The recognition of my manders is done with ManderMatcher. This is a program developed by Koen Steenhoudt and myself. It can be downloaded for free through my website.
http://www.hylawerkgroep.be/jeroen/index.php?id=85
However, it is specifically conceived for S. s. terrestris, so it won't be of use for cristatus.
Apart from AmphIdent, you may want to check this out.
http://www.fupress.net/index.php/ah/article/view/17198
http://www.sciencedirect.com/science/ar ... 4115000680
Note that both papers deal with data collected within the span of a single year (or even a single season, in 2 of the 3 datasets). The first refers to a paper by Arntzen & Wallis 1999 to substantiate that the carnifex belly pattern does not change but (1) Drechsler et al. 2015 (= the description of AmphIdent) show that this is not true for cristatus (using 3 years of data) and (2) Arntzen & Wallis do not provide evidence for that statement.
While pattern changes may hamper recognition, it doesn't mean that the software is not useful! A clear difference with AmphIdent is that they are free. I have not used any of them myself (and I don't intend to enter 10,000+ records in a different system than the one I'm already using). When choosing, you may want to focus on comparing processing time and field time. Do you need to edit the photos? Do you need to photograph animals in a standard pose? Etc.