Litcius/Paper detail

Attention-Aware Pseudo-3-D Convolutional Neural Network for Hyperspectral Image Classification

Jianzhe Lin, Lichao Mou, Xiao Xiang Zhu, Xiangyang Ji, Z. Jane Wang

2021IEEE Transactions on Geoscience and Remote Sensing34 citationsDOI

Abstract

Convolutional neural networks (CNNs) have been applied for hyperspectral image classification recently. Among this class of deep models, 3-D CNN has been shown to be more effective by learning discriminative features from abundant spectral signatures and spatial contexts in hyperspectral imagery (HSI). However, by simply imposing 3-D CNN to HSI, a large amount of initial information might be lost in this CNN pipeline. The proposed attention-aware pseudo-3-D (AP3D) convolutional network for HSI classification is motivated by two observations. First, each dimension of the 3-D HSI is not equally important, different attention should be paid to different dimensions of the initial HSI image, especially in the first convolution operation. Second, intermediate representations of the 3-D input image at different stages in the 3-D CNN pipeline represent different levels of features and should not be neglected and abandoned. Instead, a 2-D matrix of scores for each feature map should be fed to the final softmax layer. Quantitative and qualitative results demonstrate that the proposed AP3D model outperforms the state-of-the-art HSI classification methods in agricultural and rural/urban data sets: Indian Pines, Pavia University, and Salinas Scene.

Topics & Concepts

Softmax functionHyperspectral imagingConvolutional neural networkArtificial intelligencePattern recognition (psychology)Computer scienceDiscriminative modelContextual image classificationFeature (linguistics)Convolution (computer science)Feature extractionDimensionality reductionImage (mathematics)Artificial neural networkPhilosophyLinguisticsRemote-Sensing Image ClassificationRemote Sensing and Land UseAdvanced Chemical Sensor Technologies
Attention-Aware Pseudo-3-D Convolutional Neural Network for Hyperspectral Image Classification | Litcius