• Corpus ID: 36431195

Flood-survivors detection using IR imagery on an autonomous drone

  title={Flood-survivors detection using IR imagery on an autonomous drone},
  author={Sumant Sharma},
In the search and rescue efforts soon after disaster such as floods, the time critical activities of survivor detection and localization can be solved by using thermal long-wave infrared (LWIR) cameras which are more robust to illumination and background textures than visual cameras. This particular problem is especially challenging due to the limited computational power available on-board commercial drone platforms and the requirement of real-time detection and localization. However, the… 

Figures from this paper

Real-Time Low-Pixel Infrared Human Detection From Unmanned Aerial Vehicles
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, surpassing the barrier of real-time object recognition.
Flood Rescue Operations Using Artificially Intelligent UAVs
The findings indicate that the UAV is capable of performing search and rescue operations with tremendous accuracy and that too in realtime.
Guidelines for the Use of Unmanned Aerial Systems in Flood Emergency Response
Those organizations responsible for overseeing flood risk management activities including the response to flooding events will have to keep abreast of the rapid technological advances in UAS if they are to be used to best effect.


Low Resolution Person Detection with a Moving Thermal Infrared Camera by Hot Spot Classification
A two-stage person recognition approach for LWIR images is proposed: the application of Maximally Stable Extremal Regions (MSER) to detect hot spots instead of background subtraction or sliding window and the verification of the detected hot spots using a Discrete Cosine Transform based descriptor and a modified Random Naïve Bayes (RNB) classifier.
An effective approach to pedestrian detection in thermal imagery
An integrated algorithm to detect humans in thermal imagery was introduced, designed to meet the shortfall of the HOG in infrared imagery, and the pedestrians in the thermal imagery were detected.
Human Detection Based on the Generation of a Background Image by Using a Far-Infrared Light Camera
A new method for detecting human areas in thermal camera images is proposed and Experimental results with two types of databases confirm that the proposed method outperforms other methods.
Hot spot detection and classification in LWIR videos for person recognition
A solution for LWIR videos coming from a moving camera is presented that is based on hot spot classification to distinguish persons from background clutter and other objects and it is demonstrated that taking advantage of this temporal context can improve the classification performance.
Background-subtraction using contour-based fusion of thermal and visible imagery
Feature based person detection beyond the visible spectrum
  • K. Jüngling, Michael Arens
  • Computer Science
    2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
  • 2009
This paper shows the applicability of a well known, local- feature based object detector for the case of people detection in thermal data and shows how this local-feature based detector can be used to recognize specific object parts, i.e., body parts of detected people.
Robust Video-Based Surveillance by Integrating Target Detection with Tracking
A framework for video-based surveillance where detection and tracking are addressed simultaneously in a unified framework to improve detection results and improve system robustness is presented.
Thermal-Visible Video Fusion for Moving Target Tracking and Pedestrian Classification
The paper presents a fusion-tracker and pedestrian classifier for color and thermal cameras that detects, in human gait, a symmetrical double helical pattern that can then be analyzed using the Frieze Group theory.
Integral Channel Features
It is demonstrated that when designed properly, integral channel features not only outperform other features including histogram of oriented gradient (HOG), they also result in fast detectors when coupled with cascade classifiers.
On the Optimality of the Simple Bayesian Classifier under Zero-One Loss
The Bayesian classifier is shown to be optimal for learning conjunctions and disjunctions, even though they violate the independence assumption, and will often outperform more powerful classifiers for common training set sizes and numbers of attributes, even if its bias is a priori much less appropriate to the domain.