Multi-feature Digit Dataset

From Chorus
Revision as of 14:54, 30 July 2010 by Lidy (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search
Multi-feature Digit Dataset
Domain Image Recognition
Media Image
Size
Instances 2000
File Format ASCII
Creation Date
Task Handwriting Recognition
Copyright
URL http://archive.ics.uci.edu/ml/datasets/Multiple+Features


Description

This dataset consists of features of handwritten numerals (`0'--`9') extracted from a collection of Dutch utility maps. 200 patterns per class (for a total of 2,000 patterns) have been digitized in binary images.

The source image dataset is lost. Using the pixel-dataset (mfeat-pix) sampled versions of the original images may be obtained (15 x 16 pixels).

Quality

Source

Owned and donated by:

Robert P.W. Duin Department of Applied Physics Delft University of Technology P.O. Box 5046, 2600 GA Delft The Netherlands


Ground Truth Annotation

10 classes: handwritten numerals (`0'--`9')


Features

These digits are represented in terms of the following six feature sets (files):

  1. mfeat-fou: 76 Fourier coefficients of the character shapes;
  2. mfeat-fac: 216 profile correlations;
  3. mfeat-kar: 64 Karhunen-Love coefficients;
  4. mfeat-pix: 240 pixel averages in 2 x 3 windows;
  5. mfeat-zer: 47 Zernike moments;
  6. mfeat-mor: 6 morphological features.

In each file the 2000 patterns are stored in ASCI on 2000 lines. The first 200 patterns are of class `0', followed by sets of 200 patterns for each of the classes `1' - `9'. Corresponding patterns in different feature sets (files) correspond to the same original character.

number of attributes: 649 (distributed over 6 datasets,see above) (Some are integer, others are real; no missing attributes)

Attributes are SPACE separated and can be loaded by Matlab as > load filename


Copyright Remarks

Citation

M. van Breukelen, R.P.W. Duin, D.M.J. Tax, and J.E. den Hartog, Handwritten digit recognition by combined classifiers, Kybernetika, vol. 34, no. 4, 1998, 381-386. http://rexa.info/paper/76afd3f8caf9d74d3a3845ad5f7a5517bca39a64

M. van Breukelen and R.P.W. Duin, Neural Network Initialization by Combined Classifiers, in: A.K. Jain, S. Venkatesh, B.C. Lovell (eds.), ICPR'98, Proc. 14th Int. Conference on Pattern Recognition (Brisbane, Aug. 16-20),

A.K. Jain, R.P.W. Duin, J. Mao, Statisitcal Pattern Recognition: A Review, in preparation


External Links

http://archive.ics.uci.edu/ml/datasets/Multiple+Features

Personal tools
CHORUS+