Image Engineering GEOCAL User manual

Image Engineering GmbH & Co. KG · Im Gleisdreieck 5 . 50169 Kerpen . Germany
T [phone]-0 . F [phone]-10 . www.image-engineering.com
GEOCAL / GEOCAL XL
User Manual
22. November 2021

Image Engineering
Seite 2von 19
Content
1INTRODUCTION......................................................................................................... 4
Conformity ............................................................................................................ 4
Intended use......................................................................................................... 4
General safety information..................................................................................... 4
2GETTING STARTED................................................................................................... 5
Scope of delivery................................................................................................... 5
Commissioning...................................................................................................... 5
3OPERATING INSTRUCTIONS HARDWARE ............................................................... 6
GEOCAL / GEOCAL XL......................................................................................... 6
Diffractive Optical Element (DOE).......................................................................... 6
Camera settings.................................................................................................... 6
3.3.1 Camera position and suitable lenses ............................................................... 7
3.3.2 Exposure........................................................................................................ 8
4DIFFERENCES BETWEEN GEOCAL AND GEOCAL XL............................................. 9
Why a larger version of the GEOCAL?................................................................... 9
5OPERATING INSTRUCTIONS SOFTWARE...............................................................10
Installing GEOCAL software..................................................................................10
Configuration file (XML)........................................................................................10
Loading images for analysis..................................................................................12
5.3.1 Delete images ...............................................................................................12
Image analysis.....................................................................................................13
Saving results ......................................................................................................14
Export Grids.........................................................................................................15
Plotting ................................................................................................................15
5.7.1 Grid view.......................................................................................................15
4.7.2 Distortion curve.............................................................................................15
5.7.2 Quiver plot.....................................................................................................15
5.7.3 2D Color scatter.............................................................................................15
5.7.4 Histogram......................................................................................................15
Logging................................................................................................................15
Quit......................................................................................................................16
TRADEMARK AND COPYRIGHT.........................................................................16
6ADDITIONAL INFORMATION....................................................................................17

Image Engineering
Seite 3von 19
Disposal instructions.............................................................................................17
Power circuit modification for production line use...................................................17
7DATA SHEET ............................................................................................................18
Overview..............................................................................................................18
Features ..............................................................................................................18
7.2.1 Hardware ......................................................................................................18
7.2.2 Illumination....................................................................................................18
7.2.3 Software........................................................................................................18
General description hardware...............................................................................19
Requirements on the device under test (DUT).......................................................19

Image Engineering
Seite 4von 19
1 INTRODUCTION
Important information: Read the manual carefully before using the device.
Inappropriate utilization may cause damage to the device.
Conformity
We, Image Engineering GmbH & Co. KG, hereby declare that GEOCAL/GEOCAL XL corresponds to the
essential requirements of the following EC directives:
- 2011/65/EU - RoHS 2
- 2014/35/EU - Low Voltage
- 2014/30/EU - Electromagnetic Compatibility
And standards or normative documents:
- EN 61326-1:2013 –Electrical equipment for measurement, control, and laboratory use - EMC
requirements - Part 1: General requirements
- EN 61000-3-2:2014 - Electromagnetic compatibility (EMC) - Part 3-2: Limits - Limits for harmonic
current emissions
- EN 61000-3-3:2013 - Electromagnetic compatibility (EMC) - Part 3-3: Limits
- EN 62471-2:2009 - Photobiological safety of lamps and lamp systems
- DIN EN 60825-1 - Safety of laser products - Part 1: Equipment classification and requirements
Intended use
GEOCAL/GEOCAL XL consists of an illuminating hardware device and software for geometric calibration
of camera systems. The GEOCAL software is used for calculating the calibration parameters from a
single image taken with a camera system of the point grid generated by the GEOCAL hardware device.
•Only suitable for indoor use.
•Place the system in a dry and constant tempered environment. Avoid high air humidity.
General safety information
WARNING! GEOCAL / GEOCAL XL uses a laser diode for illumination.
LASER RADIATION CLASS 1M LASER PRODUCT
•Do not stare into the beam
•Do not view directly with optical instruments
•Do not open the housing of GEOCAL / GEOCAL XL under any circumstances

Image Engineering
Seite 5von 19
2 GETTING STARTED
Scope of delivery
•GEOCAL / GEOCAL XL hardware device
•The latest software version on the USB storage device
•Power supply + cable
•USB cable (type B to type A)
•Test report
Commissioning
•Remove the packaging material.
•The power socket and the main power switch are located on the back side of the device next to
the USB type B socket.
Connect GEOCAL / GEOCAL XL to a power outlet and switch it on.
Please note that the blue LED next to the power switch indicates that the power line is active
and that GEOCAL is ready for operation.
•Install the latest GEOCAL software (Windows, 64bit)

Image Engineering
Seite 6von 19
3 OPERATING INSTRUCTIONS HARDWARE
GEOCAL / GEOCAL XL
After switching on, GEOCAL / GEOCAL XL is ready for use. No warm-up phase is required.
Diffractive Optical Element (DOE)
A diffractive optical element (DOE) is used to split the incoming light beam and generate a calibration grid
of evenly distributed light spots with a wavelength of 633 nm. The DOE is mounted on the front of the
GEOCAL / GEOCAL XL device (visible glass plate mounted in circular cutout, see Figure 2).
Camera settings
Set the camera to manual exposure mode. If manual mode is not available, use the automatic exposure
mode. The image should be saved in a lossless format for best results. TIFF format is recommended, and
PNG and JPEG (lossy) are also supported. If you convert RAW image data to another format, ensure the
image is debayered in the process.
The autofocus must be turned off. Since GEOCAL / GEOCAL XL uses a collimated light beam virtually
originating from infinity, the appropriate focus distance will be near infinity. Using manual focus, ensure
the camera is focused on the light points.
Exposure program
Manual
Aperture
No default value
ISO Speed
Lowest value (e.g.,100)
Autofocus
Off
File type
TIFF is recommended. PNG
and JPEG are also supported
DOE
Figure 1: Calibration grid
(the brightest spot marks the center)
Figure 2: Location of the DOE

Image Engineering
Seite 7von 19
3.3.1 Camera position and suitable lenses
Place the camera in front of the diffractive optical element located in the circular opening on the front of
GEOCAL / GEOCAL XL.
The camera can be placed directly in front of the DOE; no minimum distance is needed (view Figure 3
below). The principal axis of the lens should be aligned approx. perpendicular to the DOE. Achieve this by
aligning the 0th diffraction order, the brightest spot in the grid center, with the center of the image. See
Figure 4 for reference. Make sure that the point grid covers the whole image and that the border of the
DOE is not visible.
The top side of GEOCAL / GEOCAL XL and the top side of the camera must have the same orientation.
A rotation of the camera around the optical axis of approx. +/- 2° will be tolerable.
No further alignment steps are necessary. The camera's translation does not influence the mapping of the
light points. The camera rotation to the DOE is part of the calibration and is determined during the
process and reported as part of the result.
The lens's front element must not be larger than 77 mm in diameter for GEOCAL or 155mm for GEOCAL
XL to capture the point grid in full format.
Figure 3: Setting the distance between camera and GEOCAL / GEOCAL XL
Figure 4: Good alignment (0th order in center of image) Bad alignment (0th order far off center)

Image Engineering
Seite 8von 19
Note: Please use caution when positioning the camera. The DOE is made of glass and has a
thickness of only 1.5 mm.
Ensure the calibration is performed in a dark environment to avoid stray light, reflections, or
similar interfering factors.
3.3.2 Exposure
For best results, the exposure should be selected so that the individual light points are not saturated (i.e.,
no white pixels or a minimal amount of white pixels in the center of each point) with the exception of the
0th diffraction order. The 0th order will always be much brighter than the points surrounding it. See Figure
5 below for an example.
Depending on your camera, you might encounter different problems:
The exposure time of my camera cannot be set short enough to avoid saturation.
Possible solutions:
- Set the ISO to the lowest native value available (ISO 100 for most cameras)
- Stop down the aperture if possible
- If the image is still too bright, you could use an ND-Filter to reduce the amount of light
reaching the image sensor
The intensity dropoff is too steep towards the edges of the image, and no more points are visible.
This problem may occur for ultra-wide-angle lenses. If the images can not be adequately analyzed, try the
following.
Generate an HDR image by taking multiple images of the point grid using a range of exposure times (do
not change aperture settings in this case) and combine them into a high dynamic range image to achieve
more or less uniform exposure across the entire sensor. Do not change the camera's orientation between
images because the points would no longer be aligned to one another through multiple images.
Figure 5: Suitable exposure (dark backgroumd, small points) Exposure too bright (red background, larger, saturated points)

Image Engineering
Seite 9von 19
4 DIFFERENCES BETWEEN GEOCAL AND GEOCAL XL
The key difference between GEOCAL XL and the standard GEOCAL is the physical size of the
diffractive optical element (DOE) that generates the point grid. While the standard GEOCAL has a
usable diameter of 77 mm, the GEOCAL XL provides a functional diameter of 155 mm.
Why a larger version of the GEOCAL?
A larger DOE is needed when the device under test (DUT) can not be placed directly in front of the
DOE, for example, when a car's windshield is involved. As mentioned in chapter 3.3.1, "Camera
position and suitable lenses," the generated point grid needs to cover the entire camera sensor to
provide a reliable calibration.
Figure 6: Use case for the larger DOE provided in the GEOCAL XL
As illustrated in Figure 6 above, the windscreen angle prevents the standard GEOCAL from being
placed close enough to the camera to provide a full-frame image of the standard DOE. This angle
would result in a less precise and less reliable calibration.

Image Engineering
Seite 10 von 19
5 OPERATING INSTRUCTIONS SOFTWARE
The GEOCAL software lets you completely calibrate your camera system in a matter of seconds.
The results can be saved as CSV or XML files.
Installing GEOCAL software
•Execute the GEOCAL installer (GEOCAL_Vx.x.x.exe, 64 bit) and follow the instructions.
•The software is now ready for use.
Configuration File (XML)
A standard config file will be installed with the software. If your calibration does not require particular
settings, you do not need to load a config file manually or manipulate the existing one. At each program
start, the default configuration is executed. You can modify this file to load your settings at the program
start if you have your own default settings. You can find and manipulate the default settings here
"…\AppData\Roaming\Image Engineering\GEOCAL".
The config file parameters are listed below if you need to make adjustments.
•Device Parameter: Wavelength
This parameter describes the wavelength of the built-in laser.
•Device Parameter: GratingConstant
This parameter describes the grating constant of the built-in DOE.
•Control Parameter: MinF
This parameter describes the stop strategy, at which optimization value the analysis is stopped
(0.0…10.0 – default: 1.0)
The smaller the value, the more accurate the analysis becomes, but the analysis time also
increases.
•Control Parameter: MinDelta
This parameter describes another stop strategy. When the change of the optimization value
between two iterations is less than the delta, the analysis stops (0.0...10.0 - default: 1e-7).
The smaller the value, the more accurate the analysis becomes, but the analysis time also
increases.
•Detection Parameter: Margin
This parameter is used to set a detection margin. It may happen that points at the edge of the
image cannot be adequately detected. With this parameter, you can ignore the edge.
•Distortion Model: Model
With this parameter, the model can be selected. Currently, the following models are implemented:
- EVEN_BROWN_MODEL, CUSTOM_MODEL_1 (for more information about the models,
please review the API documentation).
•Analysis Parameter: angles Alpha and Beta
These angles describe the orientation of the diffractive optical element to the incident expanded
and collimated beam of light. They are determined during the acceptance procedure at Image
Engineering and written in the config file. These values may not be changed unless the diffractive
optical element has to be replaced.

Image Engineering
Seite 11 von 19
Nevertheless, they can, for example, be set to zero for troubleshooting.
•Analysis Parameter: angles Omega (roll), Phi (pitch), Kappa (yaw)
These angles describe the device's orientation under test to the diffractive optical element.
These values are determined during calibration and output as results. Nevertheless, they can be
set to a fixed value for troubleshooting.
•Analysis Parameter: Radial Distortion Coefficients 1-5
The radial distortion coefficients model this type of distortion. The distorted points are denoted as:
xdistorted = x(1 + k1*r2+ k2*r4+ k3*r6+ k4*r8+ k5*r10)
ydistorted= y(1 + k1*r2+ k2*r4+ k3*r6+ k4*r8+ k5*r10)
These values are determined during calibration and output as results. However, they can be set
to a fixed value for troubleshooting.
•Analysis Parameter: Tangential Distortion Coefficients 1+2
The tangential distortion coefficients model this type of distortion. The distorted points are
denoted as:
xdistorted = x + [2p1xy + p2(r² + 2x²)]
ydistorted = y + [p1(r² + 2y²) + 2p2xy]
These values are determined during calibration and output as results. However, they can be set
to a fixed value for troubleshooting.
•Analysis Parameter: Focal Length Xand Y
The focal length values of the device under test in X and Y are determined during calibration and
output as results in pixel dimensions. Nevertheless, they can be set to a fixed value for
troubleshooting.
•Analysis Parameter: Skew coefficient
The skew coefficient is set to 0 by default and is non-zero if the image axes are not
perpendicular.
•Analysis Parameter: Principal Point
The principal point (optical center) describes the point where the optical axis of the camera lens
intersects the image sensor. It is determined during calibration and output as a result.
Nevertheless, it can be set to a fixed value for troubleshooting.
To make changes to the standard config file, open it in a text editor. It is advised to make a copy of the
standard version before editing it.
Version:
<Configuration Version="3"Description="Translation">
Installing the latest software always installs the latest configuration file. Make sure that you always use
the latest version. The latest version of the configuration file is "2".
Editing Control Parameters:
Example:
<MinF Use="false">0.1</MinF>
The "Use" flag indicates if the following value is used for the calibration procedure. If it says "false," it is
not used, and the standard value is used instead. To use a custom value, write "true" instead of "false"
and change the value ("0.1" in this example) as desired. The new value will be used in every calibration
until you set the "Use" flag back to "false" again.

Image Engineering
Seite 12 von 19
Editing Analysis Parameters:
Example:
<Alpha IsInvariant="false" UseAsStartValue ="false" Unit="Degree">0</Alpha>
The "IsoInvariant" flag indicates if the following value is calculated during calibration or set to a fixed
value. If it is set to "false," the value is calculated in each calibration. To set it to a fixed value, write "true"
instead of "false" and change the value ("0" in this example) as desired. The new value will be used in
every calibration until you set the "IsInvariant" flag back to "false" again.
With the "UseAsStartValue" flag, you can set a start value for the analysis. If you set the flag to "true," the
following value will be used as the start value. If you set the flag to "false," the default value will be used.
The default is always "0," except for the focal length and principal point. For the focal length, an estimate
is automatically made after the detection. For the principal point, the center of the image is taken as
default.
If the "IsInvariant" flag is set to "true", the "UseAsStartValue" flag is ignored.
Loading a config file:
Click File →Load configuration (ctrl+c) →navigate to the folder containing the file →select the file and
click "open."
Loading images for analysis
Click File →Load images (ctrl+o) →navigate to the folder containing the images →select one or multiple
images and click "open."
In its current state, the software supports the analysis of 8bit and 16bit images.
The following file types are supported (debayered images): .TIFF, .JPG, .PNG
Images for analysis must be taken so that the point grid fills the sensor completely.
5.3.1 Delete images
Right-click on the selected image in the list to open a delete option.

Image Engineering
Seite 13 von 19
Image analysis
•Image selection: Select an image from the list in the "Image data" section (1). The image will
then be visualized in the "Image" section (4) under the "Image" tab. You can zoom into and out of
the image using your mouse wheel. The zoom will be centered around the position of your cursor.
•Point detection: Detect the points in the image by pressing the "Detect" button in the "Detection
& analysis" (2) section.
Detection is finished when the progress bar at the bottom of the window (5) reaches 100%. By
pressing "Show Detected Grid" next to the progress bar, you can display a visualization of the
detected/undetected points.
For a successful analysis, detecting all of the points is not necessary.
If auto-detection fails, you can make a manual ROI selection by setting the checkmark in the
"Detection & analysis" section (2). Five preset ROIs will then appear in the visualized image.
The size and position of these ROIs are adjustable by click&drag.
One ROI must be positioned on the 0th diffraction order (the brightest point in the center of the
grid). Adjust the ROI size so that it contains only one point.
The four remaining ROIs must be positioned on the four points closest to the 0th diffraction order,
Figure 7: Software user interface overview

Image Engineering
Seite 14 von 19
i.e., the one above, below, and to the left and right. These four ROIs may also only contain one
point. Furthermore, the individual ROIs should not overlap too much. See Figure 8 below for
reference.
•Analysis: After detection, the actual analysis can be performed. Click on the "Analyze" button in
the "Detection & analysis" section (2) to start the calibration. The progress bar (5) indicates when
the calibration is finished.
Depending on the number of points detected, the time needed for calibration may vary slightly.
•Calibration results: the results will be listed under the "Result table" tab in the "Image" section
(4). Please find an explanation of the values in chapter 4.2 of this manual. The RMSE of the
calibration is also included in the result table. If needed, you can clear the result table by clicking
"clear table."
Under the tab "Distortion curve," you also get a graphical visualization of the determined
distortion of the camera (Geometric Distortion vs. Field).
Saving results
You have the option to either save your results to an XML or CSV-type file. If you have analyzed multiple
images, select one or multiple images of which you would like to save the results from the list in the
"Image data" section.
Then click File →Save result →Save XML or Save CSV →navigate to the desired location →click Save
If you selected multiple images, all results would be saved in one CSV or XML file.
Figure 8: Manual point detection

Image Engineering
Seite 15 von 19
Export Grids
You can export three different grids, which are saved in CSV files. For each exported grid, two files are
saved, one for the x and one for the y coordinates.
To export a grid, click File →Save grid →choose the desired grid →navigate to the desired location →
click Save.
Available grids:
- Detected grid: Describes the grid type containing the detected light points.
- Reprojected grid: Describes the grid type that contains the reprojected light points used to
determine the distortion.
- Undistorted reference grid: Describes the grid type, which contains the points of the reference
grid without lens distortion. This grid includes only distortion given by the DOE physics and
the determined extrinsic and intrinsic parameters.
Plotting
There are several plots available:
5.7.1 Grid view
After a successful analysis, the detected and reprojected points are displayed on the original image. To
see them better, you can zoom into the image. The points have a size of one pixel.
5.7.2 Distortion curve
The curve is calculated from the distortion coefficients and shows the grad of the analyzed distortion.
5.7.3 Quiver plot
The quiver plot shows the direction of deviation. The respective arrows indicate how the detected points
are located concerning the reprojected points.
5.7.4 2d Color scatter
This plot shows a grid of color-coded points, with each point representing the distance ratio between
detected and reprojected points. The scalar on the right side shows the individual color coding.
5.7.5 Histogram
Two different histograms are shown. The first histogram contains the successfully detected diffraction
points dependent on the radius. The values start at zero (principal point) and run along the image
diagonal to the image corner (one).
The second histogram contains the successfully detected diffraction points dependent on the angle. The
values start from a straight line defined from the principal point to the right edge of the image. The angles
increase counterclockwise.
Logging
The browser displays logging data for each action in the "Logging" section of the user interface. If errors
occur, this can be helpful. The software also has a status bar at the bottom, where error messages from
the API are displayed directly.

Image Engineering
Seite 16 von 19
Quit
To exit the software, click File → Quit (ctrl+q) or simply close the window.
TRADEMARK AND COPYRIGHT
Trademarks
Windows is a registered trademark of Microsoft Corp.,
Copyright Information
See separate Terms and Conditions document.

Image Engineering
Seite 17 von 19
6 ADDITIONAL INFORMATION
Disposal instructions
After the service life of GEOCAL / GEOCAL XL, it must be disposed of properly. Electrical and
electromechanical components are included in GEOCAL / GEOCAL XL. Observe all national regulations.
Ensure that third parties cannot use GEOCAL / GEOCAL XL after disposing of it.
Contact Image Engineering if assistance for disposal is required.
Power circuit modification for production line use
If you plan to use your GEOCAL / GEOCAL XL on a production line, we offer the option of modifying the
unit to eliminate the need to press the switch to turn it on and off. After the modification, moving the
switch to the "ON" position is sufficient. The unit will immediately start operating when power is supplied
without having to press the switch again.
For more information on this option, please contact Image Engineering.

Image Engineering
Seite 18 von 19
7 DATA SHEET
Overview
Product name
GEOCAL / GEOCAL XL
Principle
DOE-based geometric calibration of digital cameras
Features
7.2.1 Hardware
Diffractive Optical
Element (DOE)
Generates a very evenly distributed grid of light points,
virtually originating from infinity
Output window
GEOCAL: Usable aperture: Ø 77 mm
GEOCAL XL: Usable aperture: Ø 155mm
(camera lens needs to have an equal or smaller diameter)
Usable FoV
Approx. 30 – 120° (more extreme values need to be tested)
Dimensions (l x w x h)
GEOCAL: approx. 575 mm x 144 mm x 170 mm
GEOCAL XL: approx. 850 mm x 244 mm x 270 mm
Mounting points
GEOCAL: 3 x M5x0,8 tapped holes in the base plate
GEOCAL XL: 10 x M5x0,8 tapped holes in the base plate
7.2.2 Illumination (CAUTION: DO NOT LOOK DIRECTLY INTO THE LIGHT SOURCE!)
Light source
Frequency-stabilized diode laser
Wavelength
633 nm
Output power
5 mW
Laser Class (diode only)
3B
Laser Class (device as
sold)
1M
Lifetime
> 10.000h
Warm-up time
Not required
7.2.3 Software
System requirements
PC with Windows 7 operating system (or higher)
USB port
Functions
•Load multiple images
•View selected image
•Perform calibration
•Overlay detected point grid
•Various tools for result visualization
•Export results (CSV and XML)
Output data
Camera intrinsic and extrinsic data, the orientation of DOE

Image Engineering
Seite 19 von 19
API (C++)
Available as a separate option
General description hardware
Power supply/consumption
25W 5V/5A / Pmax = 2W
Ports
USB type B
Weight
GEOCAL: approx. 4.5 kg
GEOCAL XL: approx. 10 kg
Operating conditions
15 - 35°C
Requirements on the device under test (DUT)
Max. dimensions
GEOCAL: max. diameter of the camera lens: 77 mm
GEOCAL XL: max. diameter of the camera lens: 155mm
Usable FoV
Approx. 30 – 180° (deviating values will have to be tested)