Self-learning AI camera
TURCK Intelligent Vision (TIV) brings artificial intelligence directly to the line — robust, flexible and fast
Benefits
Fast commissioning without programming
Stand- alone device in IP67, no additional edge hardware
Real-time results, scalable through transferable networks
With its new TIV AI camera, TURCK promises a paradigm shift in industrial image processing: The intelligent camera can be change using just a few sample images, with no need for complex programming. The TIV ( TURCK Intelligent Vision) recognizes patterns and differences independently, and reliably detects good and bad parts, and different classes. Training and implementing the neural networks takes place directly on the camera, with the help of a powerful, 12-megapixel global shutter sensor (Sony Pregius S, fourth Generation) and a NVIDIA Jetson Nano GPU with 4 GB RAM. The image processing runs in real time on the device.
Flexibility and efficiency for practical applications
With four pre-installed AI apps — Difference Check, Classifier , Detector and Code Scan — the TIV12MG-Q110N takes over central tasks in industrial image processing: from completeness control and classification, to object detection and 1D/2D code recognition. The camera detects and evaluates test areas separately, and supplies coordinates, confidence scores and OK/NOK signals directly to the PLC or IT systems. The C-mount connector means common lenses can be used flexibly; an optional protective tube gives the system IP67 protection. The entire control system, lighting technology, sensors and power supply has a decentralized and robust design — ideal for use directly on the line.
Seamless integration with TAS
Operation is intuitive via the web browser. Integration into the TURCK Automation Suite (TAS) makes device management and integration into digital maintenance and monitoring processes easier. The camera can be integrated flexibly into existing systems via M12 connections for power, network, trigger and I/O. On- site operation, portable data sets and neural networks enable scalable rollouts without additional edge hardware or license costs.
