<< prev: assignment 2 up: back to handwriting recognition course

Assignment 3 and 4

Goal

Make a recognizer that transcribes a word image. The program must be executed like this:
python recognize.py wordimage.ppm
The output should be like this:


appeltaart

appelflap

appelsap

citroentaart

staart

A maximum of five lines each contain a possible transcription, in descending probability. The first word is concerned for the Top-1 performance and the edit distance. The other four are concerned for the Top-5 performance. For grading, the edit distance is most important.

Assignment 3

Make a first working version.

Assignment 4

Make a better final version.

Notes

Your grade will depend on the edit distance (50%), Top-1 recognition (30%) and Top-5 recognition(20%).
You can train your program using a part of the word labels created by you and your fellow students. They can be found in /home/student/hwr/data/words-train.
Submit your program to Nestor's Digital drop box. Your program will be tested on word images that were labeled by you and the other students. The average edit distance (Levenshtein distance) between the result and the true answer determines your grade for this assignment.
Your program will appear in a hit list, sorted by average edit distance. You can submit multiple times. The last submission before the deadline detemines your grade.
Please take care that it is important that your program does not leave temporary files. Because of the automatic testing, temporary files can easily fill up the space on my account.

Last modified: 26 May 2008, morning.