Nonlinear Feature Normalization for Hyperspectral Domain Adaptation and Mitigation of Nonlinear Effects

Gross, Wolfgang L.; Tuia, Devis; Soergel, Uwe; Middelmann, Wolfgang

doi:10.1109/tgrs.2019.2903719

Cited by 18 publications

(22 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is considered later when evaluating the classification results. In addition to the Cubert dataset, three airborne hyperspectral benchmark datasets, namely Greding, Indian Pines, and Pavia University, are used to evaluate the CNN's ability of transfer learning (Gross et al, 2019;Baumgardner et al, 2015;Plaza et al, 2006. Table 2 shows details on the all used datasets.…”

Section: Number Of Samplesmentioning

confidence: 99%

“…The offset is removed by shifting all bands by their relative offset to match the first band. (Gross et al, 2019;Baumgardner et al, 2015;Plaza et al, 2006).…”

Section: Data Pre-processingmentioning

confidence: 99%

“…The Cubert data is then processed using a min-maxnormalization (Jayalakshmi and Santhakumaran, 2011). The benchmark datasets are already pre-processed (Gross et al, 2019;Baumgardner et al, 2015;Plaza et al 2006). Hence, it is not necessary to conduct other pre-processing procedures than training / validation / test data splitting and the calculation of a normalized dataset with the same technique as used for the Cubert data.…”

Section: Cubertmentioning

confidence: 99%

“…For the Greding dataset, no study was conducted yet that used a CNN. A study accomplished by Gross et al (2019), however, achieved lower accuracies (difference around 1 to 3%) using a spectral angle mapper (SAM) and support vector machines (SVM).…”

Section: Benchmark Datamentioning

confidence: 99%

See 3 more Smart Citations

Strategic Optimization of Convolutional Neural Networks for Hyperspectral Land Cover Classification

Buehler

Schenkel

Gross

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Hyperspectral data recorded by future earth observation satellites will have up to hundreds of narrow bands that cover a wide range of the electromagnetic spectrum. The spatial resolution (around 30 meters) of such data, however, can impede the integration of the spatial domain for a classification due to spectrally mixed pixels and blurred edges in the data. Hence, the ability of performing a meaningful classification only relying on spectral information is important. In this study, a model for the spectral classification of hyperspectral data is derived by strategically optimizing a convolutional neural network (1D-CNN). The model is pre-trained and optimized on imagery of different nuts, beans, peas and dried fruits recorded with the Cubert ButterflEye X2 sensor. Subsequently, airborne hyperspectral datasets (Greding, Indian Pines and Pavia University) are used to evaluate the CNN's capability of transfer learning. For that, the datasets are classified with the pre-trained weights and, for comparison, with the same model architecture but trained from scratch with random weights. The results show substantial differences in classification accuracies (from 71.8% to 99.8% overall accuracy) throughout the used datasets, mainly caused by variations in the number of training samples, the spectral separability of the classes as well as the existence of mixed pixels for one dataset. For the dataset that is classified least accurately, the greatest improvement with pre-training is achieved (difference of 3.3% in overall accuracy compared to the non-pre-trained model). For the dataset that is classified with the highest accuracy, no significant transfer learning was observed.

show abstract

Section: Number Of Samplesmentioning

confidence: 99%

“…The offset is removed by shifting all bands by their relative offset to match the first band. (Gross et al, 2019;Baumgardner et al, 2015;Plaza et al, 2006).…”

Section: Data Pre-processingmentioning

confidence: 99%

Section: Cubertmentioning

confidence: 99%

Section: Benchmark Datamentioning

confidence: 99%

See 2 more Smart Citations

Strategic Optimization of Convolutional Neural Networks for Hyperspectral Land Cover Classification

Buehler

Schenkel

Gross

et al. 2020

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…Liu et al [23] proposed a class-wise adversarial networks and achieved a superior feature-alignment performance. Gross et al [24] proposed a nonlinear feature normalization alignment approach for domain adaptation of multitemporal hyperspectral images, which is able to mitigate nonlinear effects in hyperspectral data and transfer spectral features from one domain to another. Saha et al [25] exploited a cycle-consistent generative adversarial network to learn deep transcoding between multisensory and multitemporal domains.…”

mentioning

confidence: 99%

Joint Correlation Alignment-Based Graph Neural Network for Domain Adaptation of Multitemporal Hyperspectral Remote Sensing Images

Wang

Chen

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

In this paper, we propose a novel deep domain adaptation method based on graph neural network (GNN) for multitemporal hyperspectral remote sensing images. In GNN, graphs are constructed for source and target data, respectively. Then the graphs are utilized in each hidden layer to obtain features. GNN operates on graph structure and the relations between data samples can be exploited. It aggregates features and propagate information through graph nodes. Thus, the extracted features have an improved smoothness in each spectral neighborhood which is beneficial to classification. Furthermore, the domain-wise correlation alignment and classwise correlation alignment are jointly embedded in GNN network to achieve a joint distribution adaptation performance. By introducing the joint correlation alignment strategy in GNN, the extracted features can not only be aligned between domains but also have a superior discriminability in each domain. This domain adaptation network is named as joint correlation alignment based graph neural network (JCGNN). Experiments using multitemporal Hyperion and NCALM datasets demonstrate the effectiveness of the proposed method. Index Terms-Hyperspectral remote sensing, graph neural network, domain adaptation, classification Recently, deep learning has been applied to domain adaptation because of its excellent feature representation ability. Compared with shallow domain adaptation approaches, end-toend deep domain adaptation approaches transfer more effective knowledge by embedding adaptation modules in the network architecture. Maximum Mean Discrepancy (MMD) is a popular distribution alignment strategy, which is utilized in many deep domain adaptation methods. Tzeng et al. [17] proposed the deep domain confusion method, which firstly regularizes the single adaptation layer of deep neural network using linear-kernel MMD. Similar to MMD, correlation alignment could also measure the distribution distance between domains, and it attempts to align the second-order statistics of the source and target features. Sun et al. [18] extended the correlation alignment to deep architectures and proposed the correlation alignment for deep domain adaptation (D-CORAL) approach, so that a nonlinear transformation that aligns the correlations of features between the two domains is obtained. Besides,

show abstract

Bibliography

2021

Deep Learning for the Earth Sciences

View full text Add to dashboard Cite

Nonlinear Feature Normalization for Hyperspectral Domain Adaptation and Mitigation of Nonlinear Effects

Cited by 18 publications

References 50 publications

Strategic Optimization of Convolutional Neural Networks for Hyperspectral Land Cover Classification

Strategic Optimization of Convolutional Neural Networks for Hyperspectral Land Cover Classification

Joint Correlation Alignment-Based Graph Neural Network for Domain Adaptation of Multitemporal Hyperspectral Remote Sensing Images

Bibliography

Contact Info

Product

Resources

About