FEAT RE MACH N FEA ATURE


CHINEI E VISIONISION SYSTEM S


EMS THE FUTURE OF M CHI E VISION:


THE FUTURE OF MACHINE VI SION : DEVELOPMENTS WE WILL SEE I N 201 7


camera technology that is increasingly used in hardware and opens up entirely new dimensions formachine vision. The technology can be used to collect and process image information froma very broad electromagnetic spectrum. This paves the way formany application


scenarios in different sectors, for example the food and beverage industry.


CON NUOUS USABIL TY MPROVEMENTS


CONTIINUOUS USABILITY IIMPROVEMENTS


As the ‘eye of production’, machine vision plays an important role in industrial automation


M


achine vision will continue to develop rapidly in 2017 – also at


VISION, the world’s leading trade show for machine vision technology, which took place in Stuttgart, Germany last November, numerous trends were identified. The trade fair showed important trends, such as Industrial Internet of Things (IIoT) and Smart Factory, require faster, easier, and more robust machine vision processes. Further trends are 3D-based


technologies and identification methods such as optical character (OCR), deep-learning-ba


as well as bar and data code reading. With Halcon 13, the new release of its machine vision software, MVTec already offers improved OCR technique. By


implementing new, deep-learning-based pre-trained fonts and the corresponding classifier, the character recognition rate has been greatly improved.


In addition, the import ance o f


embedded technologies will also continue to grow. Embedded Vision makes it possible to evaluate images on compact, powerful, and energy-efficient


computing platforms. This further shifts the added value from the hardware to the software. Standard in software products for em


thus become more important.


Furthermore, a big challenge will be the comprehensive integration of two separate worlds:machin e vision an d programmable logic controller (PLC) technology. Working together with strong partners and committees, MVTec plays an important role in developing appropriate uniformstandards and interfaces for merging both technologies. Another trend can be seen in hyperspectral imaging, a


14 14 DECEMBER/JANUAR 201 ANUARY 2017 | AU AUTOMA MAT ATION


bedded systems dustrial


sed algorithms, recognition


Furthermore, machine vision


technologies will penetrate more and more industrial automation applications and thereby address a wider group of users. Therefore, it will become more important to improve machine vision usability and to make it even easier to create appropriate applications.


Dr. Maximilian Te


Lückenhaus, director marketing and sales, MVTec Software GmbH


marketing and sales, Lückenhaus, director Dr. Maximilian


ec Software GmbH


MVtec’s Merlic software is said to meet these very requirements. Due to its image-centric user interface, it reduces programming complexity, and - thanks to easyTouch - users can create machine vision applications faster. Companies thus benefit from an entirely new way to access machine vision: one that is application-driven rather than technology-driven .


Im in softw re THE HE EYE OF PRODUC ION OF PRODUCTION


Machine vision, which represents the ‘eye of production’, plays a key role in industrial automation. Thus machine vision techniques will support more and more automated production processes in a wider scope of industrial sectors. Use cases range from determining the position of obj


bjects and controlling


gripping processes, to reliably identifying many different kinds of workpieces. In this way machines and robots learn to ‘see’, improving their interactions with humans. So far, two-dimensional processes used to be the standard. Using modern 3D methods, it will be possible to track moving objects in image sequences and to id entify their directio n of movement in space. Technologies such as ‘3D scene flow’ are designed to optimise the collaboration of humans and machines. Consequently, automated manufacturing processes become much more efficient, safe, and flexible. Thus, the relevance of machine vision as an important driver in the whole robotics and automation sector will increase.


www.mv n


VTec Software www.mvtec.com


MVT T: +49 89457 6950T: +49 89457 6950


deep-learni g O R techniques A


s artificial intelligence turns from theory to reality, brought about by deep learning algorithms and ever faster processors, machine vision software developer, MVTec, launches its latest version of Halcon, which includes deep-learning for optical character recognition (OCR). A series of pre-trained fonts help to make the recognition process faster than previous classification methods. In addition, the automatic text recognition function is faster and now works with dot print fonts. Halcon 13 also reliably reads defective or occluded barcodes. Furthermore, the QR code reader is even more robust when it comes to reading unclear or distorted codes.


The ability to remotely debug HDevelop programs running under HDevEngine in a


C# or C++ application has also been improved. This is done by linking HDevelop with the corresponding HDevEngine application. Another new development is the significant performance improvement of HALCON’s key basic identification technology shape-based matching. For example, byte image searches are up to 300 per cent faster on machines with AVX2-compatible processors. In addition, the speed of related technologies, such as shape-based 3D matching, local and perspective deformable matching, as well as component-based matching has been significantly increased.


Halcon 13 also provides a new user-friendly texture inspection capability. The properties scaling and brightness, often vary a great deal. The inspection processes can now be config


bj


ured easily by simply of textures, such as


importing training images. Finally, Halcon 13 optimises surface-based 3Dmatching in flat bobjjects, as it now uses edge information in addition to 3D point clouds. In particular, this optimises applications for picking boxes. A newmethod for the high-quality reconstruction of 3D objects, usingmultiple cameras, is also available. If you are interested in learning more about this powerful software, Multipix Imaging, the UK specialist, is holding an event on 19 January, where the latest features will be explained further. See website below.


Multipix Imaging T: 01730 233332 T: 01730 233332 3i http://mult pix.com halcon-13 http://multipix.com/halcon-13


Imaging software briings advanced deep-learning OCR techniques


/AUTOMATION AT


/AUTOMATION


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44