April 2014
Tim Seymour
Vision Application Specialist, SICK Australia
: Improving Quality & Profits with Machine Vision
: SICK AG 2
Machine Vision Systems
Contents
: What is Machine vision?
: When do we use it?
: What types of Cameras can be used in Industry
: Examples of Vision Solutions
: SICK AG
Machine Vision Systems
: When do we use it?
: The benefits of Machine Vision are;
: $$$$$$$$
- Improving quality of the finished product
- Simplifying, improving or increasing work flow,
- Reducing waste , both in materials and effort.
3
: SICK AG
Machine Vision Systems
: What is Machine Vision?
: Use of cameras in the context of process control
: Is defined as the analysis of images and extraction of data for controlling an
automated process or action
4
Image
capture
Image
analysis
Decision
making
Data
output
Repeat
process
: SICK AG
What is Machine Vision
: Image capture
: 1 D, 2 D or 3 Dimensions ?
: Illuminating an Object and acquiring an image
- Front or back light?
- Visible or Infra Red?
: Moving or stationary ?
: Triggered or Free running?
5
Image
captur
e
Image
analy
sis
Decisi
on
makin
g
Data
output
Repe
at
proce
ss
: SICK AG
Image analysis
: to answer the following questions in highly repetitive situations....
Tim Seymour April 2014 6
Good/bad
Classification
PositionIdentification Dimension
OK? Where?What does it say? How large?
: SICK AG
Decision Making and Data output
: What results do you need?
- Pass or Fail?
- Sort?
- Digital or Serial?
- Settings Adjustable or fixed?
: How much time do you have?
7
Image
capture
Image
analysis
Decision
making
Data
output
Repeat
process
: SICK AG
Types of Machine Vision Components
: Vision ”Sensors”
- Easy to use
- Application-specific, or configurable but limited
- Works stand-alone
: ”Smart” camera
- Flexible, programmable with ready-to-use
toolbox
- Works stand-alone, no PC
: PC-based camera ”components”
- Fully flexible, programmable with low-level
languages (C++, .NET, etc)
- The camera only takes images, the PC
analyzes and controls the process using a
visual studio software package.
8
Vision sensor or
smart camera
PC-based
: SICK AG
Some terms explained…
: What is a pixel?
9
Pixel
: SICK AG
Some terms explained…
: FOV?
▪ Field of View;
: ROI?
▪ region of interest….?
: TOF?
▪ Time of Flight
: OCR
▪ Optical character recognition
: RGB
▪ Colour cameras
: IR or UV
▪ Type of light emitted or received?
: GUI
▪ Graphical User Interface that appears on the Human Machine Interface
10
: SICK AG
Simple Vision Sensors
: Just connect and align
- Easy to use
- Single purpose
- No parameters
- No Logic
: Examples are
- X,Y positioners
- Shingle counters
- Registration detectors
11
: SICK AG
Simple solutions using smart sensors
: object counters
: how many containers can I fit on a conveyor?
12
: SICK AG
Simple solutions using smart sensors
: X,Y positioners
: the challenge is to align the pallet
handling system to the shelf or bay
location.
13
: SICK AG
Simple solutions using smart sensors
: Labels without marks kept in register
: The saving may only be 2%.... However…...
14
: SICK AG
Complex Vision Sensors
: The Features of a Complex Vision Sensor include;
- Around 10 to 20 Algorithms or Vision ”tools”
- Ethernet or USB connectivity
- Simple logic functions
- Discrete outputs
- Fast processing times
- Low to medium image resolution
- Integrated light sources
15
: SICK AG
Example Applications…
: Interactive Edge Guiding on carpet
gluing line; 2 x cameras for 2 different
materials running simultaneously.
16
: SICK AG
Example Applications for Vision sensors…
: APPLICATION; Ultra Violet Light source to inspect a optically challenging product
17
: SICK AG
Example Applications for Vision sensors…
: APPLICATION;
- Look for correct insertion of wafer in packaging, before dosing of ice cream
18
: SICK AG
Example Applications for Vision Sensors…
: APPLICATION: Cheese Sticks
: Formation checks
:
: TECHNOLOGY: 2D Vision Camera
: LED Diffuse Back light
:
: METHODOLOGY: Back Lit image for Length and shape check
:
19
Courtesy of
: SICK AG
“Smart” Vision Cameras
: The Features of a Smart Camera are;
- Around 200 Algorithms and logical processes
- Ethernet connectivity
- Complex logical and mathematical functions
- Multiple outputs
- Fast processing times
- Medium to high image resolution
- Usually 2D ( some 3D)
- Multiple lens options
20
: SICK AG
Example Applications for Smart Cameras…
: APPLICATION: Glass Bottle Rim & Neck validation
: Break, crack & foreign matter detection
:
: TECHNOLOGY: 2D Vision system with PLC control for ejection
: LED illumination, Operator HMI for sensitivity adjustment
:
: METHODOLOGY: Front Lit image
: Greyscale value mapping.
21
Courtesy of
: SICK AG
Example Applications for Smart Cameras…
: APPLICATION: Frozen Pie packets
: Packet formation & closure verification
:
: TECHNOLOGY: 2D Vision system
: LED illumination,
:
: METHODOLOGY: Back Lit image to silhouette packet to confirm edge formation
:
22
Courtesy of
: SICK AG
Example Applications for Smart Cameras…
: APPLICATION: Sanitary pads;
: Pad thickness, feature presence & location
:
: TECHNOLOGY: “Linescan” smart cameras
: LED illumination, Operator HMI for sensitivity adjustment
:
: METHODOLOGY: Silhouetted image
: Greyscale value mapping and measurement,
23
Courtesy of
: SICK AG
Example Applications for Smart Cameras…
: APPLICATION: Lid & tub label validation
: Correct ID & position, Internal tub defect detection
:
: TECHNOLOGY: 2D smart cameras with bar code reading tools
: LED illumination, Operator HMI
:
: METHODOLOGY: Front lit image
: Greyscale; 2D Code reader & Pattern match.
24
Courtesy of
: SICK AG
Camera Components
: High end Camera components;
- High resolution, 1, 2, & 3D
- Special Intefaces
▪ GigE, Camera Link, Firewire, etc
- Most analysis functions completed by seperate software in an external PC
- Fast imaging times
- Usually high image resolutions
- Multiple lens options
25
: SICK AG
Camera Components used with 3rd party software
: APPLICATION: Batch Mixing Verification;
: Scanning Powdered Ingredients Bags
:
: TECHNOLOGY: 2D camera
: LED illumination, Operator HMI
:
: METHODOLOGY: Front lighted image,
: Intuitive pattern matching linked to customer data base for
ingredient lists and quantities.
26
Courtesy of
: SICK AG
Camera Components used with 3rd party software
: APPLICATION: Detect “Off Centre Core”;
: “News Print” Paper rolls
:
: TECHNOLOGY: Hi res 2D camera
: Infra Red LED illumination, Operator HMI
:
: METHODOLOGY: Front lighted image,
: edge detection followed by mathematically calculations to
determine centre of roll vs centre of core. .
27
Courtesy of
: SICK AG
Camera Components used with 3rd party software
: APPLICATION: Print Inspection
: Foil sealed trays
: TECHNOLOGY: Hi res 2D camera
: Diffuse LED illumination, Operator HMI
: METHODOLOGY: Front lighted image,
: complex pattern matching software.
28
Courtesy of
Missing Label Extra Print Misregistration
Blob Damaged Smudge
: SICK AG
Camera Components
: APPLICATION: Parcel Sortation ( & dimensioning)
: freight logistics
: TECHNOLOGY: Line scan camera
: Fixed LED line illumination tracked to moving objects.
: METHODOLOGY: Image software in both the Camera and
attached PC.
29
Courtesy of
: SICK AG
When to use 3D?
: 3D using
: structured light?,
: Stereoscopic?,
: triangulation?
30
: SICK AG
Camera Components integrated to Machine controls
: APPLICATION: Robotic Bin Picking
: metal parts
: TECHNOLOGY: “scanning” 3D camera
: Moving laser line illumination.
: METHODOLOGY: Image software
compatible with standard CAD files and
Robot coordinate systems.
31
: Thank you for your attention.

Improving Quality & Profits with Vision

  • 1.
    April 2014 Tim Seymour VisionApplication Specialist, SICK Australia : Improving Quality & Profits with Machine Vision
  • 2.
    : SICK AG2 Machine Vision Systems Contents : What is Machine vision? : When do we use it? : What types of Cameras can be used in Industry : Examples of Vision Solutions
  • 3.
    : SICK AG MachineVision Systems : When do we use it? : The benefits of Machine Vision are; : $$$$$$$$ - Improving quality of the finished product - Simplifying, improving or increasing work flow, - Reducing waste , both in materials and effort. 3
  • 4.
    : SICK AG MachineVision Systems : What is Machine Vision? : Use of cameras in the context of process control : Is defined as the analysis of images and extraction of data for controlling an automated process or action 4 Image capture Image analysis Decision making Data output Repeat process
  • 5.
    : SICK AG Whatis Machine Vision : Image capture : 1 D, 2 D or 3 Dimensions ? : Illuminating an Object and acquiring an image - Front or back light? - Visible or Infra Red? : Moving or stationary ? : Triggered or Free running? 5 Image captur e Image analy sis Decisi on makin g Data output Repe at proce ss
  • 6.
    : SICK AG Imageanalysis : to answer the following questions in highly repetitive situations.... Tim Seymour April 2014 6 Good/bad Classification PositionIdentification Dimension OK? Where?What does it say? How large?
  • 7.
    : SICK AG DecisionMaking and Data output : What results do you need? - Pass or Fail? - Sort? - Digital or Serial? - Settings Adjustable or fixed? : How much time do you have? 7 Image capture Image analysis Decision making Data output Repeat process
  • 8.
    : SICK AG Typesof Machine Vision Components : Vision ”Sensors” - Easy to use - Application-specific, or configurable but limited - Works stand-alone : ”Smart” camera - Flexible, programmable with ready-to-use toolbox - Works stand-alone, no PC : PC-based camera ”components” - Fully flexible, programmable with low-level languages (C++, .NET, etc) - The camera only takes images, the PC analyzes and controls the process using a visual studio software package. 8 Vision sensor or smart camera PC-based
  • 9.
    : SICK AG Someterms explained… : What is a pixel? 9 Pixel
  • 10.
    : SICK AG Someterms explained… : FOV? ▪ Field of View; : ROI? ▪ region of interest….? : TOF? ▪ Time of Flight : OCR ▪ Optical character recognition : RGB ▪ Colour cameras : IR or UV ▪ Type of light emitted or received? : GUI ▪ Graphical User Interface that appears on the Human Machine Interface 10
  • 11.
    : SICK AG SimpleVision Sensors : Just connect and align - Easy to use - Single purpose - No parameters - No Logic : Examples are - X,Y positioners - Shingle counters - Registration detectors 11
  • 12.
    : SICK AG Simplesolutions using smart sensors : object counters : how many containers can I fit on a conveyor? 12
  • 13.
    : SICK AG Simplesolutions using smart sensors : X,Y positioners : the challenge is to align the pallet handling system to the shelf or bay location. 13
  • 14.
    : SICK AG Simplesolutions using smart sensors : Labels without marks kept in register : The saving may only be 2%.... However…... 14
  • 15.
    : SICK AG ComplexVision Sensors : The Features of a Complex Vision Sensor include; - Around 10 to 20 Algorithms or Vision ”tools” - Ethernet or USB connectivity - Simple logic functions - Discrete outputs - Fast processing times - Low to medium image resolution - Integrated light sources 15
  • 16.
    : SICK AG ExampleApplications… : Interactive Edge Guiding on carpet gluing line; 2 x cameras for 2 different materials running simultaneously. 16
  • 17.
    : SICK AG ExampleApplications for Vision sensors… : APPLICATION; Ultra Violet Light source to inspect a optically challenging product 17
  • 18.
    : SICK AG ExampleApplications for Vision sensors… : APPLICATION; - Look for correct insertion of wafer in packaging, before dosing of ice cream 18
  • 19.
    : SICK AG ExampleApplications for Vision Sensors… : APPLICATION: Cheese Sticks : Formation checks : : TECHNOLOGY: 2D Vision Camera : LED Diffuse Back light : : METHODOLOGY: Back Lit image for Length and shape check : 19 Courtesy of
  • 20.
    : SICK AG “Smart”Vision Cameras : The Features of a Smart Camera are; - Around 200 Algorithms and logical processes - Ethernet connectivity - Complex logical and mathematical functions - Multiple outputs - Fast processing times - Medium to high image resolution - Usually 2D ( some 3D) - Multiple lens options 20
  • 21.
    : SICK AG ExampleApplications for Smart Cameras… : APPLICATION: Glass Bottle Rim & Neck validation : Break, crack & foreign matter detection : : TECHNOLOGY: 2D Vision system with PLC control for ejection : LED illumination, Operator HMI for sensitivity adjustment : : METHODOLOGY: Front Lit image : Greyscale value mapping. 21 Courtesy of
  • 22.
    : SICK AG ExampleApplications for Smart Cameras… : APPLICATION: Frozen Pie packets : Packet formation & closure verification : : TECHNOLOGY: 2D Vision system : LED illumination, : : METHODOLOGY: Back Lit image to silhouette packet to confirm edge formation : 22 Courtesy of
  • 23.
    : SICK AG ExampleApplications for Smart Cameras… : APPLICATION: Sanitary pads; : Pad thickness, feature presence & location : : TECHNOLOGY: “Linescan” smart cameras : LED illumination, Operator HMI for sensitivity adjustment : : METHODOLOGY: Silhouetted image : Greyscale value mapping and measurement, 23 Courtesy of
  • 24.
    : SICK AG ExampleApplications for Smart Cameras… : APPLICATION: Lid & tub label validation : Correct ID & position, Internal tub defect detection : : TECHNOLOGY: 2D smart cameras with bar code reading tools : LED illumination, Operator HMI : : METHODOLOGY: Front lit image : Greyscale; 2D Code reader & Pattern match. 24 Courtesy of
  • 25.
    : SICK AG CameraComponents : High end Camera components; - High resolution, 1, 2, & 3D - Special Intefaces ▪ GigE, Camera Link, Firewire, etc - Most analysis functions completed by seperate software in an external PC - Fast imaging times - Usually high image resolutions - Multiple lens options 25
  • 26.
    : SICK AG CameraComponents used with 3rd party software : APPLICATION: Batch Mixing Verification; : Scanning Powdered Ingredients Bags : : TECHNOLOGY: 2D camera : LED illumination, Operator HMI : : METHODOLOGY: Front lighted image, : Intuitive pattern matching linked to customer data base for ingredient lists and quantities. 26 Courtesy of
  • 27.
    : SICK AG CameraComponents used with 3rd party software : APPLICATION: Detect “Off Centre Core”; : “News Print” Paper rolls : : TECHNOLOGY: Hi res 2D camera : Infra Red LED illumination, Operator HMI : : METHODOLOGY: Front lighted image, : edge detection followed by mathematically calculations to determine centre of roll vs centre of core. . 27 Courtesy of
  • 28.
    : SICK AG CameraComponents used with 3rd party software : APPLICATION: Print Inspection : Foil sealed trays : TECHNOLOGY: Hi res 2D camera : Diffuse LED illumination, Operator HMI : METHODOLOGY: Front lighted image, : complex pattern matching software. 28 Courtesy of Missing Label Extra Print Misregistration Blob Damaged Smudge
  • 29.
    : SICK AG CameraComponents : APPLICATION: Parcel Sortation ( & dimensioning) : freight logistics : TECHNOLOGY: Line scan camera : Fixed LED line illumination tracked to moving objects. : METHODOLOGY: Image software in both the Camera and attached PC. 29 Courtesy of
  • 30.
    : SICK AG Whento use 3D? : 3D using : structured light?, : Stereoscopic?, : triangulation? 30
  • 31.
    : SICK AG CameraComponents integrated to Machine controls : APPLICATION: Robotic Bin Picking : metal parts : TECHNOLOGY: “scanning” 3D camera : Moving laser line illumination. : METHODOLOGY: Image software compatible with standard CAD files and Robot coordinate systems. 31
  • 32.
    : Thank youfor your attention.