Add to Favourites
To login click here
This article discusses the challenges of audio classification in military activities and introduces a new dataset, MAD, for training and evaluating audio classification systems. The dataset contains 8,075 sound samples from 7 classes and is extracted from various military videos. The article also presents a comprehensive sound classification study using deep learning algorithms on the MAD dataset. This dataset will be a valuable resource for evaluating existing algorithms and advancing research in acoustic-based hazardous situation surveillance systems.