About this Virtual Instructor Led Training (VILT)

Computer Vision or CV is a branch of Artificial Intelligence (AI) implementing Machine Learning (ML) and Deep Learning (DL) models to analyze visual data. Real-time analytics provide critical insights on streaming data from the multitude of sensors upstream, both surface and subsurface, and in the midstream and downstream. We can run anomaly detection algorithms on real-time data at the edge where the sensors are operationalized. Real-time data streams can address safety, sustainability, and operational efficiencies. This 4 half-day Virtual Instructor-Led Training (VILT) course introduces CV and real-time data analysis. It discusses several architectures and data quality procedures to optimize operations and maximize production. Course Highlights: Day 1 opens the critical conversation of applying data-driven analytical models upstream, midstream, and downstream. For example, what are the standard soft-computing modeling techniques for image data and operationalizing models for real-time data? Day 1 also sheds light on data-driven approaches and helps participants decide when and how to get the most out of image data and real-time data streams. Day 2 gets into the details and builds on the topics introduced on Day 1. Participants will go through an anonymized case study in each of the three areas: Computer Vision (Denoising images using segmentation regression models, developing an object detection model), Real-Time Analytics (DataStream preparation and event stream processing architectures). Days 3 and 4 will draw upon advanced Machine / Deep Learning data-driven methodologies covering historical image and static data and real-time data, edge computing, and event stream processing using Digital Signal Processing algorithms. In addition, participants will develop a suite of data-driven workflows that provide a repeatable and scalable template for a Digital Twin case study based on the Machine / Deep Learning workflows discussed on Day 1 and Day 2.