Skip to content

This repository demonstrates the work of the detector for detecting drones using the Yolo neural network version 8

License

Notifications You must be signed in to change notification settings

EVNN304/Yolo_V8_drone_detection

Repository files navigation

Yolo_V8_drone_detection

This repository demonstrates the work of the detector for detecting drones using the Yolo neural network version 8 link to the repository with the neural network https://github.com/ultralytics/ultralytics

The main operating principle of the program is that the coordinates of a moving object from the tracker are given, an area is cut out from the full-size image and transferred to a neural network for analysis

Accuracy matrix for map50 for 3 classes:

  1. drone 0.988
  2. bird 0.986
  3. plane 0.998

Accuracy matrix for map50-95

  1. drone 0.758
  2. bird 0.714
  3. plane 0.794

confusion_matrix_normalized results

val_batch2_labels val_batch0_labels val_batch1_labels val_batch1_pred

About

This repository demonstrates the work of the detector for detecting drones using the Yolo neural network version 8

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages