Abstract
An important means of remote sensing imagery interpretation, remote sensing scene classification technology has recently achieved great success, especially based on deep learning. However, most of these methods are designed for noise-free images. The scene classification performance for noisy remote sensing images (RS), i.e., SAR images with speckle noise, is poor due to the sufficient effect of noise. An intuitive solution is denoising firstly and then classifying the image, which makes the whole process cumbersome. To address this problem, we design a new complete frequency channel attention network (CFCANet) that can handle noisy RS images directly without any filtering operation. CFCANet selects part of the low-frequency information to interact with the feature map adequately. For the original feature map, a corresponding 2D DCT frequency component is assigned, from which the most significant eigenvalue of each channel is obtained by maximization. Compared with the frequency channel attention network (FcaNet), the proposed network has better anti-noise ability as it exploits low frequency information of the images. The effectiveness of our method has been proved by experiments based on public datasets and some simulated datasets. Moreover, we build a new SAR scene classification dataset: WHU-SAR6. The comprehensive evaluation shows that the proposed method consistently outperforms several advanced methods, including ResNet, SENet, CBAM, EcaNet, and FcaNet.
Highlights
R EMOTE sensing is an important tool for earth observation, which helps us explore Earth’s surface and obtain rich and valuable information [1]–[3]
When the noise level is 3, our method achieves 90.23% and 94.24% for the 20% and 50% training ratio, which is 2.05% and 1.64% higher than ResNet; 0.97% and 1.18% higher than frequency channel attention network (FcaNet)
5) Experiment Results of WHU-SAR6 dataset: As shown in Table VI, we tested different attention modules on the WHU-SAR6 dataset, complete frequency channel attention network (CFCANet) obtained the best performance with 97.65% for the 10% training ratios
Summary
R EMOTE sensing is an important tool for earth observation, which helps us explore Earth’s surface and obtain rich and valuable information [1]–[3]. With the rapid development of Earth observation technology, an increasing number of different airborne or satellite images (e.g., multi/hyperspectral [4], synthetic aperture radar [5], etc.) with resolutions up to submeters can be conveniently obtained. One of the common processing chains of remote sensing image interpretation consists of 1) pre-processing, e.g., superresolution [11], denoising [12], radiometric calibration [13], 2) feature extraction, e.g., spectral unmixing [14], radiation characteristics [15], 3) automatic machine interpretation, e.g., classification [16], [17], change detection [18], recognition [19], and 4) post-processing. We focus on the scene classification tasks of synthetic aperture radar 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.
