changes to previous version dxf_planefit

'Updated' to the previous version of dxf_planefit.m

Author: nicmenegoni

ROKA_algorithm/dxf_planefit.m

@@ -330,8 +330,6 @@
 
 
 plane3d = [dip, dipdir, radius, centroid, N, M, K, mean_err, stdev_err];
-
-
 end
 %% Functions to read and write dxf
 % by Grzegorz Kwiatek 2011
@@ -657,94 +655,3 @@ function dxf_close(FID)
   rethrow(exception);
 end
 end
-function plane3d = fit3dplane(PC)
-% This function is developed following the work of Jones et al. (2015)
-% and Seers and Hodgetts (2016)
-% 
-% INPUT
-% 'PC' (point cloud) is a vector or a matrix N x 3 (N=number of points)
-% that contain X Y Z coordinate of the points of the cloud
-
-centroid = mean(PC); % calculate centroid of point cloud
-
-%[U, S, W] = svd(PC);
-
-C = cov(PC); % calculate the covariance matrix 'C'
-
-[V, D] = eig(cov(PC)); % calculate eigenvector 'V' and eigenvalue 'D' of the
-% covariance matrix 'C'
-
-
-M = log(D(3,3)/D(1,1));% calculate vertex coplanarity 'M'
-K = log(D(3,3)/D(2,2))/log(D(2,2)/D(1,1));% calculate vertex collinearity 
-%'K'
-
-
-N = V(:,1)';
-
-% Normals have to be oriented upward!!! (Normals have to be horiented in same
-% direction)
-
-if N(1,3)<0 %if normal is oriented downward
-    N(1,1)=-N(1,1);
-    N(1,2)=-N(1,2);
-    N(1,3)=-N(1,3);    
-end
-cosAlpha = N(1,1)/norm(N);
-cosBeta = N(1,2)/norm(N);
-cosGamma = N(1,3)/norm(N);
-
-dip = 90 + rad2deg(asin(-cosGamma));
-
-if cosAlpha > 0 && cosBeta > 0
-dipdir = rad2deg(atan(cosAlpha/cosBeta));
-elseif cosAlpha > 0 && cosBeta < 0
-dipdir = 180 + rad2deg(atan(cosAlpha/cosBeta));
-elseif cosAlpha < 0 && cosBeta < 0
-dipdir = 180 + rad2deg(atan(cosAlpha/cosBeta));
-elseif cosAlpha < 0 && cosBeta > 0
-dipdir = 360 + rad2deg(atan(cosAlpha/cosBeta));
-end
-
-%calculate radius of plane = max distance point(projected onto the plane)-centroid of fitted plane
-Plane=createPlane(centroid,N);
-for i=1:length(PC(:,1))
-PT(i,:) = projPointOnPlane(PC(i,:), Plane);
-end
-
-for i=1:numel(PC(:,1))
-pdist1 (i) =sqrt((centroid(1,1)-PT(i,1))^2 + (centroid(1,2)-PT(i,2))^2 + (centroid(1,3)-PT(i,3))^2);
-end
-
-[pM1, pI1] = max(pdist1);
-
-for i=1:numel(PT(:,1))
-    if i==pI1
-        pdist2(i)=0;
-    else
-        pdist2 (i) =sqrt((PT(pI1,1)-PT(i,1))^2 + (PT(pI1,2)-PT(i,2))^2 + (PT(pI1,3)-PT(i,3))^2);
-    end
-end
-[pM2, pI2] = max(pdist2);
-
-centroid=mean([PT(pI1,:); PT(pI2,:)]);
-radius =(sqrt((PT(pI1,1)-PT(pI2,1))^2 + (PT(pI1,2)-PT(pI2,2))^2 + (PT(pI1,3)-PT(pI2,3))^2))/2;
-
-%calculation of mean error and standard deviation of fitted plane
-d = -sum(bsxfun(@times, N, bsxfun(@minus, centroid, PT)), 2);
-mean_err=mean(abs(d));
-stdev_err=std(abs(d));
-
-
-plane3d = [dip, dipdir, radius, centroid, N, M, K, mean_err, stdev_err];
-
-
-% REFERENCES:
-%
-% - Jones, R. R., Pearce, M. A., Jacquemyn, C., & Watson, F. E. (2016). 
-%   Robust best-fit planes from geospatial data. Geosphere, 12(1), 196-202.
-%
-% - Seers, T. D., & Hodgetts, D. (2016). Probabilistic constraints on
-%   structural lineament best fit plane precision obtained through
-%   numerical analysis. Journal of Structural Geology, 82, 37-47.
-end