SiamMFF: UAV Object Tracking Algorithm Based on Multi-Scale Feature Fusion

• Published in: IEEE access Vol. 12; pp. 24725 - 24734
• Main Authors: Hou, Yanli, Gai, Xilin, Wang, Xintao, Zhang, Yongqiang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Autonomous aerial vehicles; Blurring; Convolution; Convolutional neural networks; Correlation; deformable convolution; Deformable models; Feature extraction; Formability; Kernel; multi-scale feature fusion; Object tracking; Search problems; Siamese network; Tracking; unmanned aerial vehicle(UAV)
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2024.3354381

UAVs have entered various fields of life, and object tracking is one of the key technologies for UAV applications. However, there are various challenges in practical applications, such as the scale change of video images, motion blur and too high shooting angle leading to the tracked objects being too small, resulting in poor tracking accuracy. To cope with the problem that small targets are poorly tracked by UAVs due to less effective information output from the deep residual network, a SiamMFF tracking method that introduces an efficient multi-scale feature fusion strategy is proposed. The method aggregates features at different scales, and at the same time, replaces the ordinary convolution with deformable convolution to increase the sense field of convolution operation to enhance the feature extraction capability. The experimental results show that the proposed algorithm improves the success rate and accuracy of small target tracking.