NAME

s.correction

SYNOPSIS

s.correction
s.correction help
s.correction [-g] input=string output=string out_ter=string Sce=float Scn=float [lambda_c=float] [Tc=float] [tc=float]

DESCRIPTION

s.correction allows a user to refine the classification produced by s.growing command.

 This command is part of a procedure to filter airbone laser scanning (LIDAR) data in order to extract Digital Terrain Model.
The procedure involves the execution of the following commands:
1) s.edgedetection - 2) s.growing - 3) s.correction - 4) s.bspline.reg

OPTIONS

The user can run this program either interactively or non- interactively. The program will be run non-interactively if the user specifies program arguments and flag settings on the command line using the form:

s.correction [-g] input=string output=string out_ter=string Sce=float Scn=float [lambda_c=float] [Tc=float] [tc=float]

Alternately, the user can simply type s.correction on the command line without program arguments. In this case, the user will be prompted for parameter values and flag settings using the standard GRASS parser interface described in the manual entry for parser.

Flags:

-g
gridded output sites list file

Parameters:

input=string
input sites list file name (s.growing output)
output=string
output classificated sites list file name
out_ter=string
output bere earth sites list file name
Sce=float
interpolation spline step value in east direction
Scn=float
interpolation spline step value in north direction
lambda_c=float
regualarization weight in reclassification evaluation
Default: 1
Tc=float
high threshold for object to terrain reclassification
Default: 2
tc=float
low threshold for terrain to object reclassification
Default: 1


NOTES

Misclassification in the procedure output remains, due to particular geometry of some objects (the simplest we can recall is the case where we have unusual roofs with pitches at different heights).
To overcome this problem, a bilinear interpolation with Tikhonov regularising parameter lambda_c only on the points classified as bare earth is performed. The analysis of the residuals (D) between the observations and the interpolated values compared with two thresholds tc ,Tc show four cases:

·        if the point is classified as ground and D > Tc , it is reclassified as object;
·        if the point is classified as ‘double pulse ground’ and D > Tc , it is reclassified as ‘double pulse object’ (edge or vegetation);
·        if the point is classified as object and |D| < tc , it is reclassified as ground;
·        if the point is classified as ‘double pulse object’ and |D| < tc, it is reclassified as ‘double pulse ground’.

Every point of the input site file is examinated and reclassified in the output file as:
·       terrain (0)
·       terrain with double pulse (1)
·       object with double pulse (2)
·       object (3)

input format example (x|y|z):

499453.39|5418722.51|#0 %245.99 %1
499453.52|5418725.56|#0 %246.19 %1
499453.67|5418728.67|#0 %246.2 %2
499453.81|5418731.8|#0 %246.2 %3
499453.9|5418734.65|#0 %247.13 %0
499454.66|5418734.31|#0 %246.99 %0 

output format example (x|y|z%classification%bicubic_interpolated_value): 

499453.39|5418722.51|#0 %245.99 %1
499453.52|5418725.56|#0 %246.19 %2
499453.67|5418728.67|#0 %246.2 %1
499453.81|5418731.8|#0 %246.2 %0
499453.9|5418734.65|#0 %247.13 %0
499454.66|5418734.31|#0 %246.99 %0

SEE ALSO

s.edgedetection, s.bspline, s.growing, s.to.qrast

AUTHORS

Maria A. Brovelli, Politecnico di Milano - Campus Como
Massimiliano Cannata, Politecnico di Milano - Campus Como
Ulisse M. Longoni, Politecnico di Milano - Campus Como

REFERENCES

Brovelli M.A.; Cannata M.; Longoni U.M., LIDAR Data Filtering and DTM Interpolation Within GRASS
 Transactions in GIS, April 2004, vol. 8, no. 2, pp. 155-174(20)  - Blackwell Publishing Ltd