User Tools

Site Tools


visual3d:documentation:pipeline:expressions:least_squares_fitting_of_data

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
visual3d:documentation:pipeline:expressions:least_squares_fitting_of_data [2024/07/12 13:29] – removed sgrangervisual3d:documentation:pipeline:expressions:least_squares_fitting_of_data [2025/01/24 18:53] (current) – [Best Fit Plane] wikisysop
Line 1: Line 1:
 +===== Least Squares Fitting of Data =====
  
 +This page contains a list of all of [[visual3d:documentation:pipeline:expressions:expressions_overview|Evaluate_Expression]]'s functions that are used to find a line or plane of best fit to your data.
 +
 +==== Best Fit Plane ====
 +
 +**Best_Fit_Plane**(signal, start_event_signal, end_event_signal) - Finds a plane that fits the path of a point from a start event to an end event. [[https://en.wikipedia.org/wiki/Plane_%28geometry%29#Point-normal_form_and_general_form_of_the_equation_of_a_plane|The resulting plane is defined by 4 components]].
 +
 +The general form of the equation of a plane in 3D is ax+by+cz+d = 0 where 𝑎, 𝑏, and 𝑐 are the components of the normal vector which is perpendicular to the plane or any vector parallel to the plane.
 +
 +If (𝑥0,𝑦0,𝑧0)is a point that lies on the plane, then 𝑑=−(𝑎𝑥0+𝑏𝑦0+𝑐𝑧0) and as 𝑎𝑥+𝑏𝑦+𝑐𝑧−(𝑎𝑥0+𝑏𝑦0+𝑐𝑧0)=0.
 +
 +The result of the Best_Fit_Plane function is a signal with the fource components (a,b,c,d)
 +
 +<code>
 +! Example: Compute a Best_Plane_Fit for one signal across a range of frames
 +Evaluate_Expression
 +/EXPRESSION=Best_Fit_Plane(CURRENT_SIGNAL,EVENT_LABEL::ORIGINAL::START,EVENT_LABEL::ORIGINAL::END)
 +/SIGNAL_TYPES=TARGET
 +/SIGNAL_FOLDER=ORIGINAL
 +/SIGNAL_NAMES=LSK_1
 +/RESULT_TYPES=METRIC
 +/RESULT_FOLDERS=PROCESSED
 +/RESULT_NAME=LSK1_PLANE
 +/APPLY_AS_SUFFIX_TO_SIGNAL_NAME=TRUE
 +;
 +</code>
 +
 +<code>
 +! Example: Compute a Best_Plane_Fit for multiple signals at each frame of data
 +Evaluate_Expression
 +/EXPRESSION=Best_Fit_Plane(CURRENT_SIGNAL)
 +/SIGNAL_TYPES=TARGET
 +/SIGNAL_FOLDER=ORIGINAL
 +/SIGNAL_NAMES=LSK_1+LSK_2+LSK_3
 +/RESULT_TYPES=DERIVED
 +/RESULT_FOLDERS=PROCESSED
 +/RESULT_NAME=LSK_PLANE
 +/APPLY_AS_SUFFIX_TO_SIGNAL_NAME=TRUE
 +;
 +</code>
 +
 +==== Best Fit Circle ====
 +
 +**Best_Fit_Circle**(signal, start_event_signal, end_event_signal) - Fits a 2D circle to the path of a signal.
 +
 +The algorithm computes a best fit plane to the data, rotates this plane into a principal plane, computes the center and radius, the rotates these values back into the original plane. The result is 4 components:
 +  - the origin's X-coordinate;
 +  - the origin's Y-coordinate;
 +  - the origin's Z-coordinate; and
 +  - the radius.
 +
 +==== Best Fit Sphere ====
 +
 +**Best_Fit_Sphere**(signal,start_event_signal,end_event_signal) - Fits a 3D sphere to the path of a signal.
 +
 +<code>
 +! Example : Fit a sphere to 6 points on a DIAMOND
 +
 +! Create 6 TARGETS representing the Vertices
 +Create_Target
 +/SIGNAL_NAMES=V1
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES),0,0)
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=V2
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((-0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES),0,0)
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=V3
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((0,0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES),0)
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=V4
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((0,-0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES),0)
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=V5
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((0,0,0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES))
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=V6
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR((0,0,-0.5+0*FRAME_NUMBERS::ORIGINAL::FRAMES))
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Evaluate_Expression
 +/EXPRESSION=Best_Fit_Sphere(CURRENT_SIGNAL)
 +/SIGNAL_TYPES=TARGET
 +/SIGNAL_FOLDER=ORIGINAL
 +/SIGNAL_NAMES=V1+V2+V3+V4+V5+V6
 +/RESULT_TYPES=DERIVED
 +! /RESULT_FOLDERS=PROCESSED
 +/RESULT_NAME=SPHERE
 +! /APPLY_AS_SUFFIX_TO_SIGNAL_NAME=FALSE
 +;
 +</code>
 +
 +Note that TARGETS are created and the sphere computed has each TARGET on its surface.
 +
 +{{:Sphere1.jpg}}{{:sphere6view.jpg}}
 +
 +<code>
 +! Example : Create a TARGET where each frame is on a random location on the surface of a sphere of radius 1
 +Evaluate_Expression
 +/EXPRESSION=RAND(-PI(),PI(),FRAME_NUMBERS::ORIGINAL::FRAMES)
 +/RESULT_TYPES=DERIVED
 +/RESULT_FOLDERS=RADIUS
 +/RESULT_NAME=VERTICAL
 +! /APPLY_AS_SUFFIX_TO_SIGNAL_NAME=FALSE
 +;
 +
 +Evaluate_Expression
 +/EXPRESSION=RAND(0,2*PI(),FRAME_NUMBERS::ORIGINAL::FRAMES)
 +/RESULT_TYPES=DERIVED
 +/RESULT_FOLDERS=RADIUS
 +/RESULT_NAME=HORIZONTAL
 +! /APPLY_AS_SUFFIX_TO_SIGNAL_NAME=FALSE
 +;
 +
 +Create_Target
 +/SIGNAL_NAMES=SPHERE
 +! /SIGNAL_DESCRIPTION=
 +/EXPRESSION=VECTOR(
 +COS(DERIVED::RADIUS::VERTICAL)*COS(DERIVED::RADIUS::HORIZONTAL),
 +-COS(DERIVED::RADIUS::VERTICAL)*SIN(DERIVED::RADIUS::HORIZONTAL),
 +SIN(DERIVED::RADIUS::VERTICAL))
 +! /INCLUDE_CALFILE=FALSE
 +;
 +
 +Evaluate_Expression
 +/EXPRESSION=Best_Fit_Sphere(CURRENT_SIGNAL)
 +/SIGNAL_TYPES=TARGET
 +/SIGNAL_FOLDER=ORIGINAL
 +/SIGNAL_NAMES=SPHERE
 +/RESULT_TYPES=DERIVED
 +! /RESULT_FOLDERS=PROCESSED
 +/RESULT_NAME=SPHERE
 +! /APPLY_AS_SUFFIX_TO_SIGNAL_NAME=FALSE
 +;
 +</code>
 +
 +{{:random8view.jpg}}
 +{{:sphereRandom.jpg}}
 +
 +==== Simple Linear Regression ====
 +
 +**Simple_Linear_Regression**(signal1, signal2, start_event, end_event); - Fits a signal to a line given by the equation Y = mX + b where
 +
 +<code>
 +Signal = m Signal1 + b
 +</code>
 +
 +An explanation of the calculation can be found [[Visual3D:Documentation:Statistics:Compute_Linear_Regression|here]] or [[https://en.wikipedia.org/wiki/Simple_linear_regression|here]].
 +
 +Signal1 and Signal2 are both one-component signals and the resulting METRIC contains 6 components:
 +  - Slope = m -> Component 1
 +  - Intercept = b -> Component 2
 +  - Siga = uncertainty in m -> Component 3
 +  - Sigb = uncertainty in b -> Component 4
 +  - Chi2 = chi square -> Component 5
 +  - Q = The R^2 statistic -> Component 6
visual3d/documentation/pipeline/expressions/least_squares_fitting_of_data.1720790957.txt.gz · Last modified: 2024/07/12 13:29 by sgranger