Contained is the source code for performing matched filtering detection. In principle, the algorithm is quite simple in that it computes a Pearson correlation coefficient at every sample in a time series corresponding to a template. However, the actual implementation in a compiled language is tedious.

See the installation notes.

1. Super-Efficient Cross-Correlation SEC-C - if you are a Python or Matlab user looking to do template matching on a desktop computer and dislike the headache that is compilation then this may be the way to go. There is actually a good deal of similarity since this library implements in a compiled language many of the speed-ups that the SEC-C group defined in their paper.
2. Fast-Matched-Filter - this is a solid time-domain choice. The core of this appears to be written in C so you'll likely see some pretty good performance. Additionally, it has Python and Matlab bindings.

This software is distributed under the MIT license. The details are here.