Matrox Imaging’s Matrox Imaging Library (MIL) 10 Processing Pack 3 software update features a CPU-based, image classification module that makes use of deep learning technology for machine vision applications. Processing Pack 3 also includes a photometric stereo tool to bring out hard to spot surface anomalies and a new dedicated tool to locate rectangular features.
Leveraging convolutional neural network (CNN) technology, the Classification tool categorizes images of highly textured, naturally varying, and acceptably deformed goods. The inference is performed exclusively by Matrox Imaging-written code on a mainstream CPU, eliminating the dependence on third-party neural network libraries and the need for specialized GPU hardware. The intricate design and training of a neural network is carried out by Matrox Imaging, taking advantage of the accumulated experience, knowledge, and skill of its machine learning and machine vision experts.
Registration tool set gets photometric stereo
The photometric stereo technique is now available within the Registration module to produce a single image that emphasizes object surface irregularities such as embossed and engraved features, scratches and indentations. The composite image is computed from a series of images taken with light coming in from different directions as produced by illumination solutions based on the CCS Inc. Light Sequence Switch (LSS) or Smart Vision Lights LED Light Manager (LLM).
Shape-finding tool for rectangles
Part of the Geometric Model Finder (GMF) module, the Rectangle Finder tool is a faster, more flexible, and more robust option than generic geometric pattern matching. The tool can simultaneously search for multiple occurrences of rectangles with different scale and aspect ratios.
The MIL 10 Processing Pack 3 update is now available to all registered MIL users with valid maintenance subscriptions through the software’s update service as well as those interested in evaluating the software. Check out a “Deep Thoughts on Deep Learning” Q&A with Matrox Imaging. For further details, visit Matrox Imaging.