An Attention-Guided Multilayer Feature Aggregation Network for Remote Sensing Image Scene Classification

Abstract

Remote sensing image scene classification (RSISC) has broad application prospects, but related challenges still exist and urgently need to be addressed. One of the most important challenges is how to learn a strong discriminative scene representation. Recently, convolutional neural networks (CNNs) have shown great potential in RSISC due to their powerful feature learning ability; however, their performance may be restricted by the complexity of remote sensing images, such as spatial layout, varying scales, complex backgrounds, category diversity, etc. In this paper, we propose an attention-guided multilayer feature aggregation network (AGMFA-Net) that attempts to improve the scene classification performance by effectively aggregating features from different layers. Specifically, to reduce the discrepancies between different layers, we employed the channel–spatial attention on multiple high-level convolutional feature maps to capture more accurately semantic regions that correspond to the content of the given scene. Then, we utilized the learned semantic regions as guidance to aggregate the valuable information from multilayer convolutional features, so as to achieve stronger scene features for classification. Experimental results on three remote sensing scene datasets indicated that our approach achieved competitive classification performance in comparison to the baselines and other state-of-the-art methods.

Target feature matching method of remote sensing image of power tower supported by grid coding

“…In formula (1), x 、 y represents the coordinates of the image center point, m 、 n represents the neighborhood coordinates, represents the gray difference threshold. At this time, the gray difference threshold [11]of the image can be processed according to the above formula to suppress image noise.…”

Oriented Object Detection in Remote Sensing Images with Anchor-Free Oriented Region Proposal Network

“…ARCnet [44] utilizes a novel recurrent attention structure to force the scene classifiers to learn to focus on some critical areas of the very high-resolution RSIs, which often contain complex objects. AGMFA-Net [45] uses an attention-guided multi-layer feature aggregation network to capture more complete semantic regions for more powerful scene representation.…”

Remote Sensing Image Scene Classification by Multiple Granularity Semantic Learning

“…R EMOTE sensing image scene classification [1]- [4] has been widely used in fields [5] such as land surveying, nature monitoring, and urban planning [6]. It has made great progress [7], [8] with the development of deep learning [9], [10] and automatic machine learning [11], such as neural architecture search (NAS) technology [12].…”

“…The rotation angle of a scene image is an integer multiple of 23. The proportion of center crop order is 1 2 and 2 3 . After merging augmented and original samples, the numbers of samples tend to be balanced across categories.…”