Updated Machine Vision Software from MVTec for Faster AI Operations

[Hasan S. Cemkan]

## Updated Machine Vision Software from MVTec for Faster AI Operations

MVTec has announced the new HALCON 26.05 version for industrial machine vision and automation systems. This update aims to meet the high speed and accuracy requirements in production lines by enabling deep learning and rule-based algorithms to run faster and more efficiently.

### Performance and Reliability Improvements
HALCON 26.05 offers faster inference in deep learning-based object detection processes while maintaining high accuracy even for small and irregularly shaped objects. Additionally, it provides integrated data augmentation techniques to increase robustness against environmental changes such as lighting conditions, rotations, and partial occlusions.

For rule-based inspections, the new correction function allows for more reliable reading of 2D data codes on curved or deformed surfaces. This feature, aimed at the packaging and electronics sectors, facilitates marking on non-flat geometries.

Furthermore, automatic contour optimization in the Shape Matching workflow cleans up unstable or misleading contours in trained samples, leading to more consistent matching results and reduced processing time.

### New Approaches to Development Processes
The HDevelopEVO development environment, offered as a preview with the HALCON release, enables the integration of created scripts into custom applications via the .NET interface. This allows for more flexible deployment of machine vision solutions.

Additionally, multimodal interaction with large language models is supported; the direct inclusion of visual data into prompts enables more interactive and useful workflows in development and testing processes.

### Contributions to Industrial Automation
The increases in speed and accuracy in HALCON 26.05 offer significant advantages in machine vision-based inspections in sectors requiring high efficiency, such as manufacturing, logistics, and electronics. These improvements support the more effective integration of machine vision solutions into automation architectures and digital supply chain processes.

### HALCON 26.05 Technical Specifications

• Faster deep learning inference
• High recognition accuracy for small and irregular objects
• Improved data augmentation against environmental changes
• 2D data code reading on curved and deformed surfaces
• Shape Matching with automatic contour optimization
• HDevelopEVO script integration with .NET support
• Multimodal visual interaction support with large language models

With this update, MVTec provides significant advancements in both performance and development processes for industrial machine vision applications.