Our advanced solutions reject defects and improve products quality automatically with low false positive/negative rate. Our team is composed with Profesionals and Experts Integrating Automated Vision Stations to deliver products with the best quality and to improve production throughtput.
Deep Learning vs Conventional Vision? Our Robotic & Computer Vision Experts are used to upgrade production machine software and retrofit the whole process as well as building from scratch advanced top-notch fully automated visual defect detection machines to replace manual operations.
Our proven Deep Learning Machine Vision Software Suite "Retina A.I. Vision" is compatible with all known exisiting software frameworks.
We have been extensively testing & working closely with manufacturers to find out the best IPCs and Vision controllers for running Deep Learning A.I. Vision as well as conventional vision software.
+41 31 557 26 46
Bin Picking cylindrical parts with random & unstable poses. White on White Background Bin picking, also known as “part picking” or “piece picking,” is the automation of the process of identifying and grasping specific objects from a bin or...
Aims of Industrial Vision Industrial machine vision allows replacing human eye-based decisions in the production chain. Not being influenced by fatigue and moods, it improves productivity, quality and profitability, while having full control and visibility over the...
Blazing accurate “Deep Learning Retina-Cherry” Inference and Training Solution Guess how often people are looking at theira smartphone screen per day? How would you feel in case you receive a brand new smartphone with some defects on the glass? We can...
Come and visit us at Vision Stuttgart from 04-06 October 2022 Come and visit us at the booth 8E60.6! We have new products show that might be interesting for your A.I. Vision applications. Don’t hesitate to come with samples to challenge our Retina A.I....
The 4 key features that help keeping producing and selling from the most expensive country of the world !!! 1. ReliabilityProducts are conform to customer expectation independently of time and location where you purchase products. This is how famous brands are...
“3HLE” celebrated 3 Years! And more to come !!! This September 2021 “3HLE” Automation et Robotique SA has celebrated its 3rd Anniversary. It’s amazing how projects are parading quickly! And poouuuf, 3 years has gone through the on-going...
Did you know that all “PET bottles” are marked with an expiration date? The printing quality depends on the production line speed, the used technology to mark bottle. So far, conventional vision software can read quite well when it comes to the same bottle...
“3HLE” Presented as the Startup of the Month by the Specialized Machine Vision Magazine 3HLE is glad to be listed as the Startup of the Month by the specialized Vision & Metrology Magazine and presented by Vision Ventures (www.vision-ventures.eu)....
Amazing “Deep Learning Retina-Cherry” Solution for Detecting Scratches and Wear Signs on Transparent Glass and Metallic Surfaces. Only human eyes are amazing for detecting defects on shiny and transparent parts under less than ideal lightning. At...
Workflow for Deploying Quickly an OCR software for reading on Shiny, Curvy surfaces with Uneven Lightning at a rythm of more than 100’000 parts/hour. We can be surprised to see how quickly softdrinks are produced. Moreover, their quality control are permanently...
Highly Efficient “Deep Learning Retina-Cherry” Training and Inference Solutions The RETINA deep learning visual inspector is the most accurate for detecting cracks/anomalies on shiny and transparent parts such as glass jars and bottles. It can control the...
How Deep Learning OCR Successfully Reads 99.99% on Shiny, Curvy surfaces with Uneven Lightning at a rythm of more than 100’000 parts/hour. We can be surprised to see how quickly softdrinks are produced. Moreover, their quality control are permanently checked....
Merging a set of mosaics into an ultra high resolution image for cracks, scratches and dust inspection as tiny as 5 microns. We could not find any industrial camera on the market that provide a minimum resolution of 5um/pixel on a field of view of 50x50mm for watch...
Blazing accurate “Deep Learning Retina-Cherry” Inference and Training Solution Guess where these screws can end up in a human body? Validation needed? We can either offer turn-key solutions or help our customers to gain missing knowledge for validating...
OCR Blazing fast “Deep Learning Retina-Olive” Training interface Printing or engraving product codes, production/expiration dates, production line IDs on the packaging or the product itself during production is most of the time very jerky due to...
Products packaging is more important than what we think. Even if the inner product is not damaged, people tend to choose the prettiest boxes because the exterior is reflecting how well a product has been handled and produced. Most of the time, people are judging the...
The application below demonstrates how “3HLE” helps production plants become Greener & more Ecologic in addition to saving money while producing Higher Quality Coffee Capsules Cups Using Deep Learning Visual Inspection for rejecting only real Delicate...
Did you know that American “Coffee Capsules” production machines are designed and manufactured with high precision and high quality “Swiss Made” standard? The quality inspections are performed on each step of the production process on...
Microsoft Windows IoT White Paper This isn’t my father’s business environment. The days of the 100-year-old company are over. Sure, some company names from the last century will still be recognizable. And values such as customer service, honesty, and integrity will...
– What happens if a Windows 10 PC encounters a sudden power failure? – How to reset Windows 10 to a known state machine after a power failure? – Is Windows robust on production machines? – What is Windows 10 IoT? “3HLE” Automation...
Evaluation of trajectory accuracy and max speed of the robot arm by drawing and placing...
Just came across these pictures back from 2009. And they are still realistically...
Performance Evaluation of a Raspberry Pi running Deep Learning Mobilenet with a shrinked resolution of 640×480 raspicam for detecting 10 pre-trained categories/classes: bird, cow, cat, dog, sheep, horse, person, pottedplant, tvmonitor, chair. The results show...