Kaggle: Denoising Dirty Documents with MATLAB

Starter code for completing the Kaggle competition with MATLAB.

This file was created for folks who are interested in using MATLAB for the Kaggle data science competition called Denoising Dirty Documents. Specifically, it contains a useful function for converting image data into the required csv format for submission.

See the following link for more information about the competition, including submission file protocol:
https://www.kaggle.com/c/denoising-dirty-documents

The zip file contains:
1. im2csv.m -- a function that converts an input image into comma-separated value data. Optional parameters include the image ID (for this competition, the ID is equivalent to the image filename), the output filename (where to store the CSV data), and '-append' (whether to add the data to an existing file rather than creating a new file).

2. submission_raw.m -- a script that demonstrates the use of im2csv. Each image file in the test directory is read and individually converted to CSV data. When the first image is converted, an output file called raw.csv is created. For subsequent images, data is appended to this file.