Introduction to Computer Vision
Start Date: Please contact us
Course Overview
Basic filters, edge detectors, feature extractor, object (face) identifier, optical flow and additional subjects.
The students are experiencing this field by coding in matlab and python with opencv
Course Outline:
Image processing & Matching
1. Introduction to OpenCV with Python
• Installation / API
2. Basic Operators
• Median, Box, common neighbors
• convolution and kernel filters
• Coding example: filtering an image and seeing results
• Segmentation and thresholding methods
• Morphological operators: dilate erode
• Coding example: dilate/erode showing results and solving a basic problem
• Connected components and labeling
3. Edge /Corner / Line detectors
• Sobel
• Canny
• Roberts
• Laplacian
• Hough transform
• Coding example: running Sobel vs Canny and watching results
4. Image Matching
• Harris
• Scale Invariant – why??
• SIFT
• Advance Lab
– SIFT
– Effects of different params/config (bins, scaling, best match vs NN)
– Effects of Noise in the image
• SURF
• Object detection – Theory
• Face detection – Viola jones Haar Filters & Integral Image
• HoG6. Mapping transforms -optional
• Theory: Translation, Rotation, Rigid body, affine perspective
• Lab OpenCV transformations7. 3D understanding
• Camera Projection theory
• Two cameras
• Structured light
8. Optical flow and tracking
• Lucas-Kanade Theory
• Code Review in OpenCV ( Link ) & Applications
9. Deep Learning Intro
• Overview of the technology
• Tools like Keras & TensorFlow
10. Summary Exercise
• Processing path: Image processing & scaling->Computer vision feature extraction->Machine Learning classifier