Fast & Reliable
Deep learning-based classification in MVTec software enables fast and reliable assignment of images to trained classes. A simple folder structure is sufficient for training, eliminating the need for tedious individual annotation. This reduces the effort required for data preparation and development, resulting in short setup times.
The classification process is highly efficient, delivering low error rates even under varying lighting conditions or with complex objects. This makes the method ideal for automated inspection processes in industrial environments.
Sorting and grading of fruits, vegetables, or agricultural raw materials.
Classifying similar tablet variants by type or composition.
Automated decision-making to determine whether objects are defect-free or faulty.
2D & 3D MACHINE VISION TOOLBOX
HALCON 25.11 introduces Continual Learning – Classification, a new technology that makes training and maintaining classification models faster and more flexible. Users can create models with only few images per class and adapt them at any time – for example, to refine existing classes or add new ones.
Unlike conventional deep learning, this approach prevents catastrophic forgetting and keeps maintenance effort low. Based on MVTec’s pretrained models optimized for industrial scenarios, applications can be updated quickly without full retraining. Because the method requires minimal computing power, updates can even be performed directly on edge devices, eliminating the need for external training hardware while ensuring efficient, long-term operation.
The result is a flexible solution that evolves with changing production conditions and remains suitable for embedded and edge environments such as smart cameras, sensors, and inspection modules.
EASY-TO-USE NO-CODE SOFTWARE
MVTec MERLIC provides an easy entry point to deep learning-based classification with the “Classify Image” tool.
Together with the MVTec Deep Learning Tool, MVTec MERLIC enables direct training data preparation and the creation of classifiers without programming.