Welcome to the world of PC Vision and Deep Learning

Think your application is too complex to automate? Think again. Today, increasing numbers of manufacturers are turning to deep learning-based image analysis to help improve the automation of many tasks, including:

• Assembly verification and deformed part location
• Cosmetic and functional defect detection
• Texture and material classification
• Character reading, including distorted print

Cognex deep learning technology— designed specifically for factory automation—now makes it possible to automate previously unprogrammable applications, reducing error rates and quickening inspection times.

Industry is turning to deep learning technology to solve manufacturing inspections that are too complicated, time-consuming, and costly to program using traditional machine vision. Cognex ViDi is the first deep learning-based software designed to solve these complicated applications for factory automation.

What is Deep Learning?

Deep learning technology is used in advanced manufacturing practices for quality inspection and other judgment-based uses.

From the phones in our pockets to the reality of self-driving cars, the consumer economy has started to tap into the power of deep learning’s neural networks. Deep learning has emerged as a foundation technology in:

• Speech, text, and facial recognition
• Mobile and wearable devices
• Medical diagnostics
• Internet security

Deep learning technology is being used to predict patterns and make critical business decisions. This same technology is now migrating into advanced manufacturing practices for quality inspection and other judgment-based uses.

Deep Learning and Machine Vision

New deep learning-based software helps machines recognize images, distinguish trends, and make intelligent predictions and decisions.

For decades, machine vision systems have taught computers to perform inspections that detect defects, contaminants, functional flaws, and other irregularities in manufactured products. Human visual inspection prevails, however, in situations that require learning by example and appreciating acceptable deviations from the control. Machine vision, by contrast, offers the speed and robustness that only a computerized system can.

Deep learning technology uses neural networks which mimic human intelligence to distinguish anomalies, parts, and characters while tolerating natural variations in complex patterns. In this way, deep learning combines the flexibility of human visual inspection with the speed and robustness of a computerized system.

Why use Deep Learning?

How to use Deep Learning?

Deep learning software trains on a set of labeled images that represent a part’s known features, anomalies, and classes—much like a human inspector would be trained. A supervised training period teaches the system to recognize explicit defects. For defects that come in multiple forms, the system trains itself in unsupervised mode to learn the normal appearance of an object, including its significant but tolerable variations.