Abstract
Scene classification of high-resolution remote sensing (RS) images has attracted increasing attentions due to its vital role in a wide range of applications. Convolutional neural networks (CNNs) have recently been applied on many computer vision tasks and have significantly boosted the performance including imagery scene classification, object detection, and so on. However, the classification performance heavily relies on the features that can accurately represent the scene of images, thus, how to fully explore the feature learning ability of CNNs is of crucial importance for scene classification. Another problem in CNNs is that it requires a large number of labeled samples, which is impractical in RS image processing. To address these problems, a novel sparse representation-based framework for small-sample-size RS scene classification with deep feature fusion is proposed. Specially, multilevel features are first extracted from different layers of CNNs to fully exploit the feature learning ability of CNNs. Note that the existing well-trained CNNs, e.g., AlexNet, VGGNet, and ResNet50, are used for feature extraction, in which no labeled samples is required. Then, sparse representation-based classification is designed to fuse the multilevel features, which is especially effective when only a small number of training samples are available. Experimental results over two benchmark datasets, e.g., UC-Merced and WHU-RS19, demonstrated that the proposed method can effectively fuse different levels of features learned in CNNs, and clearly outperform several state-of-the-art methods especially with limited training samples.
Highlights
S CENE classification of remote sensing (RS) images has received increasing attentions
In order to combine the advantages of Convolutional neural networks (CNNs) and SRC, we proposed a novel sparse representation-based framework with deep feature fusion strategy
2) Feature Fusion and SRC: A sparse representation model is used to fuse these multilevel features for RS scene classification, which balance the contribution of features from different layers of CNNs
Summary
S CENE classification of remote sensing (RS) images has received increasing attentions. Convolutional neural network (CNN) based methods have greatly improved the performance of image classification and object detection, such as classical AlexNet [24], VGGNet [25], Inception Net [26], and ResNet [27]. These CNN-based frameworks can automatically learn to extract high-level discriminative features, which have been widely used. 3) The proposed method addresses the few-shot classification problem of RS images, since multilevel features are extracted from the well-trained CNNs. As a result, competitive results are obtained through the SRC framework based on multilayer framework.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have