fuzzy-utils If you have chosen to use the second hasing method (i.e. BerkeleyDB), then you also have the following at your disposal: fuzzy-find [--delete] hash|filename This utility searches the DB files for the digest hash (from debug output) and prints out the information stored in the DB file. It is usefull to check how many times an image has matched. Also, with the optional --delete option, you can remove the image from the DB file fuzzy-stats [days] [images] This utility generates basic statistics taken from the DB files regarding matches made [days] ago (0 by default == today) and shows some information of the top [images] (5 by default) fuzzy-cleantmp [hours] If you decide to keep debug information of the work SpamAssassin is doing, tempfiles tend to pile up. This utility can help by removing 'old' files (older than [hours] hours (12 by default)). This can safely be placed in CRON to keep your tempdir manageable TODO: Right now, the names for each DB file is hardcoded. If you change the default settings in FuzzyOcr.cf, you must update these utilities so that they can process the 'right' files.