@@ -50,6 +50,8 @@ The small frame is designed to be printable on an A4 page (or letter), and the m
...
@@ -50,6 +50,8 @@ The small frame is designed to be printable on an A4 page (or letter), and the m
## Results
## Results
### Grid scan
To quantify the accuracy of the scanning, a 25mm calibration grid was built:
To quantify the accuracy of the scanning, a 25mm calibration grid was built:
<imgsrc="img/grid.jpg"width=70%></img>
<imgsrc="img/grid.jpg"width=70%></img>
...
@@ -60,12 +62,29 @@ Here is the image used for analysis, showing the detected features (aruco marker
...
@@ -60,12 +62,29 @@ Here is the image used for analysis, showing the detected features (aruco marker
Here is the extracted image:
Here is the extracted image:
<imgsrc="img/result.jpg"width=70%></img>
<imgsrc="img/result.jpg"width=60%></img>
The grid feature points are automatically extracted:
The grid feature points are automatically extracted:
<imgsrc="img/grid_features.jpg"width=70%></img>
<imgsrc="img/grid_features.jpg"width=60%></img>
Finally, the distance between neighboring grid points is measured, revealing the accuracy of the scanning process. Here is a comparison of all frames, including the Shaper Trace, showing improved accuracy and less variance:
Finally, the distance between neighboring grid points is measured, revealing the accuracy of the scanning process. Here is a comparison of all frames, including the Shaper Trace, showing improved accuracy and less variance:
<imgsrc="img/plot.png"width=70%></img>
<imgsrc="img/plot.png"width=70%></img>
| | mean [mm] | stdev [mm] |
|------------------|-----------|------------|
| **small1** | 25.098 | 0.126 |
| **small2** | 25.017 | 0.102 |
| **medium1** | 25.048 | 0.123 |
| **medium2** | 25.056 | 0.123 |
| **large** | 25.086 | 0.112 |
| **Shaper Trace** | 24.888 | 0.201 |
### Resilience to fabrication error
To test the resilience of the scanning to process to fabrication errors, a distorted frame was designed and built. Its feature points are deviated horizontally and vertically with a standard deviation of 1mm.
<imgsrc="img/distorted_zoom.jpg"width=70%></img>
Thanks to the least squares optimization when retrieving the perspective transform, this error is successfully dampened. The results are shown in the table above, under the name **medium2**; as seen, the precision is the same as **medium1**, which was built with a tolerance of <0.4mm .