Abstract
Traditional target detection algorithms have difficulty to adapt complex environmental changes and have limited applicable scenarios. However, the deep-learning-based target detection model can automatically learn with strong generalization capability. In this article, we choose a single-stage deep-learning-based target detection model for research based on the model’s real-time processing requirements and to improve the accuracy and the robustness of target detection in remote sensing images. In addition, we improve the YOLOv4 network and present a new approach. First, we propose a classification setting of the nonmaximum suppression threshold to increase the accuracy without affecting the speed. Second, we study the anchor frame allocation problem in YOLOv4 and propose two allocation schemes. The proposed anchor frame scheme also improves the detection performance, and experimental results on the DOTA dataset validate their effectiveness.
Highlights
A LARGE number of remote sensing images have been generating regularly, and due to the rapid development of satellite and imaging technology, the task of object detection has gained significant attention of researchers
The experimental surface sets the threshold in the classification [0.5, 0.5, 0.5, 0.55]; the threshold setting of planes, ships, and small vehicles is 0.5; for large vehicle, it is set to 0.55; the model performance is good with 77.68% mean average precision (mAP), which is an increase of 2.53% compared to the original model setting
We propose an approach that first divides all targets into three scale targets of large, medium, and small according to the principle of anchor frame distribution, and the K-means algorithm is used to cluster the targets in TABLE VIII ABLATION STUDY OF DATA AUGMENTATION
Summary
A LARGE number of remote sensing images have been generating regularly, and due to the rapid development of satellite and imaging technology, the task of object detection has gained significant attention of researchers. Target detection is the process of detecting instances of semantic objects of a certain class (such as humans, vehicles airplanes, or ships) in digital images and videos Analyzing such images contributes to social and economic aspects for decision-making as they provide a valuable source of information. Fang et al [13] combined SqueezeNet with YOLOv3-tiny and get a tinier network These techniques cannot be directly applied to optical remote sensing image target detection and identification. To the best of our knowledge, target vehicle detection for the remote sensing image has not been studied in previous works This naturally brings a significant need of intelligent earth observation through automated analysis and understanding of aerial or satellite images.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
