments and Question No.3 : What is the Field of Views (FOV) and Data Collection Speed (number of points per second) of the proposed scanning system?

Thank you very much in advance for your reply

Le Chi Hieu, Cardiff University, Wales, UK

Dear Le,

Thankyou for your comments.

Question 1.1

The proposed system is aimed at applications where high-accuracy and acquisition time is not essential, and

where the system integrator needs a very low cost system. Our initial system prototype was developed for under

$1000AUD. This system is able to be programmed to scan any number of points in different locations and thus is

more suited to measuring the prescence of different objects with different 3-D locations as opposed to

performing full 3-D surface scans.

Question 1.2

This system respresents an initial prototype and there are areas which need further improvement. Our focus is

to demonstrate that a low-cost system of this type is achieveable and the challenges in developing such a

prototype.

In order to increase the accuracy of the system THEORETICALLY we could use a higher accuracy 1-D laser

rangefinder for the primary range measurement. This would of course work against the low-development cost we

have focused on.

TECHNICALLY the main inaccuracy in this system arises from the co-alignment of the pan and tilt axes of the

mirror assembly. Given the low-cost requirements of the system high-precision development proved a challenge.

Further work on this aspect would greatly improve system accuracy,

Question 2.0

The laser device itself is an off-the-shelf device performing its own calibration. We are however required to

calibrate the position and orientation between the scanner and mirror. In this system we implement a reference

plane and circle for calibration. The system is placed at a marked origin and the tilt direction is fixed at a

known angle while the mirror assembly is panned through the scanning range. The system is then manually

adjusted until the system scanning points correspond to the pre-defined calabration markings.

Question 3.0

The FOV is 330 degrees in the azimuth direction and 100 degrees in the angle of elevation. The data collection

is limited to 2 seconds per measurement, given this limitation imposed by the implemented 1-D laser

rangefinder. Given this large acquisition time, this system is most suited to scanning only multiple points of

objects in order to find the presence of features.

Thankyou for you for your comments

Kind Regards

Ben