Enhancing Gun Detection With Transfer Learning and YAMNet Audio Classification

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 58940 - 58949

Published: Jan. 1, 2024

Identification of the type gun used is essential in several fields, including forensics, military, and defense. In this research, one powerful deep learning architectures applied to identify types firearms based on their gunshot noises. For purpose extracting features from audio data, suggested technique makes use YAMNet, an effective learning-based classification model. The Mel spectrograms created collected are for multi-class classification, which it possible different guns. 1174 samples 12 distinct weapons make up study's extensive dataset, offers a varied representative collection training evaluation. We achieve remarkable accuracy 94.96% by employing best hyperparameter changes optimization methods. findings study substantial contribution domains defense, where precise identification crucial. Applying mel analyze demonstrates itself be promising strategy, providing quick accurate categorization. This research emphasizes effectiveness relevance using AI-driven model, as superior answer issues real-world weapon detection.

Language: Английский

Gaussian process regression coupled with mRMR to predict adulterant concentration in cocaine

Journal of Pharmaceutical and Biomedical Analysis, Journal Year: 2024, Volume and Issue: 248, P. 116294 - 116294

Published: June 7, 2024

Language: Английский