Requirements

In this assignment, you will solve practical and interesting problems in data visualization. The problems will gain valuable hands-on experience in data manipulation, data visualization, algorithm design and implementation.

Problem Description

You are provided the “places.zip” file with two folders “Images”, “Test” and “spider_plot” function after the file decompression. Each image in the two folders is in jpg format.

1. You resize all images in both folder to the size of 64x64 and convert them to grayscale images (hint: use imresize and rgb2gray functions).
2. For each resized image (64x64) in the “Test” folder, you plot the test image with the most similar image in the “Images” folder (hint: you concatenate all image pixels to form 4096-dimension feature for each image. Then you compute the distance between each test image and all “Images” images. Next, you plot the “Images” image with the smallest distance to the test image).
3. For all resized images (64x64) in the “Images” folder, you are required to perform Principal Component Analysis (PCA). Visualize the first 20 eigenvectors obtained from PCA (hint: use subplot function in MATLAB).
4. You use 20 eigenvectors obtained from Question c to project all resized images (64x64) in the “Images” and “Test” folders to 20-dimension space. In 20-dimension space, you are asked to plot each test image with the most similar image in the “Images” folder (hint: Here each image