UAV Target Detection Algorithm with Improved YOLOv7

Fanrun Meng; Chen Liu; Zhiren Zhu; Liming Zhou

doi:10.54097/fcis.v5i2.12803

Authors

Fanrun Meng
Chen Liu
Zhiren Zhu
Liming Zhou

DOI:

https://doi.org/10.54097/fcis.v5i2.12803

Keywords:

Unmanned Aerial Vehicle (UAV), YOLOv7, BiFPN, GAM

Abstract

The wide application of UAV technology in various fields makes UAV target detection crucial. In this study, we propose an improved algorithm based on YOLOv7 to enhance the performance and robustness of UAV target detection. We utilize YOLOv7 as the infrastructure and introduce BiFPN (Bi-directional Feature Pyramid Network) to enhance the feature fusion, while adding the GAM attention mechanism to the model, which is trained and evaluated using the VisDrone2019 dataset. The experimental results of this study show that the improved model achieves an average accuracy mAP value of 45.6%, which is 2.7% higher than the traditional model, and is able to detect and localize UAV targets more accurately.

Downloads

Download data is not yet available.

References

A, Huiyu Zhou , Y. Y. B , and C. S. C . "Object tracking using SIFT features and mean shift." Computer Vision and Image Understanding 113. 3(2009):345-352.

Yang Jinkun, et al."HOG and SVM algorithm based on vehicle model recognition." MIPPR 2019: PATTERN RECOGNITION AND COMPUTER VISION 11430.(2020).

Joachims, Thorsten . "Making Large-Scale SVM Learning Practical." Technical Reports 8.3(1998):499-526.

Viola, Paul , and M. J. Jones . "Fast and Robust Classification using Asymmetric AdaBoost and a Detector Cascade." NIPS 2001.

Chua, L. O. , and T. Roska . "The CNN paradigm." Circuits & Systems I Fundamental Theory & Applications IEEE Transactions on 40.3(1993):147-156.

R. Girshick, "Fast r-cnn," Proceedings of the IEEE international conference on computer vision, vol. 12, pp. 1440-1448, 2015.

S. Ren, K. He, Girshick R, and J. Sun, "Faster r-cnn: Towards real-time object detection with region proposal networks," Advances in neural information processing systems, vol. 28, 2015.

K. He, G. Gkioxari, and P. Dollár, R. Girshick, "Mask r-cnn," Proceedings of the IEEE international conference on computer vision. vol. 10, pp. 2961-2969, 2017.

Redmon, Joseph , and A. Farhadi. "YOLOv3: An Incremental Improvement." arXiv e-prints (2018).

Sun, Y. X. , et al. "A CLASSIFICATION AND LOCATION OF SURFACE DEFECTS METHOD IN HOT ROLLED STEEL STRIPS BASED ON YOLOV7." Metalurgija (2023).

Zhong, Lehai , et al. "Integration Between Cascade Region-Based Convolutional Neural Network and Bi-Directional Feature Pyramid Network for Live Object Tracking and Detection." Traitement du Signal: signal image parole (2021).

Treisman, Anne M. , and G. Gelade . "A feature-integration theory of attention. " Cognitive Psychology 12. 1(1980):97-136.