Expected learning outcomes

This course develops an understanding of the fundamental characteristics of digital systems used in imaging, together with general concepts of science, quantitative methods. This course covers basic algorithms for image manipulation, characterization, filtering, segmentation, feature extraction and template matching in direct space and Fourier space. The course provides the opportunity for students to explore a range of practical techniques, by developing their own simple processing functions using library facilities and tools such as Matlab.

On completion of this course the student will be able to:

• Understand (i.e. to describe, analyse and reason about) how monochrome digital images are represented, manipulated, encoded and processed, with emphasis on algorithm design, implementation and performance evaluation. methods of capturing and reproducing images in digital systems.
• Understand (i.e. to describe, analyse and reason about) how color digital images are represented, manipulated, encoded and processed.
• Make appropriate use of mathematical techniques in colour imaging. Demonstrate the use of tools such as spreadsheets and specialist maths applications to solve problems in colour imaging

Topic(s)

1. Digital image acquisition: analogue to digital conversion. Sampling and quantization. Look-up table conversions. Scaling.
2. Digital image formats: representation and description. Image encoding and image compression.
3. Image filtering: linear and non-linear filtering operations. Image convolution. Separable convolutions. Image enhancement. Image restoration.
4. Digital image processing: histogram manipulation. Thresholding. Image segmentation. Clustering techniques. Split and merge algorithms. Region processing. Edge detections. Region adjacency graph.
5. Image transformations: histogram equalization, geometric transformations, affine transformations, polynomial warps.
6. Digital image analysis: noise analysis. Texture analysis. Fourier descriptors. Features extraction. Pattern recognition. Corner detection. Saliency maps. Image interpretation. Motion analysis.
7. Color image analysis : representation, encoding, scalar and vector approaches. Clustering techniques. Color invariants. Color constancy algorithms.
8. Template matching: Similarity and dissimilarity metrics. Cross-correlation. Multiresoultion algorithms. Graph matching. Image retrieval. 2D object detection, recognition and Positioning.
9. High level image descriptors. Semantic image description MPEG7.

Teaching Methods

Lectures
Laboratory work
Net Support Learning

Teaching Methods (additional text)

Lectures by the course teacher and guest lecturers.

Lab sessions and homework assignments. Two or three of the homework assignments will be graded.

E-learning material: lectures notes in PDF and audio recordings of the lectures and important exercises.

Form(s) of Assessment

Written exam, 3 hours
Exercises

Form(s) of Assessment (additional text)

• Written exam (60%)
• Homework assignments (two or three of the homework assignments will be graded.) (40%)

• Each part must be individually approved of.

Grading Scale

Alphabetical Scale, A(best) – F (fail)

External/internal examiner

Internal examiner for the written exam and exercises.

Re-sit examination

Written exam: ordinary re-sit examination

Examination support

English translation dictionary.

Teaching Materials

Course book:

• Digital Image Processing, 3rd Edition (DIP/3e), by Rafael C. Gonzalez and Richard E. Woods, Prentice Hall (2008)

Further reading material:

• Digital Image Processing Using MATLAB (DIPUM), by Rafael C. Gonzalez, Richard E. Woods, and Steven L. Eddins, Prentice Hall (2004).
• Color Image Processing: Methods and Applications (Image Processing), by Rastislav Lukac & Kostantinos N. Plataniotis, CRC (2006)
• The Image Processing Handbook, Fifth Edition (Image Processing Handbook), by John C. Russ, CRC (2006)

Replacement course for

Partial overlap with IMT4401 Digital Image Reproduction