Sick’s Inspector83x 2D vision sensor enables plug-and-play machine vision inspection using artificial intelligence (AI), even in high-speed production conditions.
Designed by Sick, a leading global provider of sensor-based solutions for industrial applications, the Inspector83x helps eliminate the complexity of conventional machine vision. Thanks to its learning capacity, it is enough to present a few examples to the camera to obtain operational results in a few minutes in terms of fault detection.
The Sick Inspector83x is an all-in-one tool that benefits from powerful built-in lighting and up to 5 MP resolution. With a 4-core CPU and high-speed data transfer rates via industrial networks, the equipment performs AI inspections directly on the device and at significantly higher speeds than its predecessors. Its precise and rapid response does not require any control by an external machine.
According to Sick, up to 15 inspections per second can be reliably performed for machine vision applications such as defect and anomaly detection or classification.
The Inspector83x masters complex inspections over medium to large fields, including evaluating products with unpredictable characteristics and verifying complex assemblies, as well as reading and verifying OCR/OCV characters. Models with color imager will be available soon and will complete the range for color feature inspection suitable for item sorting, defect identification and color-based quality control.
AI machine vision for non-specialists
Thanks to the pre-installed Sick Nova base software, the Inspector83x is easy to use. Using a standard computer connected via the camera’s USB-C port or network interface, users follow the intuitive interface to present examples to the camera under real production conditions, then perform training and inspection. Just five examples are enough. By combining AI functionality and traditional rules-based tools, for example to add a simple measurement, it is possible to configure inspections in a very pragmatic way.
Non-specialist operators can simply add a new product example and the camera will learn autonomously. No machine vision experts or laborious preparation are required.
For complex scenes with lots of examples and large datasets, users have the ability to access the computing power of Sick dStudio cloud Service to configure their own neural network, which can then be exported as a file small size towards the Inspector83x. With Sick Nova, experienced users can also perfect their solution through custom development using Lua and Halcon programming. The Sick cloud-based dStudio service also offers the unique benefit of peer collaboration and data management, while remaining focused on usability.
High-speed data transfer with the versatile industrial network
Once installed, image inference is performed directly on the Sick Inspector83x and the results are sent to the machine control as ‘OK/NOK’ results or values from the sensor. The Inspector83x is optimized for fast data transfer to industrial networks with dual ports for EtherNet/IP or Profinet integration. A dedicated high-speed Gigabit-Ethernet port provides high-resolution image data transfer as well as data logging or TCP/IP integration. A built-in export function allows you to generate custom configurations for the most common API types with the press of a button.
The Inspector83x has seven inputs and five outputs. The device’s built-in pulse delay and queue function precisely synchronizes the camera’s image outputs based on time or encoder pulses to activate commands from the connected machine, e.g. example to trigger an ejector.
A wide range of accessories is available. A dedicated lighting connector can be connected to external light sources such as a backlight or light bars. A near infrared (NIR) variant will be available later in 2024. Using a standard C-mount thread, customers can use lenses from the wide range offered by Sick, as well as specific optics meeting specific needs. complex inspections.
Photo: High-speed conveyor inspection with Inspector83x.
