in this post i will present the results obtained bu the Santos approach from the Dietmayer Segmentation, this algorithm took a little while because it was the only one which its threshold value depends on 2 parameters, the C0 which is a constant parameter used for noise reduction (the same as in the Dietmayer Segmentation) and an additional parameter β which aims to reduce the dependence of the segmentation with respect to the distance between the laser and the objects.
Just to recall the Threshold condition:
The tests:
The procedure was similar to the testing of the other algorithms - Divied in route in 3 parts and compared the results with the ground truth results using the metric presented 2 posts ago.
However with this particular algorithm i divided the tests in 2 parts:
- To inquire the influence of the C0 parameter, I fixed the value of β = 10º
- To inquire the influence of β, I fixed the values of C0 = 1 (m)
Complete route:
β = 10º
Minimum Energy value: 148.4237
Minimum Energy C0 value: 1.375
C0 = 1.0 m
Minimum Energy value: 222.7269
Minimum Energy β value: 5º
Comments:
- The performance of the algorithm is far more dependable form the C0 parameter than the β value.
- The β curve has a global minimum for the 5º reversal, them stabilizes, once again shows that the applied metric may not be the most appropriated for testing.
Straight road:
β = 10º
Minimum Energy value: 155.0718
Minimum Energy C0 value: 1.375
C0 = 1.0 m
Minimum Energy value: 357.3028
Minimum Energy β value: 25º and 27.5º
Comments:
- The β curve has a sharp decline from 15º to 17.5º and them stabilizes again
Roundabout approach:
β = 10º
Minimum Energy value: 129.5992
Minimum Energy C0 value: 1.75
C0 = 1.0 m
Minimum Energy value: 130.2633
Minimum Energy β value: 10º
Comments:
- This time the C0 presented a sinusoidal form.
- The β curve presented a "peak" for 15º. Pretty constant for the other β values
Roundabout:
β = 10º
Minimum Energy value: 146.9297
Minimum Energy C0 value: 1.375 and 1.500
C0 = 1.0 m
Minimum Energy value: 124.3453
Minimum Energy β value: 5º
Comments:
- For this course the C0 present also small variations, minimum energy value = 146.9297 and the maximum energy value = 147.4547
- The β curve has a global minimum for the 5º reversal, them stabilizes, once again shows that the applied metric may not be the most appropriated for testing.
MAIN CONCLUSIONS:
- The performance of the algorithm is far more dependable form the C0 parameter than the β value.
- Once again the results are not very satisfying - We cant conclude what is the best threshold value for the algorithm