Some problems about using the least squares vector of lming_08

Lming_08

How are you doing.

I'm quoting you. Surface normal vector based on least squares method for estimating point clouds some problems encountered in this article, if you can see, please enlighten me.

1. The result of using _method1_ and least squares method is exactly the same.

2, the use of _method2_, the problem is as follows

I slightly changed the plane normal vector code (second coordinate system) directly applied to find the plane equation of point, In the process of using the discrete point is not a plane is fitted out of the plane equation and the least squares to fit the results of the equation difference and large (you give an example of the point cloud to fit the plane and the least squares to fit the plane equation), I can directly use whether it is possible, the question is where it is.

I'm using the point cloud to fit the plane's code

void Computepointnormal (const PCL::P OINTCLOUD<PCL::P ointxyz> &cloud, float &nx, float &ny, float
	&nz,float &nd) {#define _METHOD2_ #ifdef _method2_ eigen::matrixx4f mat_ext (Cloud.size (), 4);
		for (size_t i = 0, ind_num = cloud.size (); i < Ind_num; ++i) {Mat_ext (i, 0) = cloud.points[i].x;
		Mat_ext (i, 1) = CLOUD.POINTS[I].Y;
		Mat_ext (i, 2) = Cloud.points[i].z;
	Mat_ext (i, 3) = 1;    } eigen::matrixxf Mat_mult_ext = Mat_ext.transpose () * MAT_EXT;
	The mat is the 3*k_ matrix, Mat_ext is the k_*4 matrix//a*x = B, A.row (0). Cross (A.row (1)) = eigenvector (0);
	eigen::vector4f Row_vec[3];
	Row_vec[0] = mat_mult_ext.row (0);
	ROW_VEC[1] = Mat_mult_ext.row (1);
	ROW_VEC[2] = Mat_mult_ext.row (2);    eigen::matrix3f I_cofactor, J_cofactor, K_cofactor, L_cofactor; When calculating the cross product, the algebraic cofactor for the I, J, K, L components (size_t row_index = 0; row_index < 3; ++row_index) {for (size_t col_index = 0; col_in Dex < 3; ++col_index) {i_cofactor (Row_index, col_index) = Row_vec[row_index][col_index + 1];
			if (Col_index < 1) {J_cofactor (Row_index, col_index) = Row_vec[row_index][col_index];
			} else {j_cofactor (row_index, col_index) = Row_vec[row_index][col_index + 1];
			} if (Col_index < 2) {K_cofactor (Row_index, col_index) = Row_vec[row_index][col_index];
			} else {k_cofactor (row_index, col_index) = Row_vec[row_index][col_index + 1];
		} l_cofactor (Row_index, col_index) = Row_vec[row_index][col_index];
	}} Float I_dim = 0, J_dim = 0, K_dim = 0, L_dim = 0;
	I_dim = I_cofactor.determinant ();
	J_dim = J_cofactor.determinant ();
	K_dim = K_cofactor.determinant ();
	L_dim = L_cofactor.determinant ();
eigen::vector4f Coeff_vec (I_dim,-j_dim, K_dim,-l_dim);
eigen::vector4f argument = MAT_MULT_EXT.COLPIVHOUSEHOLDERQR (). Solve (b);
	eigen::vector4f Coeff_vec11 (argument[0], argument[1], argument[2],argument[3]);

Coeff_vec.squarednorm ();
eigen::vector3f Coeff_vec (I_dim,-j_dim, K_dim); float len = coeff_vec.sqUarednorm ();
 	Coeff_vec/= Len;

	Coeff_vec/= K_dim;
	NX = Coeff_vec[0];
	NY = coeff_vec[1];
	NZ = coeff_vec[2];
	
nd = coeff_vec[3];
 #endif}

The point cloud that I used to test

X YZ
P1 0 00.01
P2 0 10.2
P3 0 2-0.2
P4 1 00.25
P5 1 1-0.3
P6 1 2-0.1
P7 2 00.2
P8 2 1-0.2
P9 2 20.15

The equation of least squares fitting
z = 0.079 + 0.023x + -0.102y
The equation that your program fits
Z=1.645+0.563x+0.689y
Please enlighten me. Thank you

If you are willing to answer my questions, please leave a message or contact me dearpeer@126.com 719594159