OpenCV C and C++ programming for NVidia Jetson TX

Duration: 5 Days

Intended Audience

This course is for experienced C/C++ programmers who also have some familiarity with CUDA who need not to get up to speed with OpenCV Programming for Computer Vision and Image Processing Applications on NVidia Jetson TX2 devices.

Course Overview

The course covers the most important aspects of the OpenCV framework and applications that can be built using this framework.

Setup and installation of OpenCV
The OpenCV framework architecture and the various C/C++ application programming interfaces and classes.
Essential aspects of image processing using OpenCV, blurring, image morphology, geometric transforms, image histograms, Segmenting Images and Pattern Matching.
Object Recognition techniques to e.g. Detect different shapes, faces, people, and learn how to train a detector to detect custom objects for e.g. face recognition.
Learn about camera calibration and stereo images processing

Course Contents

Introduction to Image Processing

Images Capture - Devices and Sensors
Image Representation and Images as Data Structures

Introduction to the Mat class

Mat class - an overview
Loading images
Traversomg Mat objects
Lookup tables
Linear and logarithmic transformation

Image Filtering

Neighborhood of a pixel
Image averaging and Image filters
Image blurring
Gaussian filtering
Image noise filtering
Vignetting

Image Thresholding

Binary images
Basic thresholding
Adaptive thresholding
Morphological operations - Erosion and Dilation

Image Histograms

Histograms - concepts and theory
Plotting histograms
Color histograms
Multidimensional histograms

Image Derivatives and Edge Detection

Image derivatives - concepts and maths
Image derivatives in two dimensions
The Sobel derivative filter
From derivatives to edges
Edge detection using the Sobel Detector
The Canny edge detectorn
Laplacians and edge detection
Blur detection

Face Detection Using OpenCV

Image classification systems
Face detection
Haar features
Cascaded classifiers

Affine Transformations and Face Alignment

Face alignment - as the first step in facial analysis
Rotating faces
Image cropping and using it for face alignment

Feature Descriptors in OpenCV

Local binary pattern (LBP) - concepts and uses
Applying LBP to aligned facial images

Machine Learning and OpenCV

Conceptual overview of machine learning
Supervised and unsupervised learning
k-means clustering
k-nearest neighbors classifier - an overview
Support vector machines (SVMs) - an overview
Non-linear SVMs and their applications
Overfitting - how to recognise it and how to avoid it
Cross-validation
Common machine learning evaluation metrics - an overview
Precision Recall - the P-R curve

First Technology Transfer