Real-time low-pixel infrared human detection from unmanned aerial vehicles

Ignacio Martinez-Alpiste; Gelayol Golcarenarenji; Qi Wang; Jose Maria Alcaraz-Calero

doi:10.1145/3416014.3424600

Real-time low-pixel infrared human detection from unmanned aerial vehicles

Ignacio Martinez-Alpiste, Gelayol Golcarenarenji, Qi Wang, Jose Maria Alcaraz-Calero

Research output: Chapter in Book/Published conference output › Conference publication

Abstract

To improve the speed and accuracy in human detection in Search and Rescue (SAR) operations, this paper presents a novel and highly efficient machine learning empowered system by extending the You Only Look Once (YOLO) algorithm, which is designed and deployed on an embedded system. The proposed approach has been evaluated under real-world conditions on a Jetson AGX Xavier platform and the results have shown a well-balanced system in terms of accuracy, speed and portability. Moreover, the system demonstrates its resilience to perform low-pixel human detection on infrared images received from an Unmanned Aerial Vehicle (UAV) at low-light conditions, different altitudes and postures such as sitting, walking and running. The proposed approach has achieved in a constrained environment a total of 89.26% of accuracy and 24.6 FPS, surpassing the barrier of real-time object recognition.

Original language	English
Title of host publication	DIVANet 2020 - Proceedings of the 10th ACM Symposium on Design and Analysis of Intelligent Vehicular Networks and Applications
Publisher	ACM
Pages	9-15
Number of pages	7
DOIs	https://doi.org/10.1145/3416014.3424600
Publication status	Published - 16 Nov 2020

Keywords

Jetson AGX Xavier
machine learning
thermal imagery
UAV
YOLO

Access to Document

10.1145/3416014.3424600

Cite this

@inproceedings{25904de2e68149159f3c216745c96382,

title = "Real-time low-pixel infrared human detection from unmanned aerial vehicles",

abstract = "To improve the speed and accuracy in human detection in Search and Rescue (SAR) operations, this paper presents a novel and highly efficient machine learning empowered system by extending the You Only Look Once (YOLO) algorithm, which is designed and deployed on an embedded system. The proposed approach has been evaluated under real-world conditions on a Jetson AGX Xavier platform and the results have shown a well-balanced system in terms of accuracy, speed and portability. Moreover, the system demonstrates its resilience to perform low-pixel human detection on infrared images received from an Unmanned Aerial Vehicle (UAV) at low-light conditions, different altitudes and postures such as sitting, walking and running. The proposed approach has achieved in a constrained environment a total of 89.26% of accuracy and 24.6 FPS, surpassing the barrier of real-time object recognition.",

keywords = "Jetson AGX Xavier, machine learning, thermal imagery, UAV, YOLO",

author = "Ignacio Martinez-Alpiste and Gelayol Golcarenarenji and Qi Wang and Alcaraz-Calero, {Jose Maria}",

year = "2020",

month = nov,

day = "16",

doi = "10.1145/3416014.3424600",

language = "English",

pages = "9--15",

booktitle = "DIVANet 2020 - Proceedings of the 10th ACM Symposium on Design and Analysis of Intelligent Vehicular Networks and Applications",

publisher = "ACM",

address = "United States",

}

Real-time low-pixel infrared human detection from unmanned aerial vehicles. / Martinez-Alpiste, Ignacio; Golcarenarenji, Gelayol; Wang, Qi et al.
DIVANet 2020 - Proceedings of the 10th ACM Symposium on Design and Analysis of Intelligent Vehicular Networks and Applications. ACM, 2020. p. 9-15.

Research output: Chapter in Book/Published conference output › Conference publication

TY - GEN

T1 - Real-time low-pixel infrared human detection from unmanned aerial vehicles

AU - Martinez-Alpiste, Ignacio

AU - Golcarenarenji, Gelayol

AU - Wang, Qi

AU - Alcaraz-Calero, Jose Maria

PY - 2020/11/16

Y1 - 2020/11/16

N2 - To improve the speed and accuracy in human detection in Search and Rescue (SAR) operations, this paper presents a novel and highly efficient machine learning empowered system by extending the You Only Look Once (YOLO) algorithm, which is designed and deployed on an embedded system. The proposed approach has been evaluated under real-world conditions on a Jetson AGX Xavier platform and the results have shown a well-balanced system in terms of accuracy, speed and portability. Moreover, the system demonstrates its resilience to perform low-pixel human detection on infrared images received from an Unmanned Aerial Vehicle (UAV) at low-light conditions, different altitudes and postures such as sitting, walking and running. The proposed approach has achieved in a constrained environment a total of 89.26% of accuracy and 24.6 FPS, surpassing the barrier of real-time object recognition.

AB - To improve the speed and accuracy in human detection in Search and Rescue (SAR) operations, this paper presents a novel and highly efficient machine learning empowered system by extending the You Only Look Once (YOLO) algorithm, which is designed and deployed on an embedded system. The proposed approach has been evaluated under real-world conditions on a Jetson AGX Xavier platform and the results have shown a well-balanced system in terms of accuracy, speed and portability. Moreover, the system demonstrates its resilience to perform low-pixel human detection on infrared images received from an Unmanned Aerial Vehicle (UAV) at low-light conditions, different altitudes and postures such as sitting, walking and running. The proposed approach has achieved in a constrained environment a total of 89.26% of accuracy and 24.6 FPS, surpassing the barrier of real-time object recognition.

KW - Jetson AGX Xavier

KW - machine learning

KW - thermal imagery

KW - UAV

KW - YOLO

UR - https://dl.acm.org/doi/10.1145/3416014.3424600

U2 - 10.1145/3416014.3424600

DO - 10.1145/3416014.3424600

M3 - Conference publication

SP - 9

EP - 15

BT - DIVANet 2020 - Proceedings of the 10th ACM Symposium on Design and Analysis of Intelligent Vehicular Networks and Applications

PB - ACM

ER -

Real-time low-pixel infrared human detection from unmanned aerial vehicles

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this