Ensemble Learning Approaches Based on Covariance Pooling of CNN Features for High Resolution Remote Sensing Scene Classification - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Remote Sensing Année : 2020

Ensemble Learning Approaches Based on Covariance Pooling of CNN Features for High Resolution Remote Sensing Scene Classification

Résumé

Remote sensing image scene classification, which consists of labeling remote sensing images with a set of categories based on their content, has received remarkable attention for many applications such as land use mapping. Standard approaches are based on the multi-layer representation of first-order convolutional neural network (CNN) features. However, second-order CNNs have recently been shown to outperform traditional first-order CNNs for many computer vision tasks. Hence, the aim of this paper is to show the use of second-order statistics of CNN features for remote sensing scene classification. This takes the form of covariance matrices computed locally or globally on the output of a CNN. However, these datapoints do not lie in an Euclidean space but a Riemannian manifold. To manipulate them, Euclidean tools are not adapted. Other metrics should be considered such as the log-Euclidean one. This consists of projecting the set of covariance matrices on a tangent space defined at a reference point. In this tangent plane, which is a vector space, conventional machine learning algorithms can be considered, such as the Fisher vector encoding or SVM classifier. Based on this log-Euclidean framework, we propose a novel transfer learning approach composed of two hybrid architectures based on covariance pooling of CNN features, the first is local and the second is global. They rely on the extraction of features from models pre-trained on the ImageNet dataset processed with some machine learning algorithms. The first hybrid architecture consists of an ensemble learning approach with the log-Euclidean Fisher vector encoding of region covariance matrices computed locally on the first layers of a CNN. The second one concerns an ensemble learning approach based on the covariance pooling of CNN features extracted globally from the deepest layers. These two ensemble learning approaches are then combined together based on the strategy of the most diverse ensembles. For validation and comparison purposes, the proposed approach is tested on various challenging remote sensing datasets. Experimental results exhibit a significant gain of approximately 2% in overall accuracy for the proposed approach compared to a similar state-of-the-art method based on covariance pooling of CNN features (on the UC Merced dataset).
Fichier principal
Vignette du fichier
Akodad20_RS.pdf (2.39 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-02967495 , version 1 (15-10-2020)

Identifiants

Citer

Sara Akodad, Lionel Bombrun, Junshi Xia, Yannick Berthoumieu, Christian Germain. Ensemble Learning Approaches Based on Covariance Pooling of CNN Features for High Resolution Remote Sensing Scene Classification. Remote Sensing, 2020, 12 (20), pp.3292. ⟨10.3390/rs12203292⟩. ⟨hal-02967495⟩
134 Consultations
84 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More