Convolution Neural Network Based on Two-Dimensional Spectrum for Hyperspectral Image Classification

Lin, Shuo; Yang, Yuwei; Li, Chenming; Yang, Mingxiang

doi:10.1155/2018/8602103

Cited by 27 publications

(21 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we further compared the proposed FDMFN method with another three state-of-the-art deep learning based HSI classification approaches: the dilated convolution based CNN model (Dilated-CNN) [28], the 2D spectrum based CNN model [29], and the artificial neuron network with center-loss and adaptive spatial-spectral center classifier (ANNC-ASSCC) [30].…”

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

“…Following Dilated-CNN [28], for each dataset, 60% of the labeled samples per class were randomly selected for training. Next, the proposed FDMFN was compared with the 2D spectrum based CNN model [29] on the IP, KSC, and Salinas datasets. For a fair comparison, we utilized the same number of samples, as in [29], for model training.…”

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

“…Next, the proposed FDMFN was compared with the 2D spectrum based CNN model [29] on the IP, KSC, and Salinas datasets. For a fair comparison, we utilized the same number of samples, as in [29], for model training. Finally, the proposed method was compared with the ANNC-ASSCC method [30] on the Salinas dataset.…”

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

“…In [28], Devaram et al proposed a dilated convolution based CNN model for HSI classification and applied an oversampling strategy to deal with the class imbalance problem. In [29], a 2D spectrum based CNN framework was introduced for pixelwise HSI classification, which converts the spectral vector into 2D spectrum image to exploit the spectral and spatial information. In [30], Guo et al proposed an artificial neural network (ANN) based spectral-spatial HSI classification framework, which combines the softmax loss and the center loss for network training.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fully Dense Multiscale Fusion Network for Hyperspectral Image Classification

et al. 2019

View full text Add to dashboard Cite

The convolutional neural network (CNN) can automatically extract hierarchical feature representations from raw data and has recently achieved great success in the classification of hyperspectral images (HSIs). However, most CNN based methods used in HSI classification neglect adequately utilizing the strong complementary yet correlated information from each convolutional layer and only employ the last convolutional layer features for classification. In this paper, we propose a novel fully dense multiscale fusion network (FDMFN) that takes full advantage of the hierarchical features from all the convolutional layers for HSI classification. In the proposed network, shortcut connections are introduced between any two layers in a feed-forward manner, enabling features learned by each layer to be accessed by all subsequent layers. This fully dense connectivity pattern achieves comprehensive feature reuse and enforces discriminative feature learning. In addition, various spectral-spatial features with multiple scales from all convolutional layers are fused to extract more discriminative features for HSI classification. Experimental results on three widely used hyperspectral scenes demonstrate that the proposed FDMFN can achieve better classification performance in comparison with several state-of-the-art approaches.

show abstract

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

Section: Comparison With Other State-of-the-art Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fully Dense Multiscale Fusion Network for Hyperspectral Image Classification

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The CNN structure consists of the input data as a matrix or tensor with a 3D spatial structure, convolution kernel, and output data. The input signal x is convoluted by filter f to calculate signal y as follows [31]:…”

Section: The Estimated Kalman Filter Modelmentioning

confidence: 99%

Human Motion Detection through Wall Based on Micro-Doppler Using Kalman Filter Combined with Convolutional Neural Network

Saeed¹,

Abbas²,

Abed³

et al. 2019

IJIES

View full text Add to dashboard Cite

Human motion detection, classification, and perceiving, the dynamics of moving objects in the environment, are crucial in many applications. Many sensors have been used to this detection; the radar represents one of the promising sensors. Kalman filter (KF) and convolutional neural network (CNN) represents a powerful tool for estimation and classification respectively. In this paper, a combination between the KF and CNN have been proposed to detect and classify human behavior. This proposal presents two important points, the precise features map from the combination of Kalman Filter and CNN, as well as the use of the radar, which is working under all circumstances and does not break the privacy. Twenty different experiments with three scenarios for different motion with and without glass wall have been studied, and they are classified. The results show that the overperform of the proposed algorithm and the classification accuracy can reach 98.7%. This advancement of the proposed algorithm depends on the efficient Wigner-Ville short time Fourier transform (STFT) which is used as a feature extractor and make Range-Doppler (RD) map.

show abstract