Title | Research on Recognition of Criminal Suspects Based on Foot Sounds |
Publication Type | Conference Paper |
Year of Publication | 2019 |
Authors | Liu, Fuxiang, Jiang, Qi |
Conference Name | 2019 IEEE 3rd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) |
Keywords | Acoustic Fingerprints, acoustic signal detection, composability, criminal suspect recognition, feature extraction, Fingerprint recognition, Floors, foot sounds, footstep events identification, footstep sound signal, footstep sound signals, Footwear, frequency domain feature, frequency domain features, Human Behavior, main frequency band, Mel domain features, Mel frequency cepstral coefficient, MFCC, nonfootstep event identification, peak frequencies, personnel identification, police data processing, pubcrawl, Resiliency, Source separation, Time-frequency Analysis |
Abstract | There are two main contributions in this paper: Firstly, by analyzing the frequency domain features and Mel domain features, we can identify footstep events and non-footstep events. Secondly, we compared the two footstep sound signals of the same person in frequency domain under different experimental conditions, finding that almost all of their peak frequencies and trough frequencies in the main frequency band are respectively corresponding one-to-one. However for the two different people, even under the same experimental conditions, it is difficult to have the same peak frequencies and trough frequencies in the main frequency band of their footstep sound signals. Therefore, this feature of footstep sound signals can be used to identify different people. |
DOI | 10.1109/ITNEC.2019.8729307 |
Citation Key | liu_research_2019 |