Audio-Processing-Based Human Detection at Disaster Sites With Unmanned Aerial Vehicle

Abstract

This paper describes a human search system that uses an unmanned aerial vehicle (UAV). The use of robots to search for people is expected to become an auxiliary tool for saving lives during a disaster. In particular, because UAVs can collect information from the air, there has been much research into human search using UAVs equipped with cameras. However, the disadvantage of cameras is that they struggle to detect people who are hidden in shadows. To solve this problem, we mounted an array microphone on a UAV and to detect the human voice as a means of finding people that cameras cannot. Also a search method is proposed that combines voice and camera human detection to compensate for their respective shortcomings. The rate and accuracy of human detection by the proposed method are assessed experimentally.

An Optimized Multi-Task Learning Model for Disaster Classification and Victim Detection in Federated Learning Environments

“…The studies in [20] and [21] shifted their focus from aerial view to burning building and flood scenes, respectively. Apart from the aforementioned image-based victim detection, the authors in [31] proposed an integrated audio-visual human search system, in order to boost the system performance. The works in [32], [33] took another divergent approach by locating mobile terminals based on radio frequency (RF) signal.…”

“…However, this method is effective only when user equipment and victims are in the proximity of each other. Furthermore, none of the above works [18][19][20][21][31][32][33]] consider a multitask system that concurrently strives for two coupled goals.…”

“…1. Existing studies focus on solving single-task issue of disaster classification [13,16,[27][28][29] and victim detection [18][19][20][21][31][32][33] separately. In contrast, we introduce a MTL model by attaching a disaster classification head model to the backbone of a victim detection model.…”

A Hybrid Camera System for High-Resolutionization of Target Objects in Omnidirectional Images

“…ECENT improvements in image processing techniques have advanced research on object detection and classification, and we have been engaged in extensive research in that field [1][2][3][4][5][6][7][8][9][10]. In these studies, a camera photographs an object and image processing is performed to obtain information about that object.…”

Low‐slow‐small target detection using stepped‐frequency signals in a strong folded clutter environment

“…Technological developments have resulted in the wide application of low-slow-small (LSS) targets in military and civilian fields [1]. The unmanned aerial vehicle (UAV) is one of the most common LSS targets and plays an important role in a myriad of civilian activities [2,3], such as aerial photography, environmental monitoring, and small cargo transportation. In the military field, UAVs equipped with high-resolution cameras and high-sensitivity microphones can be used to perform spy missions, and UAVs armed with explosives can even be used for low-altitude raids and terrorist attacks.…”