Aircraft detection in remote-sensing images is a fundamental task in civil and military applications. Deep learning techniques to achieve end-to-end object detection have attracted the attention of the Earth observation community. One of the primary factors behind the success of deep learning techniques is the utilized data. Several previous studies focused on designing the network infrastructure. Instead, this study pays more attention to the data. With the increasing number of available public datasets, whether directly employing a large number of instances with great variation will lead to a good performance has become a research topic. The ways in which these object instances are collected differ greatly. For example, the image sizes, object sizes in the training images, and geospatial resolution are varied. Therefore, herein, the factors influencing the detection performance, such as the object size, ground sampling distance, and Google zoom view, are investigated. A you-only-look-once-v3-based detection process is proposed for automatic aircraft detection. A nonmaximum suppression algorithm strategy is applied to filter unreliable and redundant bounding boxes detected in the overlapping image blocks. The model generalization ability under different training data combinations is evaluated in several challenging cases. The results prove that more variety of training instances from a greater variety of zoom levels will result in more false alarms. Instead, more variety in the object sizes under a constant zoom level is welcome. A large range of aircraft sizes (i.e., 7 to 77 m in length in this study) can be detected, with a promising F1 score of 0.98.
With the rapid development of remote sensing, multiple techniques are now capable of producing digital elevation models (DEMs), such as photogrammetry, Light Detection and Ranging (LiDAR), and interferometric synthetic aperture radar (InSAR). Satellite-derived InSAR DEMs are particularly attractive due to their advantages of large spatial extents, cost-effectiveness, and less dependence on the weather. However, several complex factors may limit the quality of derived DEMs, e.g., the inherited errors may be nonlinear and spatially variable over an entire InSAR pair scene. We propose a segmentation-based coregistration approach for generating accurate InSAR DEMs over large areas. Two matching algorithms, including least squares matching and iterative closest point, are integrated in this approach. Three Advanced Land Observing Satellite Phased Array type L-band Synthetic Aperture Radar (ALOS PALSAR) InSAR DEMs are evaluated, and their root mean square errors (RMSEs) improved from 17.87 to 9.98 m, 51.94 to 15.80 m, and 27.12 to 12.26 m. Compared to applying a single global matching strategy, the segmentation-based strategy further improved the RMSEs of the three DEMs by 3.27, 13.01, and 9.70 m, respectively. The results clearly demonstrate that the segmentation-based coregistration approach is capable of improving the geodetic quality of InSAR DEMs.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.