Study guide for Test #1 Image Analysis and Computer Vision ECE 438

You will have 75 minutes for the test. You will be allowed to use: 1) textbook last 30 minutes (no passing), 2) calculator (your own, you cannot share during the test!)

The test will cover, in general:

1) Lectures, 2) Homework – thru 4.21, 3) textbook – Chapters/sections 1, 2, 3, 4.1, 4.2, 11, 4) Lab exercises

Notes: Most of the test material will be from lecture and homework. When you take the test, work smart – be sure to work the problems you know first. Difficult problems are worth more points.

TOPICS COVERED:

Computer Imaging Introduction

Ø definitions and relations between image processing, computer vision, image analysis

Ø typical system setup, applications

Ø digital image representation

Imaging Systems

Ø Video signals, interlaced, noninterlaced, frame grabber, synch pulse

Ø Camera interface specs, analog vs. digital

Image formation

Ø sensors, sensor equation, photons, EM waves

Ø lenses, blur equation

Ø focal length, f‑stop, field of view

Ø irradiance/radiance

Ø EM spectrum, visible, x-ray, IR, UV etc

Ø quantum efficiency

CVIPtools

Ø Various windows, viewer, options, functionality

Image representation

Ø I(r,c), pixel, vector, matrix, image, file

Ø Optical, digital, binary, gray-scale, color,, multispectral

Ø Color transforms:HSL, SCT, CCT, CIE, chromaticity, YCbCr, YUV, CMY

Ø File formats: graphics, bitmap images, vector images, key points, rendering, file headers, LUT, remapping, file types

Image Analysis

Ø Block diagram/system model: preprocessing, data reduction, feature analysis

Ø Data reduction: spatial domain via segmentation, or spectral domain via transform, followed by filtering and data extraction

Preprocessing

Ø Geometry: ROI, crop, zoom, enlarge, shrink translate, rotate, zero and first order hold, convolution

Ø Arithmetic/logic operations: Add, subtract, multiply, divide applications, AND/OR/NOT/XOR applications, masking, morphing, background subtraction

Ø Spatial filters: mean, median, enhancement, linear, nonlinear, typical coefficients

Ø Quantization: spatial and gray level, dithering variable and uniform bins, anti-aliasing filters

Binary Image Analysis

Ø Thresholding via histogram, isodata/k-means clustering

Ø Basic binary features: center of area, area, projections, axis of least 2nd moment, Euler number

Ø Connectivity types – 4,8,6; neighbors – vertical, horizontal, diagonal; connectivity dilemma – solutions:4/8,8/4,6; labeling

Edge/line detection

Ø Define segmentation, edge, line, ideal edge vs. real edge

Ø Gradient operators, 1st and 2nd derivative, magnitude & direction, Roberts, Sobel, Prewitt, laplacian

Ø Compass masks: Kirsch, Robinson

Ø Advanced edge detectors, LoG, Canny, Shen-Casten, Boie-Cox, Frei-Chen, nonmaxima suppression, hysteresis thresholding

Ø Edges in color images

Ø Edge detector performance metric: Pratt FOM, distance measures, city-block, chessboard, Euclidean

Ø Noisy images: prefilter, extend mask, use Canny, truncated pyramid

Ø Hough transform: line finding, quantization (line) search space, algorithm, edge linking (snake eating algorithm)

Ø Corner detection: Moravec, Harris, Frei-Chen

Lab

Ø labs 1‑4, CVIPlab, adding a function