MVTec goes on world tour to present Halcon 13 features
MVTec presents the new version 13 of its image processing software Halcon to a broad target group. From September 2016 until well into the first quarter of 2017, companies will get to know the comprehensive new functions down to the last detail in numerous events as part of a global road show. Approximately 1,800 participants have registered so far. This World Tour is being organized by MVTec's international sales partners.
The World Tour kicks off with the MVTec Halcon Days, an intensive training for sales partners on technical and sales-related issues, which will take place between September 26 and 28, 2016. This kickoff event will provide in-depth training in HALCON 13 for more than 70 technical support and sales representatives from MVTec's global partners, who will then pass on this specific knowledge to the users. With this approach, MVTec guarantees the best possible customer support around the world.
Optimized character recognition with deep learning
The MVTec Halcon Days and the subsequent World Tour focus on new and enhanced features in Halcon 13, which is released on November 1. These features include significantly improved identification technologies. For the first time, the latest release offers optical character recognition (OCR) functions based on deep learning technology. A series of pre-trained fonts help to make the recognition process faster than with all 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 bar codes. Moreover, 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 percent 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.
Reliable texture inspections
Halcon 13 also provides a new user-friendly texture inspection capability. The properties of textures, such as scaling and brightness, often vary a great deal. The inspection processes can now be configured easily by simply importing training images. The algorithm automatically adapts all necessary parameters based on the error-free training images. Potential texture defects are thus reliably recognized with the help of the inspection model trained in this manner.
Finally, the new Halcon version optimizes surface-based 3D matching in flat objects, as it now uses edge information in addition to 3D point clouds. In particular, this optimizes applications for picking boxes. A new method for the high-quality reconstruction of 3D objects, using multiple cameras, is also available. This method uses the information from all camera angles at the same time, thereby producing more robust results than conventional stereo approaches. By using the cameras, the method also allows an object to be reconstructed and printed in 3D (additive manufacturing/reverse engineering).
