Smart cameras based on deep learning

Cognex

Inka Krischke | Inka Krischke, 28.10.2020, 11:58

Smart cameras based on deep learning

With the In-Sight D900 vision system from Cognex, the 'In-Sight ViDi' deep learning software is embedded in a smart camera. The camera is installed in the line without a PC.

With the In-Sight D900 vision system from Cognex, the deep learning software 'In-Sight ViDi' is embedded in a smart camera. The camera is installed in the line without a PC.

Images

The modular IP67-certified image processing system includes field-replaceable lighting, lenses and filters as well as covers that can be customized according to individual application requirements. High Dynamic Range (HDR+) imaging produces evenly exposed images. The LED display enables remote pass/fail evaluation and the test results can be saved locally on an SD card.

According to Cognex, an inference engine has been integrated specifically for the execution of deep learning applications. 'In-Sight D900' works with the easy-to-use spreadsheet user interface, which simplifies application development and factory integration. Application engineers have access to the full suite of traditional vision tools such as PatMax, edge finders and measurement tools.

With In-Sight ViDi, the D900 also has three deep learning tools:
ViDi Read, ViDi Check and ViDi Detect. 'ViDi Read' is ideal for OCR tasks, as the module deciphers deformed, distorted and poorly lasered codes using optical character recognition. 'ViDi Check' is used to check assemblies and component positions, while 'ViDi Detect' helps to detect defects and other undesirable deviations.

Would you like a little more?

First-hand news every week - that's what the Computer&AUTOMATION newsletter offers you.

Subscribe to our weekly newsletter and keep your finger on the pulse of automation time. Once a week, on Tuesdays, we provide you with the latest news, exciting interviews, guest commentaries, application-oriented technical reports and new products. Stay up to date with us.

Our newsletters are of course free of charge and can be canceled at any time!

Click here to register.

Back to topic page

You might also be interested in

Cognex

Automated inspection

Until now, optical inspection in 3D was associated with high programming effort and costs, meaning that users had to resort to 2D solutions. A new type of laser-based 3D vision system is now set to put an end to this.

Cognex

Inspections in 3D as easy as in 2D

Cognex launches the In-Sight 3D-L4000 integrated vision system. The smart camera is equipped with 3D laser profile technology.

Basler

Embedded vision made easy

At embedded world 2021 DIGITAL, Basler is working with partners such as NXP and NVIDIA to simplify the path to the final embedded vision solution for customers.

Vision Components

MIPI camera and driver for Nvidia developer kit

Vision Components has put together a starter package for embedded vision projects with MIPI: the camera module and MIPI cable are simply connected to the 'Jetson Nano Developer Kit' from Nvidia and application development can begin straight away.

Cognex

Integrated OLED display

The DataMan 8700 series portable barcode readers are based on a completely redesigned platform, according to Cognex. The devices can read demanding DPM (Direct Part Mark) and label-based codes immediately, even if important elements of the code are...

3D camera series Ensenso N40/N45

It can also collaborate

The Ensenso N series is a compact and robust 3D camera system that has been specially designed for 3D applications in robotics and automated series production.

Flir

New lenses recognize even more gases

The GF77 Gas Find IR series from Flir are uncooled gas detection cameras for gas detection and thermal imaging inspection in chemical and industrial production.

Vision Components

Embedded vision combined with Raspberry Pi

With the VC MIPI Embedded Vision Kit from Vision Components and a Raspberry Pi Compute Module 3 or 3+, a fully-fledged and industrial-grade embedded vision system can be set up quickly.