Image Processing

Lecture Topics

1. Fundamentals of image processing. Human vision and its relationship to image display. Demonstration of simple image processing examples. The image as a 2D continuous and discrete function.
2. Image acquisition methods and devices: CCD, PSD, CT, LiDAR. Camera obscura and real cameras: intrinsic and extrinsic parameters and typical structure. Homogeneous linear transformations.
3. Harmonic basis functions. Fourier transform. Sampling and reconstruction. The light spectrum and color, fundamentals of color systems. Quantization and dithering.
4. Low-level image processing: histogram operations, linear filtering in the spatial and frequency domains. Separability. Nonlinear, statistical filters: median filter and its variants. Edge-preserving, bilateral filters.
5. Segmentation. Similarity-based segmentation, region-based methods. Edge detection. First and second derivatives: gradient and Hessian matrix. Sobel, Prewitt, and Canny algorithms. Hough transform for lines and circles.
6. Motion-based segmentation, foreground–background separation. Motion tracking, optical flow.
7. Stereo vision, epipolar geometry. Image matching, correlation techniques. Motion-based stereo.
8. Mathematical morphology. Measurements on binary images: position, orientation, length, cardinality. Neighborhoods. Skeletonization.
9. Fundamentals of artificial intelligence. Challenges and basic problems of computer vision. Supervised learning, perceptron, gradient methods, backpropagation. Multilayer networks, batch normalization.
10. Structure of convolutional neural networks; architectures: Inception, ResNet. Solutions for semantic segmentation and detection. Sequence processing, recurrent elements.
11. Challenges of intelligent vision, adversarial examples, representation learning. Combining learning paradigms, image generation, prediction, curiosity-driven learning.
12. Medical imaging devices: X-ray, CT, PET, MRI. Processing methods and applications, image registration. Motion analysis.
13. Fundamentals of remote sensing, objects as sources of radiation. Land cover identification. Basics and methods of spaceborne remote sensing.
14. Application of image processing in geographic information systems (GIS): data extraction, localization. Example: assessing the condition of green vegetation using image processing.

Practical topics

1. Fundamentals of image processing tools, general image manipulation techniques.

2. Color-based image processing operations, pixel-level operations, masking.

3. Implementation of filtering in the image and frequency domains.

4. Application of edge detection methods, detection of parametric curves using the Hough transform.

5. Application of segmentation methods based on intensity and edges.

6. Motion-based segmentation, robust motion detection in video.

7. Processing of binary images, basic morphological operations, object counting.

8. Investigation of automatic text recognition methods.

9. Object detection and tracking in video.

10. Fundamentals of neural networks, automatic differentiation, validation, and overfitting.

11. Image classification using convolutional neural networks.

12. Application of medical image processing methods.

13. Remote sensing, processing of satellite and aerial imagery.