loading page

SafeSpace MFNet: Precise and Efficient MultiFeature Drone Detection Network
  • +3
  • mahnoor dil ,
  • misha urooj khan ,
  • Muhammad Zeshan Alam ,
  • Farooq Alam Orakzai ,
  • Zeeshan Kaleem ,
  • Chau Yuen
mahnoor dil
Author Profile
misha urooj khan
Author Profile
Muhammad Zeshan Alam
Author Profile
Farooq Alam Orakzai
Author Profile
Zeeshan Kaleem
COMSATS University Islamabad

Corresponding Author:[email protected]

Author Profile
Chau Yuen
Author Profile


Unmanned air vehicles (UAVs) popularity is on the rise as it enables the services like traffic monitoring, emergency communications, deliveries, and surveillance. However, the unauthorized usage of UAVs (a.k.a drone) may violate security and privacy protocols for security-sensitive national and international institutions. The presented challenges require fast, efficient, and precise detection of UAVs irrespective of harsh weather conditions, the presence of different objects, and their size to enable SafeSpace. Recently, there has been significant progress in using the latest deep learning models, but those models have shortcomings in terms of computational complexity, precision, and non-scalability. To overcome these limitations, we propose a precise and efficient multiscale and multifeature UAV detection network for SafeSpace, i.e., \textit{MultiFeatureNet} (\textit{MFNet}), an improved version of the popular object detection algorithm YOLOv5s. In \textit{MFNet}, we perform multiple changes in the backbone and neck of the YOLOv5s network to focus on the various small and ignored features required for accurate and fast UAV detection. To further improve the accuracy and focus on the specific situation and multiscale UAVs, we classify the \textit{MFNet} into small (S), medium (M), and large (L): these are the combinations of various size filters in the convolution and the bottleneckCSP layers, reside in the backbone and neck of the architecture. This classification helps to overcome the computational cost by training the model on a specific feature map rather than all the features. The results show significant performance gain even for unseen feature maps with minimal loss in accuracy. Results show a significant reduction in training parameters, inference, and increased pattern in FPS and GFLOPs for \textit{MFNet} compared to YOLOv5s. \textit{MFNet-M} performance evaluation in terms of precision, recall, mean average-precision (mAP), and IOU increased around 1.8\%, 2.2\%, 0.9\%, 1.7\% compared to YOLOv5s. Furthermore, \textit{MFNet-M} achieves the best performance with 96.8\% precision, 88.4\% recall, 95.9\% mAP, and 51.1\% IoU for UAV detection. The dataset and code are available as an open source: github.com/ZeeshanKaleem/MultiFeatureNet.