Replace libLab3_Lidar.py

Labs/Lab_3_LiDAR_ML_Segmentation/libLab3_Lidar.py

@@ -109,7 +109,7 @@ def getEigenvaluePCA(a, b, c, dim, decim=2):
 	
 	"""
 	from sklearn.decomposition import PCA
-	from scipy import spatial
+	#from scipy import spatial
 	
 	
 	
@@ -123,7 +123,7 @@ def getEigenvaluePCA(a, b, c, dim, decim=2):
 	print("shape of Y " ,np.shape(Y))
 
 	# Estimate the distance of one point with every neighbour around
-	dist = spatial.distance.squareform(spatial.distance.pdist(Y))	
+	#dist = spatial.distance.squareform(spatial.distance.pdist(Y))	
 	
 
 	
@@ -139,13 +139,18 @@ def getEigenvaluePCA(a, b, c, dim, decim=2):
 	#   3. Using PCA scikit-learn function, we estimate the 3 eigenvalues of these points
 	#   4. We concatenate these 3 eigenvalues in a single array which will be return at the end.
 	for kk in range(0,n_samples):
-		pts = np.where((dist[kk,:]<=dim/2))  # Mask of values corresp. to the criteria
-											 
+		dist =  np.sqrt((Y[kk,0] - Y[:,0])**2 + (Y[kk,1] - Y[:,1])**2 + (Y[kk,2] - Y[:,2])**2)
+
+		# Mask of values corresp. to the criteria
+		pts = np.where((dist<=dim/2))
+									 
 
 		if np.size(pts)<3:			 # Check if the number of neighbour is sufficient	 
-			print("Revise dims: ")
-			print(pts)
-			print(dist)
+			#print("Revise dims: ")
+			#print(pts)
+			#print(dist)
+			eigenvalues=np.array([0., 0., 0.])
+
 		else:
 			Ytmp = Y[pts,:]			 # Apply the criteria on the dataset
 			Ytmp = Ytmp[0,:,:]		  # Reduce the depth of the array	   
@@ -483,7 +488,7 @@ def plot_3dcladd(dx,dy,dz,y,density=1):
     for xb, yb, zb in zip(Xb, Yb, Zb):
        ax.plot([xb], [yb], [zb], 'w')
     plt.show()
-	
+
 
 def plot_confMat(cnf_matrix,ClaName):
 	"""