2003-2012 MVTec Software GmbH.



Similar documents
Industrial Vision Days 2012 Making Cameras Smarter: FPGA Based Image Pre-processing Unleashed

Solution Guide III-C. 3D Vision. Building Vision for Business. MVTec Software GmbH

ni.com/vision NI Vision

Intelligent Flexible Automation

CHAPTER I INTRODUCTION

Go Faster - Preprocessing Using FPGA, CPU, GPU. Dipl.-Ing. (FH) Bjoern Rudde Image Acquisition Development STEMMER IMAGING

the power of machine vision

How To Program With Adaptive Vision Studio

VeriSens XC Complexity needs Simplicity

Basler. Area Scan Cameras

a powerful graphical developing tool for prototyping image processing tasks

Basler scout AREA SCAN CAMERAS

Overview. Proven Image Quality and Easy to Use Without a Frame Grabber. Your benefits include:

MACHINE VISION MNEMONICS, INC. 102 Gaither Drive, Suite 4 Mount Laurel, NJ USA

Introduction to GPU Computing

Basler. Line Scan Cameras

Basler pilot AREA SCAN CAMERAS

Basler beat AREA SCAN CAMERAS. High-resolution 12 MP cameras with global shutter

Shape Measurement of a Sewer Pipe. Using a Mobile Robot with Computer Vision

Parallel Computing: Strategies and Implications. Dori Exterman CTO IncrediBuild.

Optical Digitizing by ATOS for Press Parts and Tools

Computer Graphics Hardware An Overview

Unit 1: INTRODUCTION TO ADVANCED ROBOTIC DESIGN & ENGINEERING

MeshLab and Arc3D: Photo-Reconstruction and Processing of 3D meshes

Application Example: Quality Control of Injection-Molded Parts

LIST OF CONTENTS CHAPTER CONTENT PAGE DECLARATION DEDICATION ACKNOWLEDGEMENTS ABSTRACT ABSTRAK

How does the Kinect work? John MacCormick

Removing Moving Objects from Point Cloud Scenes

Medical Image Processing on the GPU. Past, Present and Future. Anders Eklund, PhD Virginia Tech Carilion Research Institute

T-REDSPEED White paper

Computing at the HL-LHC

Application Example: Quality control of sheet metal components

Basler racer. Line Scan Cameras. Next generation CMOS sensors with 2k to 12k resolution and up to 80 khz line rate

Virtual Environments - Basics -

VECTORAL IMAGING THE NEW DIRECTION IN AUTOMATED OPTICAL INSPECTION

3D Vision An enabling Technology for Advanced Driver Assistance and Autonomous Offroad Driving

Kapitel 12. 3D Television Based on a Stereoscopic View Synthesis Approach

Next Generation GPU Architecture Code-named Fermi

Professor, D.Sc. (Tech.) Eugene Kovshov MSTU «STANKIN», Moscow, Russia

RiMONITOR. Monitoring Software. for RIEGL VZ-Line Laser Scanners. Ri Software. visit our website Preliminary Data Sheet

Rapid Application Development for Machine Vision A New Approach

3. NUMBER OF PARTICIPANTS TO BE ENROLLED

Limitations of Human Vision. What is computer vision? What is computer vision (cont d)?

Brochure More information from

Programming models for heterogeneous computing. Manuel Ujaldón Nvidia CUDA Fellow and A/Prof. Computer Architecture Department University of Malaga

Manufacturing Process and Cost Estimation through Process Detection by Applying Image Processing Technique

Scope of operation and highlight of the microenable GigE Vision frame grabber family

REAL-TIME STREAMING ANALYTICS DATA IN, ACTION OUT

Graphics Cards and Graphics Processing Units. Ben Johnstone Russ Martin November 15, 2011

Masters in Human Computer Interaction

A Prototype For Eye-Gaze Corrected

Applications to Computational Financial and GPU Computing. May 16th. Dr. Daniel Egloff

Four Keys to Successful Multicore Optimization for Machine Vision. White Paper

The Flat Shape Everything around us is shaped

3D SCANNING: A NEW APPROACH TOWARDS MODEL DEVELOPMENT IN ADVANCED MANUFACTURING SYSTEM

1. INTRODUCTION Graphics 2

3D Vision An enabling Technology for Advanced Driver Assistance and Autonomous Offroad Driving

WHITEPAPER. Image processing in the age of 10 Gigabit Ethernet

White Paper. "See" what is important

Colorado School of Mines Computer Vision Professor William Hoff

Jennifer Salerno Campus Relationship Manager Credit Suisse Campus Recruiting

WEST JEFFERSON HILLS SCHOOL DISTRICT THOMAS JEFFERSON HIGH SCHOOL DIGITAL DESIGN GRADES

Introduction to Computer Graphics

High speed 3D capture for Configuration Management DOE SBIR Phase II Paul Banks

Differentiation of 3D scanners and their positioning method when applied to pipeline integrity

Petascale Visualization: Approaches and Initial Results

GeoImaging Accelerator Pansharp Test Results

How To Use Trackeye

ZEISS T-SCAN Automated / COMET Automated 3D Digitizing - Laserscanning / Fringe Projection Automated solutions for efficient 3D data capture

Alphacam Art combines Vectric s Aspire artistic design software with the market leading Alphacam manufacturing software.

HPC with Multicore and GPUs

Point Cloud Simulation & Applications Maurice Fallon

Graphical Processing Units to Accelerate Orthorectification, Atmospheric Correction and Transformations for Big Data

Automotive Applications of 3D Laser Scanning Introduction

Computer Vision. Image acquisition. 25 August Copyright by NHL Hogeschool and Van de Loosdrecht Machine Vision BV All rights reserved

Shearography in Quality Assurance and In- Field Service

BUILDING TELEPRESENCE SYSTEMS: Translating Science Fiction Ideas into Reality

Consolidated Visualization of Enormous 3D Scan Point Clouds with Scanopy

AUTOMOTIVE EXCELLENCE

New Features in TerraScan. Arttu Soininen Software developer Terrasolid Ltd

Success Story. Pharmaceutical tablet packing and inspection

Application Example: Reverse Engineering

FSI Machine Vision Training Programs

What is An Introduction

Introduction. C 2009 John Wiley & Sons, Ltd

DAMAGED ROAD TUNNEL LASER SCANNER SURVEY

Alberto Corrales-García, Rafael Rodríguez-Sánchez, José Luis Martínez, Gerardo Fernández-Escribano, José M. Claver and José Luis Sánchez

Wii Remote Calibration Using the Sensor Bar

LONG BEACH CITY COLLEGE MEMORANDUM

How To Get A Computer Science Degree

pb tec solutions GmbH, Max-Planck-Str. 11, Alzenau (Germany) Tel.: Fax:

Introducing PgOpenCL A New PostgreSQL Procedural Language Unlocking the Power of the GPU! By Tim Child

Automatic 3D Mapping for Infrared Image Analysis

Transcription:

1 MVTec Software GmbH is a leading international manufacturer of software for machine vision used in all demanding areas of imaging: semi-conductor industry, web inspection, quality control and inspection applications in general, medicine, surveillance etc. MVTec's innovative work is driven by a commitment to be the number one supplier for sophisticated technologies in machine vision. MVTec is engaged in sponsoring various activities in universities, thus participating in the challenging process of understanding how machines can be taught to see. HALCON is the comprehensive standard software for machine vision with an integrated development environment (IDE) that is used worldwide. It leads to cost savings and improved time to market: HALCON's flexible architecture facilitates rapid development of machine vision, medical imaging, and image analysis applications. HALCON provides outstanding performance and a comprehensive support of multi-core platforms, MMX and SSE2, as well as GPU acceleration. It serves all industries with a library of more than 1800 operators for blob analysis, morphology, matching, measuring, identification, and 3D vision, to name just a few. HALCON secures your investment by supporting a wide range of operating systems and providing interfaces to hundreds of industrial cameras and frame grabbers, also for standards like GenICam, GigE Vision, and IIDC 1394.

2 Go on and be a Halconist.

3

4 Cutting-edge performance in 3D vision 3D vision is revolutionizing machine vision. 3D information helps machines and robots to see and decide what to do. 3D vision technologies are getting more and more important for industrial production, robotics, and medical applications. HALCON provides all technologies needed to implement almost any conceivable 3D vision application. To name just a few, HALCON 11 now offers new features like 3D surface comparison, 3D object processing, and improved photometric stereo for advanced surface inspection. HALCON 11 sets new standards in 3D vision technology and provides state-of-the-art image processing algorithms.

5 For surface comparison a reference 3D object model is needed. It can be generated from CAD data or from reconstructed 3D data. Because 3D sensor data (or data that was reconstructed by software, e.g. with stereo) is often noisy and incomplete, the data has to be pre-processed, e.g., by registering and unifying many views to a single object. During the online phase, the test objects are acquired by a 3D sensor (or by reconstruction). After that, they have to be aligned to the reference object, e.g., by surface-based 3D matching. After the alignment, the objects are compared by calculating the distance between reference and test object with distance_object_model_3d. Points that have a too large distance are candidates for defects.

6 The new 3D surface comparison between expected and measured shape of a 3D object surface is an outstanding technology of HALCON 11. The surface can be reconstructed by any 3D technology available in HALCON like multi-view stereo, sheet of light, or by ready-to-run 3D hardware scanners which are also directly supported by HALCON. HALCON's 3D surface comparison evaluates the resulting point cloud with the trained object model. HALCON 11 controls the quality of an object's shape and thus lets you enter new markets.

7 The new 3D surface comparison between expected and measured shape of a 3D object surface is an outstanding technology of HALCON 11. The surface can be reconstructed by any 3D technology available in HALCON like multi-view stereo, sheet of light, or by ready-to-run 3D hardware scanners which are also directly supported by HALCON. HALCON's 3D surface comparison evaluates the resulting point cloud with the trained object model. HALCON 11 controls the quality of an object's shape and thus lets you enter new markets.

8 The new 3D surface comparison between expected and measured shape of a 3D object surface is an outstanding technology of HALCON 11. The surface can be reconstructed by any 3D technology available in HALCON like multi-view stereo, sheet of light, or by ready-to-run 3D hardware scanners which are also directly supported by HALCON. HALCON's 3D surface comparison evaluates the resulting point cloud with the trained object model. HALCON 11 controls the quality of an object's shape and thus lets you enter new markets.

9 Supersonic Machine Vision It is unbelievable how fast software can be. HALCON's automatic operator parallelization (AOP) is a unique feature. Let HALCON automatically split your data into multiple threads running on all available processing cores and automatically merge the output into one result. HALCON 11 supports GPU processing based on the OpenCL standard. With HALCON 11, over 75 operators are currently supported on GPUs. That's more than any other software package on the market. Moreover, depth from focus (DFF), Fast Fourier Transformation (FFT), and HALCON's local deformable matching have been significantly accelerated. HALCON 11 lets you feel the speed in machine vision.

10 Supersonic Machine Vision It is unbelievable how fast software can be. HALCON's automatic operator parallelization (AOP) is a unique feature. Let HALCON automatically split your data into multiple threads running on all available processing cores and automatically merge the output into one result. HALCON 11 supports GPU processing based on the OpenCL standard. With HALCON 11, over 75 operators are currently supported on GPUs. That's more than any other software package on the market. Moreover, depth from focus (DFF), Fast Fourier Transformation (FFT), and HALCON's local deformable matching have been significantly accelerated. HALCON 11 lets you feel the speed in machine vision.

11 Sample-based Identification Becomes Reality Imagine your machine can identify any product only by sight without any bar or data code. HALCON 11 makes it true. HALCON 11 provides its unique sample-based identification. You only have to teach HALCON 11 your objects from a few different directions, and the software identifies similar to a human being your objects in any pose. This even works with warped objects or varying perspective views. No other software provides a similar technology. With HALCON 11, genuine object identification becomes reality.

12 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

13 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

14 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

15 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets. The data set consists of 20 different pages. The training is applied with PDF images, converted to TIFF. For each page only a single training with the synthetic model was applied.

16 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets. The data set consists of 14 different vegetables and fruits. For each object group only one training sample was used.

17 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets. The data set consists of 59 different paintings. For each painting only one training sample was used.

18 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

19 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

20 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

21 With HALCON 11, genuine object identification becomes reality. Sample-based identification is capable to differentiate a large number of objects. This technology can recognize trained objects only based on characteristic features like color or texture, thereby eliminating the need to use special imprints like bar codes or data codes for object identification purposes. This even works with warped objects or varying perspective views of the object. An additional option is to learn an object from any side by using samples showing all relevant views. No other software provides a similar technology. HALCON 11 lets you enter new identification markets.

22 Go on and be a Halconist.