|
Wk |
Date |
Lecture Topic
and |
Assignments | |
|
Problems |
Exercise Due | |||
|
1 |
23 Aug |
Introduction-What are some of the problems? Enhancement (contrast, sharpening, smoothing), image interpretation, feature identification (edge detection, scale and rotation invariance) segmentation, color data representation, and getting image data |
|
|
|
2 |
28 Aug |
Digital image representation, Image file formats of interest (bmp, arf, png) See sample code AView1, AMain1, and test images arf, bmp1, bmp2, and bmp3. In addition, b1, b2, b3, b4, and b5. |
Lab 1: Using the Java awt, open, read, and render an arf file or a bmp file as an image. |
|
|
30 Aug |
Basic tools
|
Extensions to Lab 1: Develop code to zoom in or out and store the image. Load a second image and fade from the first image to the second. |
| |
|
3 |
4, 6 Sep
|
Getting pixel data from a Java Image object Image Basic
image statistics (mean, standard deviation, variance, and others),
Histograms (by image and by color)
|
|
|
|
4 |
11, 13 Sep |
Image statistics (continued): SNR Estimating noise by observing regions of uniformity |
Lab 2:Scan an image twice without moving the source and characterize the differences in the scanned images. Construct a color-by-color histogram for the images and acquire the basic statistics for each color plane |
Lab 1 |
|
Conversion from color to monochrome, Contrast enhancement |
|
| ||
|
5 |
18, 20 Sep
|
Creating a basic filter using a convolution kernel, smoothing filters |
Lab 3: Construct the software infrastructure to support convolution masks and test for a smoothing filter |
Lab
2 |
|
Smoothing, Sharpening, Speckle (median) filtering |
|
Lab
3 | ||
|
6 |
25, 27 Sep |
Edge detection, boundary contour identification, Prewitt, Roberts, Sobel, and Laplacian filters |
Lab 4: Add the necessary masks and infrastructure to apply filters to identify boundary contour points |
|
|
Flexible lab/work day |
|
| ||
|
7 |
2, 4 Oct |
Buffer and Review |
|
Lab 4 |
|
9 Oct
11 Oct |
Fall
Vacation
(begins Saturday, 6 October 2007 and ends Wednesday, 10 October 2007 when classes resume) Class resumes 11 October 2007 |
|||
|
|
Oct 16, 18 |
Test
1 · sample data, pixel representation, split color image into image planes, join color planes to form a color image · histograms, cumulative histograms, probability density functions · techniques for reversing, zooming, and fading · interpolation formulae · common image statistics and their uses, interpreting statistics · embedding geometric shapes within image data · contrast enhancement by stretching or histogram equalization · relative merits of basic smoothing filters |
sharpening and speckle filters will be covered after Test 1 |
|
|
|
|
Steganography and digital watermarking · Steganography and digital watermarking · IEEE ref |
|
|
|
|
30 Oct, 1 Nov |
Compression
|
Lab 6: Implement Huffman encoding and decoding. Apply Huffman encoding to the image containing the steganographic message. Decode the message file to reproduce the image and recover the message |
Lab
5 |
|
11 |
Nov 6, 8 |
Wavelet transform
|
|
|
|
Wavelet transform (continued)
|
Lab 7: Implement the single-band two-dimensional, DWT |
Lab
6 | ||
|
12 |
Nov 13, 15
|
Morphological image processing
|
|
|
|
Morphological image processing (continued)
|
|
| ||
|
13 |
Nov 20 |
Geometric Transformations
Representations:
|
|
Lab 7 |
|
|
Nov 27, 29 |
Segmentation |
|
|
|
Matching methods
|
Lab 8: Segment a color image based on RGB and HSI |
| ||
|
15 |
Dec 4 |
Neural network classifiers— |
|
|
|
Lab 9: Implement matching and correlation classifiers and compare their performances on an object recognition task |
| |||
|
|
|
Reading Day, Thursday, 6 Dec 2007 |
|
Lab 9 |
|
|
11 Dec 2007 |
Final Exam, Tuesday, 7:00-10:00 PM |
|
|
