Drone detection experiment based on image processing and machine learning

Drones are widely used in the field of information gathering and tracking, even they could be used to attacked targets. Therefore, the drone detection for the restricted areas or special zones is important and necessary. This paper focuses on the drone detection problem based on image processing for the restricted areas or special zones where used cameras for monitoring. The proposed solution detects drones from the captured images based on training the Haar-like features. The dataset of drone images is used in the Haar training process to generate a Haar-cascade model of drones. This model is then used to detect drones from images captured by the camera. The proposed solution is implemented and experimented with single cameras installed for any place including indoor environment and outdoor environment. Experimental results proved that the proposed solution could exactly detect drones for any zone or the restricted areas. The average accuracy of the proposed solution in the experimented environments is 91.9 %, and it provides an easy and economical solution for user.

Pham Giao N., Nguyen Phong H.

Publication type:
A1 Journal article – refereed

Place of publication:

Drone applications, Drone Security, Haar features, Haar training, Object detection


Full citation:
Pham, G. N., Nguyen, P. H., Drone detection experiment based on image processing and machine learning, International Journal of Scientific and Technological Research, ISSN: 2277-8616, Vol. 9:2, p. 2965-2971, http://www.ijstr.org/research-paper-publishing.php?month=feb2020


Read the publication here: