AT logo - The Machine Vision Specialists  
  Frame Grabbers  
  Vision Systems  
  Contact AT  
  Site Map  
Request product information


(08) 9242 5411

(02) 9905 5551

(03) 9384 1775

Defence Recognised Supplier Scheme Logo


Back to MVTec
Back to Machine Vision Software

What is HALCON?

Apply cutting-edge, deep-learning algorithms to your application with HALCON's ability to train Convolutional Neural Networks. After training, the network can be used to classify new data. Typical application areas where this deep learning technology is in the field of defect classification (circuit boards, bottle mouths, or pills), or object classification (identifying the species of a plant from one single image). Handcrafting of features is no longer necessary with Deep Learning.

Dubbed "the power behind machine vision", HALCON is the fast, comprehensive and powerful software for all demanding areas of machine vision applications such as position recognition, object identification, fault-detection, code reading, print quality inspection, surface inspection, remote sensing and aerial imaging, and all complex 3D-vision tasks.

HALCON's powerful library includes a wide variety of operators and also provides interfaces to hundreds of cameras and frame grabbers. Additionally, HALCON secures your investment by supporting most common standards, operating systems and programming languages.

Quick and efficient building of imaging solutions is possible with HALCON'S highly-interactive and integrated development environment saving costs and improving time-to-market.

Key Attributes
Enabler of the "Industrial Internet of Things" Imaging library of more than 2 000 operators
Deep-learning algorithms (HALCON Progress) Easy programming in C, C++, C# and VB. NET
Supports multiple platforms, hundreds of cameras/grabbers & common vision standards High performance via state-of-the-art instruction sets and GPU Acceleration
Key Features
Optimised for industrial imaging applications Matching to find even rotated or partly occluded objects

Blob analysis with more than 50 shape and gray-value features

High-accuracy measuring
Optical character recognition and verification
Arbitrarily shaped ROIs for significant flexibility and speed
Detection of lines, circles, ellipses with accuracy up to 1/50 pixel Extremely fast morphology Colour image processing and hyperspectral imaging
Processing of extremely large images (more than 32k x 32k) Image sequence processing (e.g., for surveillance tasks) Revolutionary software for 3D machine vision

HALCON Progress 17.12 - Deep learning, out-of-the-box capability
HALCON 13 - Professional software for all machine vision applications
Available through annual subscription, HALCON offers continuous development, improvements and maintenance with new releases approximately every six months This powerful machine vision software suits all types of applications. HALCON 13 not only masters complex 3D machine vision tasks but also offers supplier's experience, a comprehensive set of technologies and the flexibility of your choice of hardware.

Benefits and key characteristics
Optimised for short-time-to-market
HALCON comes with pretrained networks that are highly optimised for short-time-to-market, by providing a high-speed training process and retrieval of a network. These pretrained networks have been selected to provide the best foundational data set resulting in outstanding classification results.
3D matching and 3D reconstruction
HALCON views information simultaneously from multiple cameras leading to a more robust reconstruction than was possible using previous stereo methods.
Using Deep Learning
Having provided HALCON with labeled training images, training the (Convolutional Neural Network (CNN) for a specific task is easy. Once the CNN has learned the features of a classification, the trained CNN can be applied to new image data which the network then matches to the classes ("inference").
Direct debugging
HALCON's HDevEngine applications can now be de­bugged directly, making error tracking, and even remote error-tracking, much easier.
Automatic feature extraction
During training, HALCON automatically learns which features can be used to identify the desired classes and there is no need for skilled engineers to programme various classification methods.
Robust Identification Technologies
New deep-learning technology achieves higher reading rates than with previous classification methods. HALCON 13's automatic text reader is faster and supports reading of dot print characters and barcodes even if large parts of the code are defective or not visible. Dealing with blur and distortion is easier.
Powerful image classification with inference
The trained network is able to infer the corresponding class, and applications using inference can analyse large quantities of image data that belong to classes that are hard to distinguish. HALCON's Deep Learning capacity goes beyond the human capacity to distinguish classification
Easy texture inspection
Traditionally one the the more challenging machine vision tasks, dealing with an objects different characteristics has proved problematic. HALCON 13 offers an easy-to-use texture inspection, that automatically identifies defects by comparison with an image of flawless materials
Seamless integration into the powerful HALCON library
Seamlessly integrated into the HALCON library, Deep Learning blends with the comprehensive selection of available operators, providing maximum flexibility while building applications.
HALCON 13 also offers significant speedups for all related technologies, i.e., shape-based 3D matching, local and perspective deformable matching, and component-based matching.






If you like this page, please recommend and share it.

Google+ Facebook Twitter More