ELEC 4727/5727: Machine Vision

🎓 Welcome to ELEC 4727/5727 Machine Vision! This course provides an introduction to machine vision systems regarding concepts, algorithms, and applications, especially revealing how the latest vision systems can mimic and even outperform humans in specific tasks (thermal, ultrasound, MRI images, etc.). Through practical assignments and projects with Jetson Nano, students will learn to develop vision systems that can see, interpret, and act on visual data. Key topics include image/video formation and transformation, feature extraction, object recognition, and evaluating vision system performance. By the end of the course, students will have a solid background in machine vision algorithms and be able to advance their knowledge in solving real-world problems.

📚 Syllabus:

Week Topic Assignments (Tentative)
1
  • Course Overview
  • Introduction to Machine Vision: Human Visual Perception vs. Machine Vision
  • History and Evolution
  • Multispectral Visual Information: Scientific and Industrial Applications
 Assignment 1
2

Image Acquisition and Preprocessing (RGB Images)

  •  + Image acquisition devices (cameras, sensors)
  •  + Image formation and properties
  •  + Preprocessing techniques (filtering, noise reduction, normalization)
3

Image Classification (RGB Images)

  •  + Fundamentals of image classification
  •  + Feature Extraction and Description (SIFT, SURF, LBP)
  •  + Training and evaluation of image classification models
  •  + Convolutional neural networks (CNNs) for image classification
 Assignment 2
4

Image Segmentation (RGB Images)

  •  + Image segmentation techniques (thresholding, region growing)
  •  + Semantic segmentation vs. instance segmentation
  •  + Implementing segmentation algorithms with deep learning frameworks
5

Object Recognition and Detection (RGB Images)

  •  + Introduction to object recognition and detection
  •  + Classical methods vs. deep learning approaches
  •  + Implementing object detection algorithms (YOLO, SSD, Faster R-CNN)
 Assignment 3
6

Midterm
7

Automatic Image Annotation and Dataset Creation (RGB Images)

  •  + Importance of automatic image annotation
  •  + Techniques for automatic image annotation (object detection, semantic segmentation)
  •  + Creating and curating labeled datasets for machine learning tasks
8

Medical Imaging (MRI, Ultrasound)

  •  + Introduction to medical imaging modalities
  •  + MRI image acquisition and processing
  •  + Ultrasound image acquisition and processing
 Assignment 4
9

Satellite and Remote Sensing (Satellite Images)

  •  + Satellite image acquisition and characteristics
  •  + Remote sensing applications and techniques
  •  + Image preprocessing for satellite imagery
10

Thermal Imaging (Infrared Images)

  •  + Principles of thermal imaging
  •  + Thermal camera technology and applications
  •  + Preprocessing and analysis of thermal images
11

Depth Perception and 3D Vision (Depth Images)

  •  + Depth sensing technologies (stereo vision, structured light)
  •  + Depth estimation algorithms
  •  + 3D reconstruction methods (point clouds, surface fitting)
12

Motion Analysis and Tracking

  •  + Optical flow estimation
  •  + Object tracking techniques (Kalman filter, particle filter)
  •  + Multi-object tracking frameworks (MOT)
 Assignment 5
13

Advanced Topics in Machine Vision

  •  + Biometric recognition systems
  •  + Medical imaging applications
  •  + Autonomous navigation
14
  • Final Project Review
15
  • Final Project Demo and Presentation