Skip to content
2664 images of drones, labeled, with trained YOLO weights.
Branch: master
Clone or download
Latest commit 4188486 Apr 29, 2019


DroneNet is Joseph Redmon's YOLO real-time object detection system retrained on 2664 images of DJI drones, labeled.
The original and labeled images used for retraining can be found under the image and label folders respectively.

Setting up

  1. Install the Ubuntu Linux distribution.

  2. Open terminal and enter the following lines to build Darknet:

git clone
cd darknet
  1. Move, drone.names, and yolo-drone.cfg under the cfg folder, create a weights directory and move yolo-drone.weights into the folder, move drone.jpg under the data folder, and move test.txt and train.txt in the root directory of your cloned darknet.

  2. Change lines 2 and 3 to your path in


Open terminal in the root directory of the darknet executable and enter:

./darknet detector test cfg/ cfg/yolo-drone.cfg weights/yolo-drone.weights data/drone.jpg


You can’t perform that action at this time.