This course provides an overview of the essential techniques of processing, analysis and interpretation of images. The course is structured according to the complexity of the information extracted from the images and the level of interpretation of the scene. The first two blocks present useful models for the initial extraction of information from a single image, in particular the models of vector space, mathematical morphology and level sets. The process of interpreting the images generally involves a step of grouping the previously extracted information. The techniques related to this medium/high vision process combine segmentation and detection algorithms that are the content of the third and fourth block of the subject. Finally, the last block presents the necessary tools to analyze information from video sequences.
Person in charge
Javier Ruiz Hidalgo (
Philippe Salembier Clairon (
CE5 - Design and apply techniques of signal processing, choosing between different technological tools, including those of Artificial vision, speech recognition and multimedia data processing.
CT6 - Autonomous Learning. Detect deficiencies in one's own knowledge and overcome them through critical reflection and the choice of the best action to extend this knowledge.
CB5 - That the students have developed those learning skills necessary to undertake later studies with a high degree of autonomy
Generic Technical Competences
CG1 - To design computer systems that integrate data of provenances and very diverse forms, create with them mathematical models, reason on these models and act accordingly, learning from experience.
CG2 - Choose and apply the most appropriate methods and techniques to a problem defined by data that represents a challenge for its volume, speed, variety or heterogeneity, including computer, mathematical, statistical and signal processing methods.
CG5 - To be able to draw on fundamental knowledge and sound work methodologies acquired during the studies to adapt to the new technological scenarios of the future.
Acquire basic knowledge of frequency representation and advanced image filters.
Understand the use of tools for geometric processing.
Understand how to use object segmentation and detection techniques.
Acquire the basic knowledge of motion estimation and tracking.
Filtering and Frequency Analysis
Frequency representation: FT, DFT
Advanced filters: linear, non-local, bilateral
Multiscale image analysis I: Downsampling / Upsampling, Interpolation, pyramid, wavelets & CNNs
Mathematical morphology and lattice
Filters by reconstruction
Variational model and level sets
Transition-based segmentation: Edge detection, Active contour
Homogeneity-based segmentation: Classification, Region growing & Watershed
Object recognition: Local features, Bag of words Region proposals, Regression
Motion estimation, Optical flow
Theory, exercise and laboratory classes corresponding to Unit 1
Theory: Theory classes corresponding to Unit 1
Laboratory: Laboratory classes corresponding to Unit 1
Guided learning: Driven learning corresponding to Unit 1
Autonomous learning: Autonomous learning corresponding to Unit 1
The subject is based on classroom theory classes and laboratory. The theory classes follow the program defined in this teaching guide. Within the lectures, the dialogue between professors and students is promoted by proposing exercises and activities to be carried out jointly based on particular aspects of the topic being dealt with. The laboratory classes exemplify the contents developed in the theory classes.
Final mark is obtained from:
- Parcial exam: P (20%)
- Final exam: F (50%)
- Laboratory: L (30%)