Reference implementation for Bloom filter-based iris indexing proposed in [1] and extended in [2,3].
This work is licensed under license agreement provided by Hochschule Darmstadt (h_da-License).
- See example_binary_template.txt file for Iris-Code data format.
- Currently, the number of enrolled subjects and block width must be powers of 2.
- Implement the split_dataset and extract_source_data methods as needed for your experiment and data filenames.
- Add result processing code as needed.
bloomfilter.py [-h] [-v] -d DIRECTORY -n ENROLLED -bh HEIGHT -bw WIDTH -T CONSTRUCTED -t TRAVERSED
required named arguments:
- -d DIRECTORY, --directory DIRECTORY : directory where the binary templates are stored
- -n ENROLLED, --enrolled ENROLLED : number of enrolled subjects
- -bh HEIGHT, --height HEIGHT : filter block height
- -bw WIDTH, --width WIDTH : fitler block width
- -T CONSTRUCTED, --constructed CONSTRUCTED : number of trees constructed
- -t TRAVERSED, --traversed TRAVERSED : number of trees traversed
optional arguments:
- -h, --help : show this help message and exit
- -v, --version : show program's version number and exit
- [1] Christian Rathgeb, Frank Breitinger, Harald Baier, Christoph Busch, "Towards Bloom Filter-based Indexing of Iris Biometric Data", In Proceedings of the 8th IAPR International Conference on Biometrics (ICB'15), 2015.
- [2] Pawel Drozdowski, Christian Rathgeb, Christoph Busch, "Bloom Filter-based Search Structures for Indexing and Retrieving Iris-Codes", in IET Biometrics, 2017.
- [3] Pawel Drozdowski, Christian Rathgeb, Christoph Busch, "Multi-Iris Indexing and Retrieval: Fusion Strategies for Bloom Filter-based Search Structures", in Proc. International Joint Conference on Biometrics (IJCB), Denver, USA, October 2017.