Homework answers / question archive / Assignment 3 CPS563 - Data Visualization Requirements In this assignment, you will solve two practical and interesting problems in data visualization

Assignment 3 CPS563 - Data Visualization Requirements In this assignment, you will solve two practical and interesting problems in data visualization

Statistics

Share With

Assignment 3 CPS563 - Data Visualization
Requirements In this assignment, you will solve two practical and interesting problems in data visualization. These problems will gain valuable hands-on experience in data manipulation, data visualization, algorithm design and implementation.
You are provided the “footprint.zip” file with two folders “data_footprint” and “test_footprint” after the file decompression.
Each image in the two folders above has the following format A-B.jpg, where A is the human ID, whereas B is the sample ID. One person has many footprint samples. For example, 7-1.jpg is the 1st footprint image of person 7 and 7-3.jpg is the 3rd footprint image of person 7.
Problem 1
There are four (4) tasks to be performed as below.
a. Resize each image to 50x70 and convert it to grayscale (hint: use imresize and rgb2gray functions in MATLAB, 50: width, 70: height)
b. For all images in the “data_footprint” folder, you are required to perform Principal Component Analysis (PCA). Visualize the first 15 eigenfootprints obtained from PCA.
c. Then, you project all images to principal components obtained from PCA with the first 15 components. Then, you utilize Parallel Coordinate to visualize all images (with the corresponding human ID) in the “data_footprint” folder with the data you projected (with the first 15 components).
d. For each image in the “test_footprint” folder, you project it to principal components obtained from PCA (in b), and then visualize that “test_footprint” image with all images in “data_footprint” folder via Parallel Coordinate, again with the first 15 components. List all observations on the Parallel Coordinate results visualizing all images in the “data_footprint” folder with each image in the “test_footprint” folder.
Problem 2
There are three (3) tasks to be performed as below.
a. Resize all images in the “data_footprint” folder to the size of 60 x 60 and perform K means (K = 10) by using 3600 dimension features (raw pixels). Display all footprint of each clusters. (hint: use subplot function in MATLAB). Display the average footprint of each clusters.
b. Resize all images in the “data_footprint” folder to the size of 60 x 60 and perform PCA on all footprints to reduce the number of dimension to 100 (project all footprints to the first 100 principal components). Then, perform K means (K = 10) on the projected data (100 dimensions). Display all footprints of each clusters. Display the average footprint of each clusters.
c. Discuss the difference between the clusters obtained in (a) and (b).
What to Submit
1. A well-documented program that implements all problems in the Assignment 3. You must submit your program source code.
2. A well-written, concise project report. It should include: (a) title and names of group members; (b) the analysis of each problem; (c) the issues during the implementation; (d) the solutions to overcome the issues in (c); (e) the contribution of each individual member.
3. The powerpoint slides (maximum 20 slides) used in the Assignment grading.
For each group, you must submit the files above in a single zipped folder named after your student IDs.
Note: If you cannot submit zipped file to isidore, please change the filename extension to doc or docx and then submit it.