The new EasyDeepLearning library is a Convolutional Neural Network-based classification library which learns by example. It has been tailored, parametrised and optimised for analysing images, particularly for machine vision applications. It learns how to distinguish defects or classify parts by being shown many images of the parts to be inspected. It includes functions for dataset creation, classifier training and image classification. Compatible with CPU and GPU processing it is able to detect defective products or sort products into various classes. It supports data augmentation, works with as few as one hundred training images per class. EasyDeepLearning has a simple API and the user can benefit from the power of deep learning with only a few lines of code. It includes a free Studio application to ease the creation of deep learning applications.
The new Easy3D library is a set of software tools enabling the development of 3D Machine Vision Inspection Applications. It is able to generate a depth map from a series of images that contain a laser line projected on the inspected object and computes a calibration model applied to depth maps to transform them into calibrated 3D point clouds. It provides functions to generate such ZMaps. The Easy3D Studio application provides an easy way to configure a complete 3D laser line inspection setup.