Planeamento
Aulas Laboratoriais
Lab 1
Introduction to Simulink for image processing. HSV space to sort objects by color. (inglês)Introduction to Simulink software for processing images captured by webcams. . Using the HSV space to sort objects by color.
Lab 2
Practice the Point Processing of Gray level Images” with Simulink.
Lab3: Practice on mathematical morphology
Use of different morphological based techniques for counting the number of spare parts on a pallet.
Lab 4:Shape based objects description and classification
Implementation of a real time shape based classifier
Practice in linear filtering
Practice in linear filtering for image smoothing and enhancement in real time
Image Description Based on Texture. Interest Points.
Practice with features for grayscale images in particular those based on interest points. Comparision between different methods to find interesting points. Classification based on features associated to these points. Experiments of real time classification of objects with simulink.
Grayscale images features
Use of SURF descriptors and a group specific descriptors for classification and tracking.
Conclusion of Project 1
Finalization of the works corresponding to Project 1
Planar Projective transforms
Practice with Planar Projective transforms. Application to image mosaicing
Practice of cameras calibration
Practice of cameras calibration. Measuring Planar Objects with a Calibrated Camera.
Introduction to stereo vision
Calibration of a stereo pair. Image rectification. Depth from disparity. A complete disparity.map. Reconstruction of 3D scenes.
Practice stereo vision with uncalibrated cameras.
Estimation the movement : estimation the relative position between cameras
Other applications of computer vision
Implementation of different matlab examples.
Conclusions
Conclusion of ongoing projects.
Aulas de Problemas
Introduction to Simulink for image processing. HSV space to sort objects by color.
Introduction to Simulink software for processing images captured by webcams. Using the HSV space to sort objects by color.
Practice with binary mathematical morphology
Use the Euler number and the detection of the connected components for counting the number of spare parts on a pallet.
Operations on binary shapes
Use of different morphological based techniques for counting the number of spare parts on a pallet.
Shape based objects description and classification
Evaluation of shape based object descriptors for objects characterization and classification.
Linear filters
Practice in linear filtering for image smoothing and enhancement.
Image Description Based on Texture. Interest Points.
Practice with features for grayscale images in particular those based on interest points. Comparision between different methods to find interesting points. Classification based on features associated to these points.
Grayscale images features
Use of SURF descriptors and a group specific descriptors for classification and tracking.
Conclusion of Project 1
Finalization of the works corresponding to Project 1
Planar Projective transforms
Practice with Planar Projective transforms. Application to image mosaicing
Practice of cameras calibration
Practice of cameras calibration. Measuring Planar Objects with a Calibrated Camera.
Introduction to stereo vision
Calibration of a stereo pair. Image rectification. Depth from disparity. A complete disparity.map. Reconstruction of 3D scenes.
Practice stereo vision with uncalibrated cameras
Estimation the movement : estimation the relative position between cameras
Other applications of computer vision
Implementation of different matlab examples.
Conclusions
Conclusion of ongoing projects.
Aulas Teóricas
Introduction to the course. Introduction to image processing and computer vision.
Introduction to the course, assessment methods and bibliography. Introduction to image processing and computer vision. Color Spaces. Conversion Between Color Spaces. Representation of Images in Grayscale. Presention of the problems for the "problems and laboratory" classes of this week.
Point Processing of images.Image binarization. Introduction to mathematical morphology
Point Processing of images. Histogram based operations. Automatic and manual image binarization. Otsu's method. Concept of shape. Characterization of forms. Introduction to mathematical morphology. Morphological binary operations: erosion, dilation, opening and closing.
Mathematical morphology
Continuation of the study of mathematical morphology for binary images. Boundary extraction; Region filling; Hit-or-Miss transformation; Skeletonization; Morphological reconstruction; Convex Hull; Euler Number; Ultimate erosion
Image Description Based on Shape
Introduction to image description. Shape as a region: area, Euler number, eccentricity, geometric moments,invariant moments. Descriptors based on the shape skeleton.Countour based descriptors: perimeter, shape signatures, chain code, Fourier descriptors. Introduction to classifiers.
Spatial Linear and no linear filters for gray level images
Introduction. Types of image noise. Mechanics of spatial filtering.Smoothing spatial filters: type of filters. Sharpening spatial filters: image gradient, smoothed derivative, Image Laplacian, Laplacian of Gaussian filter, high-boost filtering.Brief introduction to non linear filters: median filter; morphological gray level filters.
Image Description Based on Texture (Part 1)
Definition of texture. Texture primitives.Texture descriptors: statistical, structural and spectral approach.The auto-correlation. Frequency Descriptors.Co-Occurrence matrices. Law’s texture energy measures. Patches descriptors from image interest points. Methods to find interest points. SIFT algorithm overview: evaluate keypoints and build keypoints descriptors.
Grayscale images features (continuation
Binary descriptors: BRIEF, ORB, BRISK, FREAK. Performance Evaluation of Local Descriptors. Presentations by students of other types of descriptors.
An industrial approach to image processing
Invited session with a representative of Infaimon: new trends in hardware and software to computer vision in industry.
Projective Geometry for Computer Vision
Basics of Projective Geometry.Projective spaces P1 and P2: Points at infinity, Homography,Cross-Ratio Invariance.Duality between lines and points in P2. Introduction to conics. Types of transformations.
Camera Models and Calibration
Camera as a projective device:pinhole model.Other types of projections. Calibration matrix.Intrinsec and extrinsec camera parameters. Camera calibration: basic equations.The Direct Linear Transformation algorithm.
Introduction to stereo vision
Inferring 3D from 2D. Stereo principle. Depth from triangulation. Image rectification. Disparity. Depth from disparity. A complete disparity.map. Reconstruction of 3D scenes. Introduction to epipolar geometry. Examples.
Epipolar Geometry
Epipolar Geometry: points, lines and planes. Epipolar equations. The homography matrices H and G. The essencial matrix E. Pose Recovery from the Essential Matrix. The fundamental matrix F. Estimate the Essential Matrix from the Fundamental Matrix. Computing the fundamental matrix. The correspondence problem. Examples.
Revisions
Revisions
Presentations
Presentation of lab 13 results