MovAlyzeR records all data that the pen tablet produces:
o x,y=Horizontal, vertical position, resp. (Origin is left bottom, Calibration via Input Device Settings)
o z=Axial Pen pressure in tablet units
Optionally (limited processing options):
o Altitude=Pen tilt
o Azimuth=Orientation of the projection of the pen onto the tablet
With additional, specific hardware (limited processing options):
o 2 pens or a pen and a mouse can be recorded simultaneously on one pen tablet
To check the data delivered by Wacom pen tablets:
Start >Control Panel >Wacom Tablet Properties
Press Control + Click Diagnosis

AXIAL PEN PRESSURE

Many pen tablet pens have a built-in force sensor for axial pen pressure. Pen pressure is stored as the z-coordinate in the 3rd column of the raw data HWR files. Processing carries over the pen pressure as is.

The number of pen pressure levels may be up to 2048 depending upon the pen tablet. The maximum pen pressure can be set in the Input-Device Settings. In MovAlyzeR an alarm can be enabled when the writer presses the pen beyond a settable pressure level.

The relation between pen pressure and z-coordinate can be altered in the tablet properties. We assume that the relationship is set to linear (i.e., not hard or soft), which is the factory setting (For Wacom: Connect tablet; Click Start >Control Panel >Wacom Tablet Properties; Bring pen close to tablet. Pen Toll will show up.)
Name:  WacomPenProperties.PNG
Views: 652
Size:  34.4 KB

Click Customize >Default. This will ensure a reproducible output.
Name:  WacomPenPressureDefault.PNG
Views: 525
Size:  16.4 KB

Move Threshold, Sensitivity, and Maximum Pressure output (within the digitization range). Here is an example of a hard setting where you have to press really hard to get some output. A soft setting would be where you get high output with the softest pressure like a soft pencil (Some may reverse hard and soft to make it consistent with photographic paper).
Name:  WacomPenPressureHard.PNG
Views: 673
Size:  16.8 KB

Axial pen pressure is the force in the direction of the pen axis. The normal (orthogonal) pressure on paper depends also on pen tilt and internal friction of the refill (as pressing the pen may require the refill to slide inside the pen barrel). The friction may show as hysteresis of the pressure (pressure output is increased after a high pressure burst).

Pen pressure may also lag behind relative to the x and y movements.

The pen has also its own weight (normally 10-20 grams). Postage scales may suffice (1 Oz =28.3495 grams). This is a force but not caused by active pressure by the writer. You may assign this value to zero-grams point as there is no active pen pressure.

The measure for force is Newton (N). A force can accelerate a mass (in grams). Force is a Mass * Acceleration. A mass has a predictable weight because of the Earth's standard gravity acceleration of 9.80665 m/s2. (Gravity acceleration varies less than 1%with location and height on Earth -- http://en.wikipedia.org/wiki/Gravity_of_Earth).

9.81 N is the force due to the weight of of a mass of 1 kg on Earth (9.81 kg m / s2). For our purpose we can use grams (g), implying this is the force due to the gravity acceleration exerted on a mass of 1 g.

EASY TO REMEMBER (metric made easy)
Exactly is as good as can be.
Precisely is close to exactly.
Otherwise: approximately.


1 kg =10 N (1 g = 0.01 N)
Weight of 5 US Cent = 5 g (exactly).
Weight of 1 US Cent = 2.5 g (exactly)
Weight of a cube of 10 cm of water = 1 kg (exactly), 1 cm3 (1 cc) of water weighs 1 g (precisely).
Weight of 1 Oz = 30 g
Weight of a pen = 15 g
High pen pressure = 400 g
Weight of a hand = 400 g
Test: Dip hand in pitcher full of water.
Volume of a hand = 400 cm3. Volume of water pushed out = 400 cm3 = 0.4 liter of water (precisely) = 0.4 kg (precisely).
Because Specific mass of limbs = 1 g/cm3 (like water) it means that the hand weighs 0.4 kg = 400 g (exactly)
PEN PRESSURE CALIBRATION

o Start MovAlyzeR
o Make a construction that positions the pen vertically with minimal friction, e.g., a plastic cup, upside down with an ample hole in the bottom.
o Read and log the P=... value in the recording window (top left).
o Place the pen in the cup inside the recording window. The pen rests by itself on the tablet. The pressure due to the weight of the pen can be considered as exerting 0 pen pressure.
o Tape rolls of 5 or 10 5-US-Cent pieces together. Tape these rolls to the top of the pen so you add 50 g, 100 g, 200 g, ... to the top of the pen. Write down the reading of the P= value.

You will then get a table like:
Test..........................P=
Pen in air......................0
Pen + 0 g..................100 <== Point of 0 pen pressure for free-standing pen
Pen + 50 g................250
Pen + 100 g...............400
Pen + 200 g...............700
Pen + 300 g.............1000
See for Pressure-calibration curves of various tablets: www.neuroscript.net/tablets >Tablet reviews

Hopefully your data can be approximated by a linear curve (not necessarily through the origin). See example of Wacom Intuos3 Ink Pen.
Name:  penpressure.png
Views: 557
Size:  5.3 KB

HOW TO CALIBRATE OUTPUT OF MOVALYZER

MovAlyzeR has deliberately selected centimeter (cm) and second (s) as its default measures for all features because research results have to be reported in metric measures and these units keep the figures close to 1. See Input Device Settings how to calibrate those settings. Always double-check the settings by drawing a square of 5 cm in 4 s, for example. MovAlyzeR must measure size and duration values within the accuracy margins or the settings are wrong. Never trust automatic settings.

For pen pressure no easy test exists and the pen pressure varies due to many factors. Therefore no calibration of pen force was attempted. However, it is possible to convert the raw data into gram-force data using External Apps or post-hoc calculations, e.g., in MS Excel making a scatter plot with straight lines between the measurements. Hopefully the chart is linear

The first step is to make a chart of the ADDED weight (in gram) horizontally (i.e., the chart's x-axis with the controlled variable) and the output z or P value vertically (i.e., the chart's y-axis with the dependent variable).

You can also chose to report the absolute pen pressure.

All is acceptable as long as details are added that suffice for others to replicate.

A generic formula to convert Units (i.t., the abstract pen tablet output units for pen force) into grams can be built as follows:

(Measured Value in Units - Offset from zero in Units) * (Range in grams) / (Range in Units)

You easily can recognize that the output of this formula is in grams.
In the above example we will thus get:
(z Value in Units - 100 Units) * (300 g - 0 g) / (1000 Units - 100 Units) =
(z Value in Units - 100 Units) * 300 g / 900 Units =
(z Value - 100) * 300 / 900
Verify this is correct for 2 values:
z Value =100 Units yields Pen Pressure = 0 g
z Value = 1000 Units yields Pen Pressure = 300 g
An automatic calibration can be done in MovAlyzeR using an External App which reads the raw HWR data and updates the TF processed output data.
In the Input Device Settings a new minimum pressure setting is needed, e.g., -100 because -100 means now pen is lifted.

PEN PRESSURE IN HANDWRITING

The control of pen pressure in handwriting: A subtle point
John Wann, Dept. of Psychology, University of Edinburgh, Edinburgh EH8 9YL, UK
Ian Nimmo-Smith, MRC Applied Psychology Unit, Cambridge, UK
http://dx.doi.org/10.1016/0167-9457(91)90005-I, How to Cite or Link Using DOI
http://www.sciencedirect.com/science...6794579190005I

Biometric Authentication
Lecture Notes in Computer Science Volume 3087, 2004, pp 283-294
A Novel Digitizing Pen for the Analysis of Pen Pressure and Inclination in Handwriting Biometrics
Christian Hook, Juergen Kempf, Georg Scharfenberg
http://link.springer.com/chapter/10....6-3_26?LI=true

The relation between pen force and pen-point kinematics in handwriting
L. R. B. Schomaker, R. Plamondon
Biological Cybernetics (impact factor: 1.59). 07/1990; 63(4):277-289. DOI:10.1007/BF00203451
http://citeseerx.ist.psu.edu/viewdoc...p=rep1&type=ps
https://www.researchgate.net/publica...in_handwriting