- 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 - Global Spatial and Local Spectral Similarity-Based Group Sparse Representation for Hyperspectral Imagery Classification
摘要： Spectral-spatial classification has been widely exploited for hyperspectral imagery. However, current methods either focus on local spatial similarity or global nonlocal self-similarity (NLSS). In this paper, we propose novel methods to couple both global spatial similarity and local spectral similarity together in a single framework. In particular, our approaches exploit global spatial similarity by searching non-overlap nonlocal patches, whereas spectral similarity is determined locally within the found patches. Experimental results on two real hyperspectral data sets demonstrate the efficiency of the proposed methods, with 5%-7% (overall classification accuracy) improvements over approaches that only consider either global or local similarity.
关键词： classification,nonlocal self-similarity,Hyperspectral image,group sparse representation
更新于2025-09-09 09:28:46
-
[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 - Spectral-Spatial Hyperspectral Image Classification via Locality and Structure Constrained Low-Rank Representation
摘要： Low-rank representation (LRR) has been applied widely in most fields due to its considerable ability to explore the low-dimensional subspace embedding in high-dimensional data. However, there are still some problems that LRR can’t effectively exploit the local structure and the representation for the given data is not discriminative enough. To tackle the above issues, we propose a novel locality and structure constrained low-rank representation (LSLRR) for hyperspectral image (HSI) classification. First, a distance metrics, which combines spectral and spatial similarity, is proposed to constrain the local structure. This makes two pixels in HSI with small distance have high similarity. Second, we exploit the classwise block-diagonal structure for the training data to learn the more discriminative representation for the testing data. And the experimental results verify the effectiveness and superiority of LSLRR comparing with other state-of-the-art methods.
关键词： low-rank representation,block-diagonal structure,hyperspectral image classification
更新于2025-09-09 09:28:46
-
Low-High-Power Consumption Architectures for Deep-Learning Models Applied to Hyperspectral Image Classification
摘要： Convolutional neural networks have emerged as an excellent tool for remotely sensed hyperspectral image (HSI) classification. Nonetheless, the high computational complexity and energy requirements of these models typically limit their application in on-board remote sensing scenarios. In this context, low-power consumption architectures are promising platforms that may provide acceptable on-board computing capabilities to achieve satisfactory classification results with reduced energy demand. For instance, the new NVIDIA Jetson Tegra TX2 device is an efficient solution for on-board processing applications using deep-learning (DL) approaches. So far, very few efforts have been devoted to exploiting this or other similar computing platforms in on-board remote sensing procedures. This letter explores the use of low-power consumption architectures and DL algorithms for HSI classification. The conducted experimental study reveals that the NVIDIA Jetson Tegra TX2 device offers a good choice in terms of performance, cost, and energy consumption for on-board HSI classification tasks.
关键词： hyperspectral image (HSI) classification,Deep learning (DL),low-power consumption architectures,embedded computing
更新于2025-09-09 09:28:46
-
A Low-rank Tensor Model for Hyperspectral Image Sparse Noise Removal
摘要： Hyperspectral image (HSI) has been widely used in target detection and classification. However, various kinds of noise in HSIs affect the applications of HSIs. In this paper, we propose a low-rank (LR) tensor recovery model to remove noise. Considering that the HSI is a 3-D HSI data, and the underlying LR tensor property is used in the model. Then, according to the similarity of adjacent bands images, the regularization on the difference of adjacent bands images is considered. The experiments of removing noise from different noisy HSIs show that our method can achieve better performance on removing sparse noise, especially for strips removal.
关键词： low-rank,Hyperspectral image,tensor,sparse noise removal
更新于2025-09-09 09:28:46
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Hyperspectral Image Refined Plant Classification By Graph Based Composite Kernel
摘要： Recently, the popularity of using hyperspectral image to study and monitor plant characteristics and conditions has been increased. The use of hyperspectral image improves the breeding process and increases profits. In the case of hyper-spectral data with high spectral resolution characteristics suitable for intraclass classification, this paper focuses on the application of hyperspectral image analysis in distin-guishing among different plant species. Plant intraclass clas-sification is sophisticated due to its small spectral differ-ences. Hence, a refined hyperspectral image classification method for plant, referred as SI-GCK which uses Spectral Index (SI) to represent plant spectral, and take advantage of semi-supervised graph-based composite kernel (GCK) method to combine spectral information and spatial location of pixels for classification is presented in this paper. As a comparison, sequential floating forward selection (SFFS) is used to select spectral bands for SVM learning. Its accuracy of plant classification is nearly equal to result by means of SI, and the proposed method in this paper is better than afore-mentioned.
关键词： spectral index,plant classification,graph-based composite kernel,hyperspectral image
更新于2025-09-04 15:30:14
-
Trilateral Smooth Filtering for Hyperspectral Image Feature Extraction
摘要： Traditional bilateral ?ltering (BF) cannot extract hyperspectral image (HSI) features well when the center pixel of the neighborhood pixel set is a noise point in the process of ?ltering the HSI. In this letter, a trilateral smooth ?ltering (TRSF) is presented. The proposed algorithm avoids the above-mentioned limitation problem in the BF algorithm. TRSF is successfully applied to the feature extraction of three actual HSIs. To prove the effectiveness of the proposed algorithm, support vector machines are used to classify the extracted features. Experimental results show that the proposed feature extraction method is simple and effective.
关键词： hyperspectral image (HSI),feature extraction,Bilateral ?ltering (BF),trilateral smooth
更新于2025-09-04 15:30:14
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Can We Generate Good Samples for Hyperspectral Classification? — A Generative Adversarial Network Based Method
摘要： The insufficiency of training samples is really a great challenge for hyperspectral image (HSI) classification. Samples generation is a commonly used technique in deep learning based remote sensing field which can extend the training set. However, previous methods ignore the real distribution of the training samples in the feature space and thus can hardly ensure that the generated samples possess the same patterns with the real ones. In this paper, we propose a generative adversarial network based method (SpecGAN) to handle this problem. Different from traditional GAN framework where the generated samples have no categories, for the first time we take the label information into consideration for hyperspectral images. Feeding a random noise z and a class label vector y into the generator, we can get a spectral sample of the corresponding category. The experiments on the Pavia University data set demonstrate the potential of the proposed SpecGAN in spectral samples generation.
关键词： hyperspectral image classification,generative adversarial network,Sample generation,deep learning
更新于2025-09-04 15:30:14
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Local Similarity Regularized Sparse Representation for Hyperspectral Image Super-Resolution
摘要： Recently, performance of hyperspectral image super-resolution (SR) has been significantly improved via sparse representation. However, most of these existing methods fail to consider the local geometrical structure of the sparse coefficients. To take this crucial issue into account, this paper proposes an effective method, which exploits the location related constraint about the sparse coefficients and incorporates their local similarity into the sparse coding process. Thus, the proposed method can preserve the properties of the aforementioned local geometrical structures. Based on the experimental results, the proposed method is demonstrated to be more effective than previous efforts in the task of hyperspectral image SR.
关键词： Local similarity,Sparse representation,Hyperspectral image,Super-resolution
更新于2025-09-04 15:30:14
-
Differentiation of Deciduous-Calyx and Persistent-Calyx Pears Using NIR Hyperspectral Imaging Analysis
摘要： Korla Fragrant pears are small oval pears characterized by light green skin, crisp texture, and a pleasant aroma after which they are named. The flesh of deciduous-calyx pears is considered more desirable in taste and texture attributes than that of persistent-calyx pears; Chinese packaging standards require that each packed case of highly demanded “superior” class pears contain 5% or less of the persistent-calyx fruits. Near-infrared hyperspectral imaging was investigated as a potential method for automatic sorting of the two types of pears. The hyperspectral images were analyzed, and wavebands at 1190 nm and 1199 nm were selected for differentiating deciduous-calyx fruits from persistent-calyx fruits. A multispectral differentiation algorithm using the ratio of the pears’ relative intensities at 1190 nm and 1199 nm was developed. The results showed that the algorithm correctly classified 89.3% to 94.0% of deciduous-calyx pears and effectively differentiated pears such that the number of persistent-calyx pears misclassified as deciduous-calyx pears comprised only 2.4% to 4.9% of all pears classified as deciduous-calyx pears, performing well within the targeted packaging standards.
关键词： Classification,Korla fragrant pear,Hyperspectral image,Fruit quality
更新于2025-09-04 15:30:14