- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Deep Tensor Factorization for Hyperspectral Image Classification
摘要： High-dimensional spectral feature and limited training samples have caused a range of difficulties for hyperspectral image (HSI) classification. Feature extraction is effective to tackle this problem. Specifically, tensor factorization is superior to some prominent methods such as principle component analysis (PCA) and non-negative matrix factorization (NMF) because it takes spatial information into consideration. Recently, deep learning has gotten more and more attention for efficiently extracting hierarchical features for various tasks. In this paper, we propose a novel feature extraction method, deep tensor factorization (DTF), to extract hierarchical and meaningful features from observed HSI. This method takes advantage of tensor in representing HSI and the merits of convolutional neural network (CNN) in hierarchical feature extraction. Specifically, a convolution operation is firstly applied in the spectral dimension of HSI to suppress the effect of noise. Then, the convolved HSI is fed into tensor factorization to learn a low rank representation of data. After that, the above two process are repeated to learn a hierarchical representation of HSI. Experimental results on two real hyperspectral datasets show the superiority of the proposed method.
关键词： Hyperspectral image (HSI) classification,feature extraction,convolutional neural network (CNN),tensor decomposition
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Similarity-Preserving Deep Features for Hyperspectral Image Classification
摘要： Recently, deep learning has been introduced to extract hierarchical features of hyperspectral images (HSIs) and achieved good classification performance. However, the previous deep learning based methods only consider the semantic information of individual pixel, which cannot effectively deal with the complex spectral-spatial characteristic of HSIs. In this paper, we propose a novel deep learning based framework to learn the similarity-preserving deep features (SPDF) for HSI classification. Specifically, we firstly introduce a deep network that can take pairs of image patches as training samples, and then a loss function is elaborately designed to minimize the feature distance of similar pairs and maximize the feature distance of dissimilar pairs in feature space. Once the deep network is well trained, the SPDF can be obtained by propagating the samples through the trained network. Finally, these features are fed into the support vector machines (SVM) for HSI classification. Experimental results demonstrate the proposed method outperforms other competitive methods.
关键词： similarity,feature extraction,deep learning,Hyperspectral image classification
更新于2025-09-10 09:29:36
-
Convolutional Neural Network Trained by Joint Loss for Hyperspectral Image Classification
摘要： In this letter, is proposed the hyperspectral image classification method based on the convolutional neural network, which is trained jointly by the reconstruction and discriminative loss functions. In the network, small convolutional kernels are cascaded with the pooling operator to perform feature abstraction, and a decoding channel composed of the deconvolutional and unpooling operators is established. The unsupervised reconstruction, performed by the decoding channel, not only introduces priors to the network training but also is made use to enhance the discriminability of the abstracted features by the control gate. By the experiments, it is shown that the proposed method performs better than the state-of-the-art neural network-based classification methods.
关键词： Control gate,unsupervised reconstruction,convolutional neural network (CNN),joint loss (JL),hyperspectral image (HSI) classification
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Dual-Channel Densenet for Hyperspectral Image Classification
摘要： Deep neural networks provide deep extracted features for image classification. As a high dimension data, hyperspectral image (HSI) feature extraction is unlike an RGB image whose feature representation could not be simply generated in the spatial domain. To take full advantage of HSI, a dual-channel convolutional neural network (CNN) is applied, 1D convolution for the spectral domain and 2D convolution for spatial domain. For pixel-wise classification of HSI, in our network model, one-dimensional customized DenseNet is for extracting the hierarchical spectral features and another customized DenseNet is applied to extract the hierarchical spatial-related feature. Furthermore, we experimentally tuned the several widen factors and dense-net growth rates to evaluate the impact of hyper-parameter. To compare our proposed method with HSI classification methods, we test other three DNNs based method in two real-world HSI dataset. The result demonstrated our approach outperformed the state-of-art method.
关键词： Dual-channel DenseNet,Hyperspectral image classification,spatial-spectral
更新于2025-09-10 09:29:36
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Wide Contextual Residual Network with Active Learning for Remote Sensing Image Classification
摘要： In this paper, we propose a wide contextual residual network (WCRN) with active learning (AL) for remote sensing image (RSI) classification. Although ResNets have achieved great success in various applications (e.g. RSI classification), its performance is limited by the requirement of abundant labeled samples. As it is very difficult and expensive to obtain class labels in real world, we integrate the proposed WCRN with AL to improve its generalization by using the most informative training samples. Specifically, we first design a wide contextual residual network for RSI classification. We then integrate it with AL to achieve good machine generalization with limited number of training sampling. Experimental results on the University of Pavia and Flevoland datasets demonstrate that the proposed WCRN with AL can significantly reduce the needs of samples.
关键词： SAR,Residual networks,classification,hyperspectral image,remote sensing,active learning
更新于2025-09-10 09:29:36
-
[IEEE 2018 37th Chinese Control Conference (CCC) - Wuhan (2018.7.25-2018.7.27)] 2018 37th Chinese Control Conference (CCC) - Deep Forest-Based Classification of Hyperspectral Images
摘要： The classi?cation of hyperspectral images (HSIs) is a hot topic in the ?eld of remote sensing technology. In recent years, convolutional neural network (CNN) has achieved great success for HSI classi?cation. However, CNN has to do a great effort in parameters tuning which is time-consuming. Furthermore, a large number of samples are required to train CNN, nevertheless, it is expensive to obtain enough training samples from HSIs. In this paper, we propose a novel classi?cation approach based on deep forest. To reduce the dimension of hyperspectral data, principal component analysis (PCA) is performed during the pre-processing. In contrast to the CNN, our method has fewer hyper-parameters and faster training speed. To the best of our knowledge, this is among the ?rst deep forest-based hyperspectral spectral information classi?cation. Extensive experiments are conducted on two real-world HSI datasets to show the proposed method is signi?cantly superior to the state-of-the-art methods.
关键词： Deep Neural Network(DNN),Hyperspectral Image (HSI),Principal Component Analysis (PCA),Deep Forest
更新于2025-09-10 09:29:36
-
Hyperspectral Target Detection via Adaptive Information—Theoretic Metric Learning with Local Constraints
摘要： By using the high spectral resolution, hyperspectral images (HSIs) provide significant information for target detection, which is of great interest in HSI processing. However, most classical target detection methods may only perform well based on certain assumptions. Simultaneously, using limited numbers of target samples and preserving the discriminative information is also a challenging problem in hyperspectral target detection. To overcome these shortcomings, this paper proposes a novel adaptive information-theoretic metric learning with local constraints (ITML-ALC) for hyperspectral target detection. The proposed method firstly uses the information-theoretic metric learning (ITML) method as the objective function for learning a Mahalanobis distance to separate similar and dissimilar point-pairs without certain assumptions, needing fewer adjusted parameters. Then, adaptively local constraints are applied to shrink the distances between samples of similar pairs and expand the distances between samples of dissimilar pairs. Finally, target detection decision can be made by considering both the threshold and the changes between the distances before and after metric learning. Experimental results demonstrate that the proposed method can obviously separate target samples from background ones and outperform both the state-of-the-art target detection algorithms and the other classical metric learning methods.
关键词： target detection,hyperspectral image,local constraints,metric learning
更新于2025-09-10 09:29:36
-
Fast Active Learning for Hyperspectral Image Classification using Extreme Learning Machine
摘要： Due to undulating and complexity of the earth’s surface, obtaining the training samples for remote sensing data is time consuming and expensive. Therefore, it is highly desirable to design a model that uses as few labelled samples as possible and reducing the computational time. Several active learning (AL) algorithms have been proposed in the literature for the classification of hyperspectral images (HSI).However, its performance in term of computational time has not been focused yet. In this paper, we have proposed AL approach based on Extreme Learning Machine (ELM) that effectively decreases the computational time while maintaining the classification accuracy. Further, the effectiveness of the proposed approach has been depicted by comparing its performance with state-of-the-art AL algorithms in terms of classification accuracy and computational time as well. The ELM based active learning (ELM-AL) with different query strategies were conducted on two HSI data sets. The proposed approach achieves the classification accuracy up to 90% which is comparable to support vector machine (SVM) based AL (SVM-AL) approach but effectively reduces the computational time significantly by 1000 times. Thus proposed system shows the encouraging results with adequate classification accuracy while reducing the computation time drastically.
关键词： Uncertainty sampling,Remote Sensing Image,Extreme learning machine,Classification,Active learning,Uncertainty measure,Hyperspectral Image
更新于2025-09-09 09:28:46
-
[Smart Innovation, Systems and Technologies] Smart Intelligent Computing and Applications Volume 105 (Proceedings of the Second International Conference on SCI 2018, Volume 2) || Dimensionality Reduction Using Subset Selection Method in Framework for Hyperspectral Image Segmentation
摘要： This paper presents a dimensionality reduction method using subset selection for hyperspectral image segmentation framework. This framework consists of three stages—dimensionality reduction, hierarchical image fusion, and segmentation. A methodology based on subset construction is used for selecting k informative bands from d bands dataset. In this selection, similarity metrics such as Average Pixel Intensity (API), Histogram Similarity (HS), Mutual Information (MI) and Correlation Similarity (CS) are used to create k distinct subsets and from each subset, a single band is selected. Hierarchical fusion is used to create a single high quality image. After getting fused image, Fuzzy c-means (FCM) algorithm is used for segmentation of image. The qualitative and quantitative analysis shows that CS similarity metric in dimensionality reduction algorithm gets high-quality segmented image.
关键词： subset selection,hierarchical image fusion,hyperspectral image segmentation,dimensionality reduction,Fuzzy c-means (FCM)
更新于2025-09-09 09:28:46
-
[American Society of Agricultural and Biological Engineers 2018 Detroit, Michigan July 29 - August 1, 2018 - ()] 2018 Detroit, Michigan July 29 - August 1, 2018 - Classification of Tomato Impact Bruise Using Hyperspectral Imaging Based on Spatial-spectral Method
摘要： Tomatoes have various drop impacts on post-harvest process, which causes the quality deterioration. It is required to evaluate impact injuries quickly in a non-destructive method. Hyperspectral image is commonly of with multi-modal classes and ambiguous class boundary, and spatially adaptive classification of land cover with hyperspectral image is one of challenging problems in accurate classification image community. As hyperspectral image includes many interesting objects whereas each object contains variant spectral signature and the discrimination among them is less efficient. This paper presents a new spatial-spectral fusion method, which extracts patch analysis and combines spectral features to perform fruit quality classification. In spectral features, a method of tomato quality classification based on mean-square-error curve fitting and peak-feature matching is presented. It extracts peak features from known drop injury tomatoes’ spectra and unknown tomato samples spectra to compute their similarity values through multiple similarity measures, respectively. Then, the unknown sample is assigned by selecting the known quality tomato with the largest similarity value. At last, in comparison with the proposed method and the method such as partial-least square discriminate analysis (PLS-DA), support vector machine (SVM), the result shows the practicality and accuracy of the proposed method.
关键词： Tomato,Quality Classification,Peak-feature matching,Patch Analysis,Hyperspectral Image
更新于2025-09-09 09:28:46