Introduction to Industrial Image Processing

Overview

Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!

Grab it

The aim of this course is to provide an overview of current technologies and trends in image processing and their industrial application. The theory and practice of industrial image processing is covered using the complete chain - from image creation and image processing to the derivation of a measurement result or decision. You will learn the basics and skills for model-based evaluation of spatially resolved image information so that not only innovative but also robust solutions for industrial applications can be implemented.

Syllabus

Week 1

Introduction to the basics and history of industrial image processing
- What are typical use-cases and applications of industrial image processing?
- How does image acquisition relates to the human vision?
- How did visual analysis and processing evolve over time?
Basics of image acquisition
- Overview over the physical foundations of light
- Explanation of the hardware components of an industrial machine vision system: optical elements (e.g. lenses), imaging sensors and light sources
Overview over illumination strategies****

Week 2

Algorithms for feature amplification
- Digital representation of images
- Procedures for pixel-wise image modification
- Algorithms for image pre-processing, such as smoothing and sharpening
- Morphological operations for preparing object detection
- Filtering in the frequency domain using Fourier transform

Week 3

Detection of Image Features
- Extraction and localization of shapes such as edges, circles and squares using the Canny-Edge-Detection and Hough-Transformation algorithms
- Image segmentation for object detection based on the Two-Pass-Algorithm
- Detection and description of repeatable image features using the Harris-Corner-Detector and SIFT-Features

Week 4

Object Detection and Classification
- Definition of classification and general procedure
- Introduction and comparison of different classification techniques: Nearest-Neighbor, Bayesian Decision Rule, Support Vector Machines
- Object recognition and feature matching based on SIFT-Features and homography estimation

Week 5

Photogrammetry – From 2D images to 3D information
- Mathematical and geometrical foundations of stereo camera pairs
- Methods and algorithms for camera calibration
- Rectification of images for efficient image comparison
- Disparity computation and triangulation for determination of a 3D point
- Industrial applications of photogrammetry
Summary
- Summary of all five modules
- Outlook to advanced image processing techniques using machine learning and AI