AI Camera: Automatic Optical Inspection with AI

Self-learning AI Camera for Intelligent Inspections – Fast, Precise, Reliable!

TURCK Intuitive Vision (TIV)

TURCK Intuitive Vision (TIV) revolutionizes demanding industrial applications such as quality control, kitting, and more with simple commissioning and self-learning functions. The powerful, AI-based all-in-one vision solution learns what is important in the target state from application images. This enables it to detect production errors or incomplete kits – without the need for time-consuming manual rule definition. Experience faster inspections, error-free kitting, and more, and test TIV in your specific application.

AI Camera Ensures Error-Free Kitting

TIV camera uses machine learning to recognize objects and check the correct composition of sets

Benefits:

High productivity thanks to minimal errors and rework
High flexibility with changing sets, as several set configurations can be taught in
Fast feedback thanks to powerful hardware
Simple integration without programming effort
More cost-effective than fully automated systems with robots

Hide All Open All

Challenge

In picking and packing processes known as kitting, employees put together sets from various components. It can happen that sets are loaded incorrectly due to human error. Besides the frustration caused and damage to the company's image, this also leads to high costs for the subsequent delivery of missing parts. Fully automated processes do not pay off in many applications due to the low unit quantities involved or the degree of flexibility required. A camera system that checks and releases the set in the shortest possible time is needed.

Solution

The TIV AI camera system is taught in with a set of training images. A complex definition of rules is not necessary. The camera thus reliably recognizes different items and checks whether the detected objects correspond to the defined set. The system reliably detects even the smallest differences in the millimeter range or very similar objects such as washers of different sizes, regardless of the arrangement of the components. Various correct configurations can also be taught in. When putting together kits for sensors, employees for example can teach in two sets for different sizes. This prevents an M18 sensor, for example, from being supplied with mounting accessories for an M12 sensor. The camera transmits the OK or NOK result of the test to the PLC, other systems or displays and other visualization devices via a conventional digital output.

Hide All Open All

AI Camera Guarantees Error-Free Placement

Difference Check enables the AI-supported TIV camera to reliably detect errors in assembly processes – intuitive commissioning without any programming required

Benefits:

Simple integration into existing processes without complex programming
AI-supported error detection that sees more than the human eye
Fast teach-in from as few as 10 images per state guarantees high flexibility for product changes
Clear OK/NOK output and feedback to the PLC

Hide All Open All

Challenge

At manual workstations several test points on products often have to be checked during final assembly. Different states can be correct at each point,such as the presence or correct positioning of components. In order to minimize the amount of costly rework and number of rejects, correct placement must be checked automatically during production. It is important here that the complexity of the quality control to be implemented is not determined by the complexity of the inspection task.

Solution

The Difference Check with TIV makes it possible to teach in objects with multiple test regions (ROIs). The neural network of the TIV is trained by teaching in at least 10 images classified as 'good' (OK) and at least 10 'wrong' (NOK) images. The camera determines the rules behind this on its own. It detects deviations in the test regions and outputs the result via a PLC interface. An image is considered OK if all test regions are correct, otherwise it is considered NOK. A consistent background is essential, as this is also taught in and contributes to object detection. The camera evaluates each test region individually and determines the overall result. If there are faulty test regions, an NOK signal is output, which activates a signal light or, for example, causes the product to be ejected. Employees are notified with a visual indication and can react accordingly.

Hide All Open All

Experience AI vision live – test TIV now!

Want to know what our intelligent vision solution can really do? Then simply send us some images from your application – our camera experts will analyze your requirements and carry out a free, customized feasibility test with TIV. This will give you a concrete idea of how TIV makes a difference: fast, precise, and reliable.

Fill out the form now!

Get in touch with our AI camera experts!

Please select a salutation

Please enter your first name

Please enter your last name

Please enter the name of your company

Please enter your city

Please enter your country

Please enter your e-mail address

Please enter your phone number

Please read the privacy policy and confirm your acceptance using the checkbox

I accept the privacy policy* i

* required

About Turck

Turck is your global partner for factory, process and logistics automation in numerous industries. With its digitally networkable solutions for efficient automation systems, the family-owned company supports the digital transformation and is one of the pioneers of Industry 4.0 and IIoT. As a specialist in smart sensor technology and decentralized automation, Turck brings intelligence to machines and ensures reliable detection, transmission and processing of relevant production data – from sensor to cloud – not only with robust I/O solutions in IP67, but also with user-friendly software and services.