HOME Representation preprocessing Representations Representation by pixelsRepresentation by features

Representation by features

The commands in the table can be used to find some simple features to represent images. Unless otherwise stated these are fixed mappings. All these mappings operate on datafiles as well as on datasets in which object images are stored. Note the feature values found for most of the routines are image size dependent. Users may normalize image sizes by im_resize if appropriate.

> Measurement of Image Features
histm Convert images to histograms (trainable)
im_harris* Find Harris points in images
im_moments Computes moments as features from object images
im_mean Computes center of gravity
im_measure* Computes some measurements
im_profile Computes image profiles
im_stat Compute some simple statistics
filtm General mapping for applying user defined routines to all objects in the dataset or datafile
filtim General mapping for applying user defined routines to all images in the dataset or datafile

*The DIPimage package should be in the path for this command.


R.P.W. Duin, January 28, 2013


HOME Representation preprocessing Representations Representation by pixelsRepresentation by features