====== Details about 3D ellipsoid fitting ====== The [[plugin:analysis:3d_analysis:start|3D ellipsoid fitting plugin]] will fit a 3D ellipsoid to objects in a labelled image. The ellipsoid axes and radii are computed using 3D moments : * sx2 = ∑(x-Cx)2/Vobj * sy2 = ∑(y-Cy)2/Vobj * sz2 = ∑(z-Cz)2/Vobj * sxy = ∑(x-Cx)(y-Cy)/Vobj * sxz = ∑(x-Cx)(z-Cz)/Vobj * syz= ∑(y-Cy)(z-Cz)/Vobj The three eigen values and vectors are then computed from this matrix : | sx2 | sxy | sxz | | sxy | sy2 | syz | | sxz | syz | sz2 | The three radii are then : √(5*eigen value) The different results are : * Cx,Cy,Cz : **centre** of the object (in pixel coordinate) * Vx, Vy, Vz : coordinate of the **main elongation** axis * R1, R2, R3 : the tree radii of the fitted ellipsoid * XY, XZ, YZ : **angle** between the main axis of the ellipsoid and the 3 planes * Vobj : **volume** of the object * Vell : volume of the fitted ellipsoid * Vbb : volume of the bounding box around the object * Vbbo (experimental) : volume of the bounding box oriented in the direction of the main axis * The two **poles** of the object as the two voxels used to compute Feret diameter, and the two poles of the fitting ellipsoid. {{:tutorial:plugins:ellipsoid.png|Object and fitted ellipsoid with main elongation axis}} Object and fitted ellipsoid with main elongation axis {{:tutorial:plugins:ellipsoid1.png|Bounding box and oriented bounding box}} Bounding box and oriented bounding box