Scorpion Vision Training in Oslo 11. – 13. October 2017

Fly in to the best machine vision training and learn from the experts behind Scorpion Vision Software® and Scorpion 3D Stinger.

The next generation
Scorpion 3D Stinger Scanner – 2MP – 3500 fps
and
the Scorpion 3D Venom Camera – 1 micron resolution
will be presented for the first time

Scorpion Vision Family - 1000px

Scorpion 3D Stinger Family

Learn more about the cutting-edge Scorpion Compact Vision systems combined with our impressive range of industrial cameras:
Scorpion Box, Scorpion 2D Stinger,
Scorpion 3D Venom and Scorpion 3D Stinger Cameras.

Attend one, two or all three days.

Invitation – Scorpion Vision Training Course Oslo October 2017

Sep15_Tordivel

The course is based on the newest Scorpion Vision Software® release XII including 3D Stereo Vision, Dense 3D Image generation and 3D Robot Vision.

Video : Thor Vollset – Tordivel presents 3D Stereo Vision and Vision Guided Robots

After its positive première in 2014, the inspect application forum followed up with a successful and increasingly international event on the Vision in Stuttgart this year. Altogether, more than 350 visitors to the fair, half of them from abroad, took the opportunity to attend the diverse 30-minute talks and gain detailed insights into the current application and development possibilities in machine vision.

thorvollset_ceo

Check the video at Inspect

3D Stereo Vision and Vision Guided Robots

Under the motto “Vison 4.0 – Smart Vision for Smart Factories,” innovative systems and solutions were presented from the fields of Best Practice, 3D Inspection, Automotive, Robotics, Virtual Reality, Food & Beverage and Embedded Vision.

inspect

Tordivel – We are hiring in Oslo, Norway

cones

We are growing fast with profit need to hire good people to meet the market demand for Scorpion Vision Software and Scorpion 3D Stinger Technology.

Tordivel is a Nordic leader in 2D and 3D machine vision and a partner for minor and major companies providing cutting-edge Scorpion 3D Stinger Products, Scorpion Vision Solutions and Scorpion Stinger OEM Services.

We have the following position open for application:

scorpionsalesengineer

seniordeveloper

automationvisionengineer

Company

Tordivel is a Nordic leader in 2D and 3D machine vision and a partner for minor and major companies providing cutting-edge Scorpion 3D Stinger Products, Scorpion Vision Solutions and Scorpion Stinger OEM Services.

is-2016-0010 Scorpion 3D Stinger for Robot Vision - VIII

Tordivel has an extensive and global network of partners and distributors around the world with a head office in Oslo, Norway. Scorpion Vision Ltd is our subsidariry located in Southmapton, United Kingdom.

In 1996 a team of vision and image processing specialists was established to provide state-of the-art vision solutions. From 1997 to 2000 Tordivel developed and deployed a large number of inline advanced vision systems for mobile telephone inspection in Norway and Sweden.

In 2000 we created the unique Scorpion Vision Software®. The Scorpion Vision Framework has since been the basis for thousands of machine vision systems and Scorpion Vision OEM solutions. In 2006 we decided to develop the best easy-to-use 3D Machine Vision Framework.

From 2006 to 2011 we participated in a series of large R&D projects within 3D Machine Vision. In 2011 we invented the unique Scorpion 3D Stinger Camera – today we market products and services based on 3D Stereo Vision with our unique Scorpion 3D Stinger Technology.

The Tordivel Group is growing fast with profits within OEMs, 2D and 3D machine vision software and 3D Robot Vision.

Visit us at Aluminium-2016 – 29.11-1.12

Visit us a Aluminium-2016 in Dusseldorf  where will present our new Scorpion 3D Stinger for the Aluminium Industry.

is-2016-0004-stub3dcollage

Scorpion 3D Stinger for Alumium Industry

Read more about Scorpion 3D Stinger for Stub Measurement

Dusseldorf 29. november to 1. december 2016

Scorpion 3D Stinger for the Aluminium Industry

Scorpion 3D Stinger for the Aluminium Industry is a proven, complete, cost effective and expandable platform for robust measurement, classification and tracking in Rodding Shops. The system consists of these application:

  • Scorpion 3D Stinger for Stub Measurement
  • Stinger 3D Stinger for Butt Measurement
  • Scorpion Stinger ID Reading

Tordivel has worked in the Aluminium Industry since 2004 and has an extensive set of reference systems with leading smelters around the world.

Tordivel AS cooperates with machine builders with experience with the aluminium industry and all technical details are based on Tordivel’s standard product is subject to change due to a continous improvement process.

Scorpion Vision XII – SMARTscript is Dynamic Configuration

scorpionxii-64x64In this post we describe another important element in the Scorpion Vision framework were we want to explain that

 Python SMARTscript is Dynamic Configuration

Scripting supplements the ease of point & click configuration to develop more accurate, robust and smarter 2D and 3D machine vision solutions.

This example, SDP-0147 DynamicResultPythonScript, uses an image of three valves. The image is from an old and simple 2D robot vision solution. The example is made to be an inspiration for Scorpion Vision developers.

is-2016-0020-precornerfilter_input

Image of three valves

The example shows how a simple script can do the following operations:

  • PreCornerFilter image using  the STC-0011-ImageFilter 4.1.0.52
  • FindCornerCandidates using the Blob4 tool
  • Display the filter image using a color palette.
  • Cluster the candidates using a smart python script
  • Display the clusters and the results using the STC-0013-DisplayResult 2.2.0.28
  • Create the ClusterArea for each Cluster – STC-0100-PolygonCreator2D 1.0.0.5
  • Display the ClusterArea
is-2016-0020-cluster_result_2

The updated PythonScript tool support dynamic results – the cluster results is shown

This is the 4th post in series of Scorpion Vision Software important tool improvements included in upcoming – Scorpion Vision XII.

Python makes Scorpion an Open Flexible Development Platform

Python is an integral part of Scorpion Vision Software. It makes Scorpion Vision an open, extendable and flexible development platform for the most advanced machine vision solutions.

Easy to import major Open-Source libraries like OpenCV 3, Numpy and SciPy.

is-2016-0020-cluster_opensource

The Zen of Python (PEP 20), which includes aphorisms such as:[45]

  • Beautiful is better than ugly
  • Explicit is better than implicit
  • Simple is better than complex
  • Complex is better than complicated
  • Readability counts

PreCornerFilter image using  the STC-0011-ImageFilter 4.1.0.52

The ImageFilter Scorpion Tool Component is very useful. The filter string does the following:

  • m3 – median filter image block size 3
  • p11,0.0005 – precornerfilter image with size 11 and scale 0.0005
  • X normalises the image
is-2016-0020-precornerfilter

The filter string to precorner filter an image

The image below shows the input image and the filtered image with a color palette.

is-2016-0020-pythonfilterresult

FindCornerCandidates using the Blob4 tool

The Blob4 tool locates the strongest peaks from the precorner filter to find corner candidates

is-2016-0020-blob4-intensitysort

New option to sort Blob4 result by intensity value

Cluster the candidates using a Smart Script

The core clustering uses a KDTree from Scipy to cluster all corner candidates. The script locates the largest cluster – removes repeatably the candidates in the largest cluster MaxIterations times and clusters again.

is-2016-0020-cluster_corescript_2

The clustering is fast – typical 1 ms

Display the clusters and the results using the STC-0013-DisplayResult 2.2.0.28

The DisplayResult displays the cluster string C1=(3,(315.7,334.2),11) in the image.

is-2016-0020-stc-dynamiclinking

The STC support %1i tool indexing that extract the index from the toolname Cluster1

Create the ClusterArea for each Cluster – STC-0100-PolygonCreator2D 1.0.0.5

The polygon creator uses the result from ClusterScript to create a polygon describing the cluser area.

is-2016-0020-polygoncreator

Display the ClusterArea

is-2016-0020-clusterarea

The Cluster Area shown – inside the Blob visualises all the corner candidates

Scorpion Tools in Action

The following Scorpion Tools and STCs, Scorpion Tool Components, are used in the demo profile SDP-0147 DynamicResultPythonScript – 1.0.0.5

  • PreCornerFilter – STC-0011-ImageFilter 4.1.0.52
  • FindCornerCandidates – Tool 87 – Blob4
  • ClusterScript – Tool 3 – PythonScript
  • ClusterN – STC-0013-DisplayResult 2.2.0.28
  • CircleN – STC-0100-PolygonCreator2D 1.0.0.5
is-2016-0020-cluster_script_2

The complete ClusterScript that defines the datamodel and performs clustering using the Spatial module in SciPy

STC-0124 ObjectFinder3D – Fast and robust 3D Pattern Matching

scorpionxii-64x64In this post we have used 3D images captured in the Auto3D research project. The image below captured with a stripelight projector to prove that a Battery Pack could be located in 3D and changed by a robot without human interaction on an offshore platform.

This is the third post in series of Scorpion Vision Software important tool improvements included in upcoming – Scorpion Vision XII.

is-2016-0019-valvepointcloud3d

The 3D image of the Battery pack is shown.

In the Auto3D project we developed a 3D Pattern Matcher together with SINTEF named 3DMama – this was one of the first 3D pattern matchers in 2009. We have now taken a big step forward developing a significant better 3D Pattern Matcher STC-0124 ObjectFinder3D replacing 3DMama. The 3D pattern matcher has the following important features:

  • Automatic model training and feature point selection and verification
  • User defined number of featurepoints
  • Locates multiple objects in 3D
  • Very fast and easy training process

The ObjectFinder3D is optimised for dense 3D points cloud generated by 3D Stereo Vision like the Scorpion 3D Stinger Camera.

3D Model to locate Battery Pack

The image below shows the Learning page  and the Battery Pack Model with 3D origin visualised.

is-2016-0019 OF3D-Template.PNG

STC-0124 ObjectFinder3D Learning Page

Fast and robust location of Battery Pack

The battery is located in a 3D point cloud with 360.000 points in 102.5 ms. The 3D bounding and the pose is shown in red. The 3D pose – (x,y,z) and (rx,ry,rz) is calculated.

is-2016-0019-of3d_locatebatterypack

Use 3D Pose to verify Top Cover Shape

Once locating the battery pack it is easy to verify that all cover knobs are present. In 3D we can easily do the following with Scorpion Vision Software XII:

  • Count the knobs
  • Verify the 3D shape by
    • Height
    • Top Area
    • Volume
  • Check the rotation angle for each knob

All of these features makes it easy to verify the type of top cover. This important information can be used to select the right tool for replacing the battery pack or identification of the battery pack type to be replaced.

is-2016-0019-of3d-locateknob

Height Map of Top Cover where the six knobs are counted. The HeightMap is color code – position with no information is coded grey

Scorpion Tools in Action

The following Scorpion Tools and STCs, Scorpion Tool Components, are used in the demo profile SDP-0006 ObjectFinder3D.

  • 3DROI – STC-0094-ROICalculator 1.0.0.4
    • Defines the 3D Region of interest
  • ClusterFilter – STC-0111-ClusterFilter3D 1.0.0.22
    • Can filter noise from a 3D Point Cloud
  • FindValve – STC-0124 ObjectFinder3D 0.9.0.7
    • Locates the valve in 3D based on a 3D Model
  • LocateValvePlane – STC-0114 PlaneFit3D 0.9.0.16
    • Precise plane location of the top cover
  • HMapValve – Tool 122 – ExtractMap3D
    • Extract the HeightMap of the Top Cover perpendicular to the top cover plane
  • HeightHistogram –  Tool 75 – Intensity
    • Calculates the height map histogram
  • FindKnobs – Tool – 87 – Blob4
    • Locates the knobs in 3D in the heightmap

Calibrator2D – Automatic UltraWide angle lens calibration

scorpionxii-64x64This post is the first in series of Scorpion Vision Software tool improvements included in upcoming – Scorpion Vision XII.

Working with 3D Stereo Vision we have learned that completely removing lens distortion is extremely important – this require the best sub-pixel grid location algorithms. In 3D Robot Vision significant value is added by moving closer using wideangle lenses.

We also use this capability to make high precision 3D Stereo Vision system for our OEM customer using low cost lenses and cameras in the smallest form-factor.

is-2016-0016-ultrawidelensimage

Calibration Grid with Ultra Wide Lens

We are proud to state that we have added “Auto Grid Detection” that starts in the center of the images and locates all points that is located with proper quality. This means that Calibrator2D will use all visible grid points that are visible in the image – this will ensure almost 100 % coverage of the image.

is-2016-0016-settings-calibrator2d

Calibrator 2D – Auto Grid Settings

The Calibrator2D.Setup page shows the user feedback after calibrating this image. We see that only the corners are left uncalibrated – where the lens more or less sees to infinity 🙂

The average error is subpixel 1/5th of pixel. The maximum correction of a pixel is 48.9 pixels – the pixel size is 0.9 mm equal to 45 mm.

is-2016-0016-ultra-wide-calibrator2d