Discover the benefits of Overview
AI QC keeps factories competitive
The most accurate and flexible AI vision product
Applications and examples
The flexible calculator to measure your needs
Learn more about Overview
Our principles in work and life
Our values and goals
Meet our professional team
Deep Learning is a domain of Machine Learning which is a domain of Artificial Intelligence. Our cameras mimic human intelligence like better and faster eyes. In the context of manufacturing and Overview.ai, computer vision deep learning refers to the analysis of image data using state-of-the-art deep neural networks to generate insights beyond simple measurement. Deep Learning models learn from existing labeled and unlabeled data to create these predictions. According to Forbes, machine learning-based automated quality testing can increase detection rates by up to 90%.
With a state-of-the-art deep learning system you can distinguish blemishes, defects and anomalies from natural variations in the environment increasing your true accuracy and reducing false positives and false negatives. Deep Learning combines the flexibility of a human inspector or quality control engineer with the speed and repeatability of a traditional system.
Deep Learning Based Inspection is fundamentally better because it puts the power with the customer. Not only does it beat traditional machine vision systems in head-to-head accuracy every time, but it also can be trained by a factory operator or quality control engineer ,and not a specialized and expensive vision engineer. This means that the system can adapt and get better over time. No more dead cameras in the factory. Additionally, by enhancing your dataset with synthetic data you can make your system even more flexible and train it to catch defects that it has never seen before.
Traditional Machine Vision, also referred to as classical computer vision, requires developing a custom algorithm every time that is rules based. These algorithms can be very sophisticated and require a highly trained and expensive engineer to program. Once that engineer finishes development and walks away from the project there is no way to improve the inspection. Even worse if the engineer tells you that he can only get to a subpar accuracy, there is nothing more you can do. So not only does traditional machine vision have a ceiling, but it can also only inspect limited and clearly defined tasks. Because of these fundamental issues and the fact that it can’t improve over time, traditional machine vision systems are decreasing in popularity compared to new deep learning systems.
Overview has 9 different deep learning vision modules that can be combined in any way. This means that the platform is perfectly adaptable to every use case. Whether you need a high speed inspection or you want to validate a part with 200+ different inspection points, we have got you covered. Our algorithms are constant improving and your inspection system will never go stale. Thats our guarantee.
