Abstract
Although viable tools are available for the identification of unknown deceased individuals, recognition rates with these methods are greatly impacted by the degree to which decomposition has occurred. Therefore, identifying highly decomposed remains poses a major challenge. This paper analyzes the effect of facial decomposition on the recognition rates of several facial recognition commercial-off-the-shelf systems and research-grade systems, as well as algorithms contained in a custom recognition library. The custom dataset of facial images used in the experiment is composed of 42 subjects at stages of decomposition ranging from recently deceased to later stages where the soft tissues are severely decomposed and facial features are deformed. It is shown that an algorithm's ability to correctly detect a decomposing face is a crucial first step that not all face models can accurately handle. However, some of the evaluated Convolution Neural Network (CNN)-inspired methods provide promising results even in cases of severely decomposed faces.
| Original language | English |
|---|---|
| Title of host publication | 2019 IEEE 10th International Conference on Biometrics Theory, Applications and Systems, BTAS 2019 |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| ISBN (Electronic) | 9781728115221 |
| DOIs | |
| State | Published - Sep 2019 |
| Event | 10th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2019 - Tampa, United States Duration: Sep 23 2019 → Sep 26 2019 |
Publication series
| Name | 2019 IEEE 10th International Conference on Biometrics Theory, Applications and Systems, BTAS 2019 |
|---|
Conference
| Conference | 10th IEEE International Conference on Biometrics Theory, Applications and Systems, BTAS 2019 |
|---|---|
| Country/Territory | United States |
| City | Tampa |
| Period | 09/23/19 → 09/26/19 |
Funding
This research was supported in part by an appointment to the Oak Ridge National Laboratory Post-Bachelor’s Research Associate Program, sponsored by the US Department of Energy and administered by the Oak Ridge Institute for Science and Education. This manuscript has been authored in part by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).