• Corpus ID: 242757572

Extended Abstract Version: CNN-based Human Detection System for UAVs in Search and Rescue

@article{Mesvan2021ExtendedAV,
  title={Extended Abstract Version: CNN-based Human Detection System for UAVs in Search and Rescue},
  author={Nikite Mesvan},
  journal={ArXiv},
  year={2021},
  volume={abs/2111.02870}
}
This paper proposes an approach for the task of searching and detecting human using a convolutional neural network and a Quadcopter hardware platform. A pre-trained CNN model is applied to a Raspberry Pi B and a single camera is equipped at the bottom of the Quadcopter. The Quadcopter uses accelerometer-gyroscope sensor and ultrasonic sensor for balancing control. However, these sensors are susceptible to noise caused by the driving forces such as the vibration of the motors, thus, noise… 

Figures and Tables from this paper

References

SHOWING 1-9 OF 9 REFERENCES

Human detection in uncluttered environments: From ground to UAV view

The very well known HOG detector is adapted for UAV use and a new kind of training dataset is proposed in order to increase the detector's angular robustness and to reduce the search space and consequently the computation time.

Real-time people and vehicle detection from UAV imagery

An approach for the automatic detection of vehicles based on using multiple trained cascaded Haar classifiers with secondary confirmation in thermal imagery based on a similar cascaded classification technique combining additional multivariate Gaussian shape matching.

Robust U-Net-based Road Lane Markings Detection for Autonomous Driving

  • Le-Anh TranM. Le
  • Computer Science
    2019 International Conference on System Science and Engineering (ICSSE)
  • 2019
A new method is proposed to detect road lane markings for supporting surveillance and autonomous driving, constructed based on U-Net architecture, a convolutional neural network developed for biomedical image segmentation, and Hough Transform method is implemented.

A Vision-based Method for Autonomous Landing on a Target with a Quadcopter

A new methodology is developed, including two major tasks, for vision-based autonomous landing systems executed by a quadcopter: (1) a vision- based algorithm is used to detect and predict an object’s future motion using Kalman Filter; (2) PID algorithm is implemented in a Quadcopter platform to autonomously balance and land this one on a stationary target.

Enhancement of Robustness in Object Detection Module for Advanced Driver Assistance Systems

An environmental condition detector that helps to send a warning signal to the cloud in case the self-driving car needs human actions due to the significance of the situation and the empirical results prove the reliability and the scalability of the proposed system to realistic scenarios.

SSD: Single Shot MultiBox Detector

The approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location, which makes SSD easy to train and straightforward to integrate into systems that require a detection component.

Object Detection Streaming and Data Management on Web Browser

  • 2020

Towards Improvement of Object Detection Accuracy for Deep Convolutional Neural Networks

  • Master's Thesis, Myongji University,
  • 2021

Object Detection Streaming and Data

  • Management on Web Browser”,
  • 2020