Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Spectral residual methods.
Статті в журналах з теми "Spectral residual methods"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 статей у журналах для дослідження на тему "Spectral residual methods".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте статті в журналах для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Canuto, Claudio, and Alfio Quarteroni. "Preconditioned minimal residual methods for chebyshev spectral calculations." Journal of Computational Physics 60, no. 2 (September 1985): 315–37. http://dx.doi.org/10.1016/0021-9991(85)90010-5.
Повний текст джерела
2
TSUCHIDA, Satoshi, Soichiro TANAKA, and Takayuki ODAJIMA. "Spectral Pattern Index for Logarithmic Residual and the similar methods." Journal of the Japan society of photogrammetry and remote sensing 32, no. 1 (1993): 25–35. http://dx.doi.org/10.4287/jsprs.32.25.
Повний текст джерела
3
Wong, Yau Shu, Thomas A. Zang, and M. Yousuff Hussaini. "Preconditioned conjugate residual methods for the solution of spectral equations." Computers & Fluids 14, no. 2 (January 1986): 85–95. http://dx.doi.org/10.1016/0045-7930(86)90001-0.
Повний текст джерела
4
Schiff, N. D., K. P. Purpura, and J. D. Victor. "Gating of Local Network Signals Appears as Stimulus-Dependent Activity Envelopes in Striate Cortex." Journal of Neurophysiology 82, no. 5 (November 1, 1999): 2182–96. http://dx.doi.org/10.1152/jn.1999.82.5.2182.
Повний текст джерелаАнотація:
Neuronal activity often is treated as a composition of a stimulus-driven component and a second component that corrupts the signal, adding or deleting spikes at random. Standard quantitative methods such as peristimulus histograms and Fourier analysis use stimulus-locked averaging to enhance detection of the driven component of neuronal responses and de-emphasize the “noise.” However, neural activity also includes bursts, oscillations, and other episodic events that standard averaging methods overlook. If this activity is stimulus independent, it can be characterized by standard power spectral analysis (or autocorrelation). But activity that is excited by (but not temporally locked to) the visual stimulus cannot be characterized by averaging or standard spectral analysis. Phase-locked spectral analysis (PLSA) is a new method that examines this “residual” activity—the difference between the individual responses to each cycle of a periodic stimulus and their average. With PLSA, residual activity is characterized in terms of temporal envelopes and their carriers. Previously, PLSA demonstrated broadband interactions between periodic visual stimuli and fluctuations in the local field potential of macaque V1. In the present study, single-unit responses (SUA) from parafoveal V1 in anesthetized macaque monkey are examined with this technique. Recordings were made from 21 neurons, 6 of which were recorded in pairs along with multiunit activity (MUA) from separate electrodes and 8 of which were recorded along with MUA from the same electrode. PLSA was applied to responses to preferred (orientation, direction, and spatial frequency) and nonpreferred drifting gratings. For preferred stimuli, all cells demonstrated broadband (1–10 Hz and higher) residual activity that waxed and waned with the stimulus cycle, suggesting that changes in the residual activity are introduced routinely by visual stimulation. Moreover, some reconstructed envelopes indicate that the residual activity was sharply gated by the stimulus cycle. Oscillations occasionally were seen in the power spectrum of single units. Phase-locked cross-spectra were determined for 3 SUA/SUA pairs and 11 SUA/MUA pairs. Residual activity in the cross-spectra was generally much less than the residual activity determined separately from each neuron. The reduction in the residual activity in the cross-spectra suggests that nearby neurons may gate inputs from distinct and relatively independent neuronal subpopulations that together generate the background rhythms of striate cortex.
5
Li, Shaodan, Shiyu Fu, and Dongbo Zheng. "Rural Built-Up Area Extraction from Remote Sensing Images Using Spectral Residual Methods with Embedded Deep Neural Network." Sustainability 14, no. 3 (January 24, 2022): 1272. http://dx.doi.org/10.3390/su14031272.
Повний текст джерелаАнотація:
A rural built-up area is one of the most important features of rural regions. Rapid and accurate extraction of rural built-up areas has great significance to rural planning and urbanization. In this paper, the spectral residual method is embedded into a deep neural network to accurately describe the rural built-up areas from large-scale satellite images. Our proposed method is composed of two processes: coarse localization and fine extraction. Firstly, an improved Faster R-CNN (Regions with Convolutional Neural Network) detector is trained to obtain the coarse localization of the candidate built-up areas, and then the spectral residual method is used to describe the accurate boundary of each built-up area based on the bounding boxes. In the experimental part, we firstly explored the relationship between the sizes of built-up areas and the kernels in the spectral residual method. Then, the comparing experiments demonstrate that our proposed method has better performance in the extraction of rural built-up areas.
6
Kim, Cheolsun, Dongju Park, and Heung-No Lee. "Compressive Sensing Spectroscopy Using a Residual Convolutional Neural Network." Sensors 20, no. 3 (January 21, 2020): 594. http://dx.doi.org/10.3390/s20030594.
Повний текст джерелаАнотація:
Compressive sensing (CS) spectroscopy is well known for developing a compact spectrometer which consists of two parts: compressively measuring an input spectrum and recovering the spectrum using reconstruction techniques. Our goal here is to propose a novel residual convolutional neural network (ResCNN) for reconstructing the spectrum from the compressed measurements. The proposed ResCNN comprises learnable layers and a residual connection between the input and the output of these learnable layers. The ResCNN is trained using both synthetic and measured spectral datasets. The results demonstrate that ResCNN shows better spectral recovery performance in terms of average root mean squared errors (RMSEs) and peak signal to noise ratios (PSNRs) than existing approaches such as the sparse recovery methods and the spectral recovery using CNN. Unlike sparse recovery methods, ResCNN does not require a priori knowledge of a sparsifying basis nor prior information on the spectral features of the dataset. Moreover, ResCNN produces stable reconstructions under noisy conditions. Finally, ResCNN is converged faster than CNN.
7
Liu, Qin, Letong Han, Rui Tan, Hongfei Fan, Weiqi Li, Hongming Zhu, Bowen Du, and Sicong Liu. "Hybrid Attention Based Residual Network for Pansharpening." Remote Sensing 13, no. 10 (May 18, 2021): 1962. http://dx.doi.org/10.3390/rs13101962.
Повний текст джерелаАнотація:
Pansharpening aims at fusing the rich spectral information of multispectral (MS) images and the spatial details of panchromatic (PAN) images to generate a fused image with both high resolutions. In general, the existing pansharpening methods suffer from the problems of spectral distortion and lack of spatial detail information, which might prevent the accuracy computation for ground object identification. To alleviate these problems, we propose a Hybrid Attention mechanism-based Residual Neural Network (HARNN). In the proposed network, we develop an encoder attention module in the feature extraction part to better utilize the spectral and spatial features of MS and PAN images. Furthermore, the fusion attention module is designed to alleviate spectral distortion and improve contour details of the fused image. A series of ablation and contrast experiments are conducted on GF-1 and GF-2 datasets. The fusion results with less distorted pixels and more spatial details demonstrate that HARNN can implement the pansharpening task effectively, which outperforms the state-of-the-art algorithms.
8
He, S., H. Jing, and H. Xue. "SPECTRAL-SPATIAL MULTISCALE RESIDUAL NETWORK FOR HYPERSPECTRAL IMAGE CLASSIFICATION." International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLIII-B3-2022 (May 30, 2022): 389–95. http://dx.doi.org/10.5194/isprs-archives-xliii-b3-2022-389-2022.
Повний текст джерелаАнотація:
Abstract. In recent years, deep neural networks (DNN) are commonly adopted for hyperspectral image (HSI) classification. As the most representative supervised DNN model, convolutional neural networks (CNNs) have outperformed most algorithms. But the main problem of CNN-based methods lies in the over-smoothing phenomenon. Meanwhile, mainstream methods usually require a large number of samples and a large amount of computation. A multi-task learning spectral-spatial multiscale residual network (SSMRN) is proposed to learn features of objects effectively. In the implementation of the SSMRN, a multiscale residual convolutional neural network (MRCNN) is proposed as spatial feature extractors and a band grouping-based bi-directional gated recurrent unit (Bi-GRU) is utilized as spectral feature extractors. To evaluate the effectiveness of the SSMRN, extensive experiments are conducted on public benchmark data sets. The proposed method can retain the detailed boundary of different objects better and yield a competitive performance compared with two state-of-the-art methods especially when the training samples are inadequate.
9
Meng, Zhe, Lingling Li, Xu Tang, Zhixi Feng, Licheng Jiao, and Miaomiao Liang. "Multipath Residual Network for Spectral-Spatial Hyperspectral Image Classification." Remote Sensing 11, no. 16 (August 13, 2019): 1896. http://dx.doi.org/10.3390/rs11161896.
Повний текст джерелаАнотація:
Convolutional neural networks (CNNs) have recently shown outstanding capability for hyperspectral image (HSI) classification. In this work, a novel CNN model is proposed, which is wider than other existing deep learning-based HSI classification models. Based on the fact that very deep residual networks (ResNets) behave like an ensemble of relatively shallow networks, our proposed network, called multipath ResNet (MPRN), employs multiple residual functions in the residual blocks to make the network wider, rather than deeper. The proposed network consists of shorter-medium paths for efficient gradient flow and replaces the stacking of multiple residual blocks in ResNet with fewer residual blocks but more parallel residual functions in each of it. Experimental results on three real hyperspectral data sets demonstrate the superiority of the proposed method over several state-of-the-art classification methods.
10
Shamsipour, Pejman, Michel Chouteau, and Denis Marcotte. "Data analysis of potential field methods using geostatistics." GEOPHYSICS 82, no. 2 (March 1, 2017): G35—G44. http://dx.doi.org/10.1190/geo2015-0631.1.
Повний текст джерелаАнотація:
Processing of potential field data is commonly done by spectral methods because of their low computational complexity. However, we have studied some geostatistical methods to process the potential field data, and we find the advantages of using these spatial methods. First, we investigate transformation of data by kriging using a gravimetric model of covariance, we compare this approach with the spectral method, and we find its advantage when the data were sparse and not on a regular grid using a synthetic example as well as a field data example. Then, we use factorial kriging for noise reduction and separation of the regional and residual components. This method does not have some of the practical limitations that the spectral-based methods encounter. Finally, we determine the flexibility of interpolation using nonstationary covariances.
11
Chen, Wenjing, Xiangtao Zheng, and Xiaoqiang Lu. "Hyperspectral Image Super-Resolution with Self-Supervised Spectral-Spatial Residual Network." Remote Sensing 13, no. 7 (March 26, 2021): 1260. http://dx.doi.org/10.3390/rs13071260.
Повний текст джерелаАнотація:
Recently, many convolutional networks have been built to fuse a low spatial resolution (LR) hyperspectral image (HSI) and a high spatial resolution (HR) multispectral image (MSI) to obtain HR HSIs. However, most deep learning-based methods are supervised methods, which require sufficient HR HSIs for supervised training. Collecting plenty of HR HSIs is laborious and time-consuming. In this paper, a self-supervised spectral-spatial residual network (SSRN) is proposed to alleviate dependence on a mass of HR HSIs. In SSRN, the fusion of HR MSIs and LR HSIs is considered a pixel-wise spectral mapping problem. Firstly, this paper assumes that the spectral mapping between HR MSIs and HR HSIs can be approximated by the spectral mapping between LR MSIs (derived from HR MSIs) and LR HSIs. Secondly, the spectral mapping between LR MSIs and LR HSIs is explored by SSRN. Finally, a self-supervised fine-tuning strategy is proposed to transfer the learned spectral mapping to generate HR HSIs. SSRN does not require HR HSIs as the supervised information in training. Simulated and real hyperspectral databases are utilized to verify the performance of SSRN.
12
Zhao, Rui, and Shihong Du. "Spectral-Spatial Residual Network for Fusing Hyperspectral and Panchromatic Remote Sensing Images." Remote Sensing 14, no. 3 (February 8, 2022): 800. http://dx.doi.org/10.3390/rs14030800.
Повний текст джерелаАнотація:
Fusing hyperspectral and panchromatic remote sensing images can obtain the images with high resolution in both spectral and spatial domains. In addition, it can complement the deficiency of high-resolution hyperspectral and panchromatic remote sensing images. In this paper, a spectral–spatial residual network (SSRN) model is established for the intelligent fusion of hyperspectral and panchromatic remote sensing images. Firstly, the spectral–spatial deep feature branches are built to extract the representative spectral and spatial deep features, respectively. Secondly, an enhanced multi-scale residual network is established for the spatial deep feature branch. In addition, an enhanced residual network is established for the spectral deep feature branch This operation is adopted to enhance the spectral and spatial deep features. Finally, this method establishes the spectral–spatial deep feature simultaneity to circumvent the independence of spectral and spatial deep features. The proposed model was evaluated on three groups of real-world hyperspectral and panchromatic image datasets which are collected with a ZY-1E sensor and are located at Baiyangdian, Chaohu and Dianchi, respectively. The experimental results and quality evaluation values, including RMSE, SAM, SCC, spectral curve comparison, PSNR, SSIM ERGAS and Q metric, confirm the superior performance of the proposed model compared with the state-of-the-art methods, including AWLP, CNMF, GIHS, MTF_GLP, HPF and SFIM methods.
13
Khotimah, Wijayanti Nurul, Mohammed Bennamoun, Farid Boussaid, Ferdous Sohel, and David Edwards. "A High-Performance Spectral-Spatial Residual Network for Hyperspectral Image Classification with Small Training Data." Remote Sensing 12, no. 19 (September 24, 2020): 3137. http://dx.doi.org/10.3390/rs12193137.
Повний текст джерелаАнотація:
In this paper, we propose a high performance Two-Stream spectral-spatial Residual Network (TSRN) for hyperspectral image classification. The first spectral residual network (sRN) stream is used to extract spectral characteristics, and the second spatial residual network (saRN) stream is concurrently used to extract spatial features. The sRN uses 1D convolutional layers to fit the spectral data structure, while the saRN uses 2D convolutional layers to match the hyperspectral spatial data structure. Furthermore, each convolutional layer is preceded by a Batch Normalization (BN) layer that works as a regularizer to speed up the training process and to improve the accuracy. We conducted experiments on three well-known hyperspectral datasets, and we compare our results with five contemporary methods across various sizes of training samples. The experimental results show that the proposed architecture can be trained with small size datasets and outperforms the state-of-the-art methods in terms of the Overall Accuracy, Average Accuracy, Kappa Value, and training time.
14
Pavarino, Luca F. "Preconditioned conjugate residual methods for mixed spectral discretizations of elasticity and Stokes problems." Computer Methods in Applied Mechanics and Engineering 146, no. 1-2 (July 1997): 19–30. http://dx.doi.org/10.1016/s0045-7825(96)01224-8.
Повний текст джерела
15
Canuto, Claudio, Alessandro Russo, and Vincent van Kemenade. "Stabilized spectral methods for the Navier-Stokes equations: residual-free bubbles and preconditioning." Computer Methods in Applied Mechanics and Engineering 166, no. 1-2 (November 1998): 65–83. http://dx.doi.org/10.1016/s0045-7825(98)00083-8.
Повний текст джерела
16
Saranathan, S., M. van Noort, and S. K. Solanki. "Correction of atmospheric stray light in restored slit spectra." Astronomy & Astrophysics 653 (August 31, 2021): A17. http://dx.doi.org/10.1051/0004-6361/201937100.
Повний текст джерелаАнотація:
Context. A long-standing issue in solar ground-based observations has been the contamination of data due to stray light, which is particularly relevant in inversions of spectropolarimetric data. Aims. We aim to build on a statistical method of correcting stray-light contamination due to residual high-order aberrations and apply it to ground-based slit spectra. Methods. The observations were obtained at the Swedish Solar Telescope, and restored using the multi-frame blind deconvolution restoration procedure. Using the statistical properties of seeing, we created artificially degraded synthetic images generated from magneto-hydrodynamic simulations. We then compared the synthetic data with the observations to derive estimates of the amount of the residual stray light in the observations. In the final step, the slit spectra were deconvolved with a stray-light point spread function to remove the residual stray light from the observations. Results. The RMS granulation contrasts of the deconvolved spectra were found to increase to approximately 12.5%, from 9%. Spectral lines, on average, were found to become deeper in the granules and shallower in the inter-granular lanes, indicating systematic changes to gradients in temperature. The deconvolution was also found to increase the redshifts and blueshifts of spectral lines, suggesting that the velocities of granulation in the solar photosphere are higher than had previously been observed.
17
Zhang, Jing, Minhao Shao, Zekang Wan, and Yunsong Li. "Multi-Scale Feature Mapping Network for Hyperspectral Image Super-Resolution." Remote Sensing 13, no. 20 (October 19, 2021): 4180. http://dx.doi.org/10.3390/rs13204180.
Повний текст джерелаАнотація:
Hyperspectral Image (HSI) can continuously cover tens or even hundreds of spectral segments for each spatial pixel. Limited by the cost and commercialization requirements of remote sensing satellites, HSIs often lose a lot of information due to insufficient image spatial resolution. For the high-dimensional nature of HSIs and the correlation between the spectra, the existing Super-Resolution (SR) methods for HSIs have the problems of excessive parameter amount and insufficient information complementarity between the spectra. This paper proposes a Multi-Scale Feature Mapping Network (MSFMNet) based on the cascaded residual learning to adaptively learn the prior information of HSIs. MSFMNet simplifies each part of the network into a few simple yet effective network modules. To learn the spatial-spectral characteristics among different spectral segments, a multi-scale feature generation and fusion Multi-Scale Feature Mapping Block (MSFMB) based on wavelet transform and spatial attention mechanism is designed in MSFMNet to learn the spectral features between different spectral segments. To effectively improve the multiplexing rate of multi-level spectral features, a Multi-Level Feature Fusion Block (MLFFB) is designed to fuse the multi-level spectral features. In the image reconstruction stage, an optimized sub-pixel convolution module is used for the up-sampling of different spectral segments. Through a large number of verifications on the three general hyperspectral datasets, the superiority of this method compared with the existing hyperspectral SR methods is proved. In subjective and objective experiments, its experimental performance is better than its competitors.
18
Guspí, Fernando, and Beatriz Introcaso. "A sparse spectrum technique for gridding and separating potential field anomalies." GEOPHYSICS 65, no. 4 (July 2000): 1154–61. http://dx.doi.org/10.1190/1.1444808.
Повний текст джерелаАнотація:
The separation of regional and residual potential field anomalies, regarded as a spectral problem, can be greatly facilitated when a spectrum estimate shows a clear break between low‐ and high‐frequency components, a feature that normal fast‐Fourier‐transform (FFT) methods fail to present. In this work, we model the discrete Fourier transform of a potential field, measured at stations irregularly distributed on a surface, by means of a high‐resolution sparse estimate derived originally for seismic signal processing. The coefficients of this estimate, which are distributed according to the Cauchy probability law, produce a model with only few components having a significant value. A steepest‐descent algorithm gives a computing alternative to large matrix multiplications and inversions. Advantages of taking this approach are twofold. First, the high‐resolution transform can be used as a gridding tool to evaluate the potential field either on a horizontal plane or on the topographic surface. The enhancement of the spectral peaks and the virtual absence of sidelobes prevents oscillations and edge effects in the result. Secondly, the sparse distribution of the spectral elements allows the interpreter to locate clearly the low‐frequency components related to the regional field. After a second and faster pass, the values of those coefficients can be redefined in order to obtain a more robust separation, ajusting the residuals by the Cauchy criterion. A theoretical noise‐free example to separate the magnetic anomaly of a prism from a polynomial background illustrates well the difference between sparse and FFT spectra. An example with real Bouguer anomalies in the Interserrana basin, Argentina, shows that gridding results, in this case reduced to sea level, compare well with those obtained by other gridding methods, and that the separation procedure is able to outline well defined areas of positive and negative residual anomalies.
19
Mu, Kai, Ziyuan Zhang, Yurong Qian, Suhong Liu, Mengting Sun, and Ranran Qi. "SRT: A Spectral Reconstruction Network for GF-1 PMS Data Based on Transformer and ResNet." Remote Sensing 14, no. 13 (July 1, 2022): 3163. http://dx.doi.org/10.3390/rs14133163.
Повний текст джерелаАнотація:
The time of acquiring remote sensing data was halved after the joint operation of Gao Fen-6 (GF-6) and Gao Fen-1 (GF-1) satellites. Meanwhile, GF-6 added four bands, including the “red-edge” band that can effectively reflect the unique spectral characteristics of crops. However, GF-1 data do not contain these bands, which greatly limits their application to crop-related joint monitoring. In this paper, we propose a spectral reconstruction network (SRT) based on Transformer and ResNet to reconstruct the missing bands of GF-1. SRT is composed of three modules: (1) The transformer feature extraction module (TFEM) fully extracts the correlation features between spectra. (2) The residual dense module (RDM) reconstructs local features and avoids the vanishing gradient problem. (3) The residual global construction module (RGM) reconstructs global features and preserves texture details. Compared with competing methods, such as AWAN, HRNet, HSCNN-D, and M2HNet, the proposed method proved to have higher accuracy by a margin of the mean relative absolute error (MRAE) and root mean squared error (RMSE) of 0.022 and 0.009, respectively. It also achieved the best accuracy in supervised classification based on support vector machine (SVM) and spectral angle mapper (SAM).
20
Massie, Christine, Keren Chen, and Andrew J. Berger. "Calibration Technique for Suppressing Residual Etalon Artifacts in Slit-Averaged Raman Spectroscopy." Applied Spectroscopy 76, no. 2 (October 1, 2021): 255–61. http://dx.doi.org/10.1177/00037028211046643.
Повний текст джерелаАнотація:
Back-illuminated charged-coupled device (BI-CCD) arrays increase quantum efficiency but also amplify etaloning, a multiplicative, wavelength-dependent fixed-pattern effect. When spectral data from hundreds of BI-CCD rows are combined, the averaged spectrum will generally appear etalon-free. This can mask substantial etaloning at the row level, even if the BI-CCD has been treated to suppress the effect. This paper compares two methods of etalon correction, one with simple averaging and one with row-by-row calibration using a fluorescence standard. Two BI-CCD arrays, both roughened by the supplier to reduce etaloning, were used to acquire Raman spectra of murine bone specimens. For one array, etaloning was the dominant source of noise under the exposure conditions chosen, even for the averaged spectrum across all rows; near-infrared-excited Raman peaks were noticeably affected. In this case, row-by-row calibration improved the spectral quality of the average spectrum. The other CCD’s performance was shot-noise limited and therefore received no benefit from the extra calibration. The different results highlight the importance of checking for and correcting row-level fixed pattern when measuring weak Raman signals in the presence of a large fluorescence background.
21
Zou, Liang, Zhifan Zhang, Haijia Du, Meng Lei, Yong Xue, and Z. Jane Wang. "DA-IMRN: Dual-Attention-Guided Interactive Multi-Scale Residual Network for Hyperspectral Image Classification." Remote Sensing 14, no. 3 (January 23, 2022): 530. http://dx.doi.org/10.3390/rs14030530.
Повний текст джерелаАнотація:
Deep learning-based fusion of spectral-spatial information is increasingly dominant for hyperspectral image (HSI) classification. However, due to insufficient samples, current feature fusion methods often neglect joint interactions. In this paper, to further improve the classification accuracy, we propose a dual-attention-guided interactive multi-scale residual network (DA-IMRN) to explore the joint spectral-spatial information and assign pixel-wise labels for HSIs without information leakage. In DA-IMRN, two branches focusing on spatial and spectral information separately are employed for feature extraction. A bidirectional-attention mechanism is employed to guide the interactive feature learning between two branches and promote refined feature maps. In addition, we extract deep multi-scale features corresponding to multiple receptive fields from limited samples via a multi-scale spectral/spatial residual block, to improve classification performance. Experimental results on three benchmark datasets (i.e., Salinas Valley, Pavia University, and Indian Pines) support that attention-guided multi-scale feature learning can effectively explore the joint spectral-spatial information. The proposed method outperforms state-of-the-art methods with the overall accuracy of 91.26%, 93.33%, and 82.38%, and the average accuracy of 94.22%, 89.61%, and 80.35%, respectively.
22
Zhang, Tianyu, Cuiping Shi, Diling Liao, and Liguo Wang. "Deep Spectral Spatial Inverted Residual Network for Hyperspectral Image Classification." Remote Sensing 13, no. 21 (November 7, 2021): 4472. http://dx.doi.org/10.3390/rs13214472.
Повний текст джерелаАнотація:
Convolutional neural networks (CNNs) have been widely used in hyperspectral image classification in recent years. The training of CNNs relies on a large amount of labeled sample data. However, the number of labeled samples of hyperspectral data is relatively small. Moreover, for hyperspectral images, fully extracting spectral and spatial feature information is the key to achieve high classification performance. To solve the above issues, a deep spectral spatial inverted residuals network (DSSIRNet) is proposed. In this network, a data block random erasing strategy is introduced to alleviate the problem of limited labeled samples by data augmentation of small spatial blocks. In addition, a deep inverted residuals (DIR) module for spectral spatial feature extraction is proposed, which locks the effective features of each layer while avoiding network degradation. Furthermore, a global 3D attention module is proposed, which can realize the fine extraction of spectral and spatial global context information under the condition of the same number of input and output feature maps. Experiments are carried out on four commonly used hyperspectral datasets. A large number of experimental results show that compared with some state-of-the-art classification methods, the proposed method can provide higher classification accuracy for hyperspectral images.
23
Dupuy, Nathalie, Jean Pierre Huvenne, Ludovic Duponchel, and Pierre Legrand. "Classification of Green Coffees by FT-IR Analysis of Dry Extract." Applied Spectroscopy 49, no. 5 (May 1995): 580–85. http://dx.doi.org/10.1366/0003702953964174.
Повний текст джерелаАнотація:
Principal component analysis (PCA) of infrared spectra has been used as a classification method for the green beans of coffee from various origin. Before spectral acquisition, sampling methods were tested for 45 samples, and we chose dry extract of water-soluble compounds on SiCaF2 supports. After PCA of the first derivatized spectra, the first four loadings were examined. The scores of the second principal component appear to be directly correlated by their sign to the species arabica or robusta. This result allows an easy classification. In the same way, the pigmentation is well characterized into two groups on the scattergram of the samples with respect to the PC1 and PC3 components. Another feature of this method is that the analysis of the spectral data in terms of residual variance separate components which are correlated with properties. This approach provides assistance in the interpretation of infrared spectra of complex mixtures.
24
Melgaard, David K., David M. Haaland, and Christine M. Wehlburg. "Concentration Residual Augmented Classical Least Squares (CRACLS): A Multivariate Calibration Method with Advantages over Partial Least Squares." Applied Spectroscopy 56, no. 5 (May 2002): 615–24. http://dx.doi.org/10.1366/0003702021955178.
Повний текст джерелаАнотація:
A significant extension to the classical least-squares (CLS) algorithm called concentration residual augmented CLS (CRACLS) has been developed. Previously, unmodeled sources of spectral variation have rendered CLS models ineffective for most types of problems, but with the new CRACLS algorithm, CLS-type models can be applied to a significantly wider range of applications. This new quantitative multivariate spectral analysis algorithm iteratively augments the calibration matrix of reference concentrations with concentration residuals estimated during CLS prediction. Because these residuals represent linear combinations of the unmodeled spectrally active component concentrations, the effects of these components are removed from the calibration of the analytes of interest. This iterative process allows the development of a CLS-type calibration model comparable in prediction ability to implicit multivariate calibration methods such as partial least squares (PLS) even when unmodeled spectrally active components are present in the calibration sample spectra. In addition, CRACLS retains the improved qualitative spectral information of the CLS algorithm relative to PLS. More importantly, CRACLS provides a model compatible with the recently presented prediction-augmented CLS (PACLS) method. The CRACLS/PACLS combination generates an adaptable model that can achieve excellent prediction ability for samples of unknown composition that contain unmodeled sources of spectral variation. The CRACLS algorithm is demonstrated with both simulated and real data derived from a system of dilute aqueous solutions containing glucose, ethanol, and urea. The simulated data demonstrate the effectiveness of the new algorithm and help elucidate the principles behind the method. Using experimental data, we compare the prediction abilities of CRACLS and PLS during cross-validated calibration. In combination with PACLS, the CRACLS predictions are comparable to PLS for the prediction of the glucose, ethanol, and urea components for validation samples collected when significant instrument drift was present. However, the PLS predictions required recalibration using nonstandard cross-validated rotations while CRACLS/PACLS was rapidly updated during prediction without the need for time-consuming cross-validated recalibration. The CRACLS/PACLS algorithm provides a more general approach to removing the detrimental effects of unmodeled components.
25
He, Luxiao, Mi Wang, Ying Zhu, Xueli Chang, and Xiaoxiao Feng. "Image Fusion for High-Resolution Optical Satellites Based on Panchromatic Spectral Decomposition." Sensors 19, no. 11 (June 9, 2019): 2619. http://dx.doi.org/10.3390/s19112619.
Повний текст джерелаАнотація:
Ratio transformation methods are widely used for image fusion of high-resolution optical satellites. The premise for the use the ratio transformation is that there is a zero-bias linear relationship between the panchromatic band and the corresponding multi-spectral bands. However, there are bias terms and residual terms with large values in reality, depending on the sensors, the response spectral ranges, and the land-cover types. To address this problem, this paper proposes a panchromatic and multi-spectral image fusion method based on the panchromatic spectral decomposition (PSD). The low-resolution panchromatic and multi-spectral images are used to solve the proportionality coefficients, the bias coefficients, and the residual matrixes. These coefficients are substituted into the high-resolution panchromatic band and decompose it into the high-resolution multi-spectral bands. The experiments show that this method can make the fused image acquire high color fidelity and sharpness, it is robust to different sensors and features, and it can be applied to the panchromatic and multi-spectral fusion of high-resolution optical satellites.
26
Sun, Hezhi, Ke Zheng, Ming Liu, Chao Li, Dong Yang, and Jindong Li. "Hyperspectral Image Mixed Noise Removal Using a Subspace Projection Attention and Residual Channel Attention Network." Remote Sensing 14, no. 9 (April 26, 2022): 2071. http://dx.doi.org/10.3390/rs14092071.
Повний текст джерелаАнотація:
Although the existing deep-learning-based hyperspectral image (HSI) denoising methods have achieved tremendous success, recovering high-quality HSIs in complex scenes that contain mixed noise is still challenging. Besides, these methods have not fully explored the local and global spatial–spectral information of HSIs. To address the above issues, a novel HSI mixed noise removal network called subspace projection attention and residual channel attention network (SPARCA-Net) is proposed. Specifically, we propose an orthogonal subspace projection attention (OSPA) module to adaptively learn to generate bases of the signal subspace and project the input into such space to remove noise. By leveraging the local and global spatial relations, OSPA is able to reconstruct the local structure of the feature maps more precisely. We further propose a residual channel attention (RCA) module to emphasize the interdependence between feature maps and exploit the global channel correlation of them, which could enhance the channel-wise adaptive learning. In addition, multiscale joint spatial–spectral input and residual learning strategies are employed to capture multiscale spatial–spectral features and reduce the degradation problem, respectively. Synthetic and real HSI data experiments demonstrated that the proposed HSI denoising network outperforms many of the advanced methods in both quantitative and qualitative assessments.
27
Khodasevich, M. A., E. A. Scorbanov, and M. V. Rogovaya. "Application of Multivariate Analysis of Broadband Transmission Spectra for Calibration of Physico-Chemical Parameters of Wines." Devices and Methods of Measurements 10, no. 2 (June 24, 2019): 198–206. http://dx.doi.org/10.21122/2220-9506-2019-10-2-198-206.
Повний текст джерелаАнотація:
The use of multivariate processing of spectral information has recently been favored due to the express nature of this method, the ease of use of mathematical packages, and the lack of the need to add chemical reagents. The aim of the work is using the methods of multivariate analysis of broadband transmission spectra to calibrate the physicochemical parameters of wines and to improve the accuracy of this calibration by selecting spectral variables.Using the interval projection to latent structures of the transmission spectra in the range of 220– 2500 nm, the physicochemical characteristics of the varietal unblended Moldovan wine are calibrated. Interval methods of multivariate data analysis allow signifi reducing the root mean square calibration error in comparison with the broadband multivariate methods. Residual predictive deviations exceed the threshold value of 2.5 for K, Ca, Mg, oxalic, malic and succinic acids, 2,3-butylene glycol, ash and phenolic compounds for red wines and Mg, tartaric, citric and lactic acids, 2,3-butylene glycol, ash, phenolic compounds and soluble salts for white wines. These values demonstrate good calibration quality.The application of the proposed method for calibrating the physicochemical parameters of wines makes it possible to replace traditional methods with spectral measurements, which are available not only in laboratory but also in the fi and characterized by small values of the root mean square error of calibration.
28
Shen, Weizheng, Qingming Kong, Jianbo Wang, Nan Ji, and Zhongbin Su. "Bioagriculture Outlier Elimination Based on 3D View ofX-YVariance and Leverage Measurement." Mathematical Problems in Engineering 2015 (2015): 1–9. http://dx.doi.org/10.1155/2015/375827.
Повний текст джерелаАнотація:
Aiming at effective outlier elimination in the biological near-infrared spectral and achieving high accuracy predictive modeling, this paper proposes a novel outlier elimination method based onX-Yvariance and leverage analysis. Firstly, the characters of near-infrared spectral are summarized; then residual sampleX-variance, leverage, and residual sampleY-variance are concatenated as a divergence measurement. We further compared the proposed method withX-Yvariance, Mahalanobis distance, and HotellingT2statistical analysis; the experiment results demonstrate that the proposed methods have competitive outlier elimination and better performance in time complexity and accuracy. The proposed method can also be adopted for other outlier elimination tasks.
29
Yang, Ying, Yong Xie, Xunhao Chen, and Yubao Sun. "Hyperspectral Snapshot Compressive Imaging with Non-Local Spatial-Spectral Residual Network." Remote Sensing 13, no. 9 (May 6, 2021): 1812. http://dx.doi.org/10.3390/rs13091812.
Повний текст джерелаАнотація:
Snapshot Compressive Imaging is an emerging technology that is based on compressive sensing theory to achieve high-efficiency hyperspectral data acquisition. The core problem of this technology is how to reconstruct 3D hyperspectral data from the 2D snapshot measurement in a fast and high-quality manner. In this paper, we propose a novel deep network, which consists of the symmetric residual module and the non-local spatial-spectral attention module, to learn the reconstruction mapping in a data-driven way. The symmetric residual module uses symmetric residual connections to improve the potential of interaction between convolution operations and further promotes the fusion of local features. The non-local spatial-spectral attention module is designed to capture the non-local spatial-spectral correlation in the hyperspectral image. Specifically, this module calculates the channel attention matrix to capture the global correlations between all of the spectral channels, and it fuses the channel attention attained feature maps and the spatial attention weighted features as the module output, thus both of the spatial-spectral correlations of hyperspectral images can be fully utilized for reconstruction. In addition, a compound loss, including the reconstruction loss, the measurement loss, and the cosine loss, is designed to guide the end-to-end network learning. We experimentally evaluate the proposed method on simulation and real datasets. The experimental results show that the proposed network outperforms the competing methods in terms of the reconstruction quality and running time.
30
Zhang, Tianyu, Cuiping Shi, Diling Liao, and Liguo Wang. "A Spectral Spatial Attention Fusion with Deformable Convolutional Residual Network for Hyperspectral Image Classification." Remote Sensing 13, no. 18 (September 9, 2021): 3590. http://dx.doi.org/10.3390/rs13183590.
Повний текст джерелаАнотація:
Convolutional neural networks (CNNs) have exhibited excellent performance in hyperspectral image classification. However, due to the lack of labeled hyperspectral data, it is difficult to achieve high classification accuracy of hyperspectral images with fewer training samples. In addition, although some deep learning techniques have been used in hyperspectral image classification, due to the abundant information of hyperspectral images, the problem of insufficient spatial spectral feature extraction still exists. To address the aforementioned issues, a spectral–spatial attention fusion with a deformable convolution residual network (SSAF-DCR) is proposed for hyperspectral image classification. The proposed network is composed of three parts, and each part is connected sequentially to extract features. In the first part, a dense spectral block is utilized to reuse spectral features as much as possible, and a spectral attention block that can refine and optimize the spectral features follows. In the second part, spatial features are extracted and selected by a dense spatial block and attention block, respectively. Then, the results of the first two parts are fused and sent to the third part, and deep spatial features are extracted by the DCR block. The above three parts realize the effective extraction of spectral–spatial features, and the experimental results for four commonly used hyperspectral datasets demonstrate that the proposed SSAF-DCR method is superior to some state-of-the-art methods with very few training samples.
31
Zabalza, Maialen, and Angela Bernardini. "Super-Resolution of Sentinel-2 Images Using a Spectral Attention Mechanism." Remote Sensing 14, no. 12 (June 16, 2022): 2890. http://dx.doi.org/10.3390/rs14122890.
Повний текст джерелаАнотація:
Many visual applications require high-resolution images for an adequate interpretation of the data stored within them. In remote sensing, the appearance of satellites such as Sentinel or Landsat has facilitated the access to data thanks to their free offer of multispectral images. However, the spatial resolution of these satellites is insufficient for many tasks. Therefore, the objective of this work is to apply deep learning techniques to increase the resolution of the Sentinel-2 Read-Green-Blue-NIR (RGBN) bands from the original 10 m to 2.5 m. This means multiplying the number of pixels in the resulting image by 4, improving the perception and visual quality. In this work, we implement a state-of-the-art residual learning-based model called Super-Resolution Residual Network (SRResNet), which we train using PlanetScope-Sentinel pairs of images. Our model, named SARNet (Spectral Attention Residual Network), incorporates Residual Channel Attention Blocks (RCAB) to improve the performance of the network and the visual quality of the results. The experiments we have carried out show that SARNet offers better results than other state-of-the-art methods.
32
Yuan, Q., Y. Ang, and H. Z. M. Shafri. "HYPERSPECTRAL IMAGE CLASSIFICATION USING RESIDUAL 2D AND 3D CONVOLUTIONAL NEURAL NETWORK JOINT ATTENTION MODEL." International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLIV-M-3-2021 (August 10, 2021): 187–93. http://dx.doi.org/10.5194/isprs-archives-xliv-m-3-2021-187-2021.
Повний текст джерелаАнотація:
Abstract. Hyperspectral image classification (HSIC) is a challenging task in remote sensing data analysis, which has been applied in many domains for better identification and inspection of the earth surface by extracting spectral and spatial information. The combination of abundant spectral features and accurate spatial information can improve classification accuracy. However, many traditional methods are based on handcrafted features, which brings difficulties for multi-classification tasks due to spectral intra-class heterogeneity and similarity of inter-class. The deep learning algorithm, especially the convolutional neural network (CNN), has been perceived promising feature extractor and classification for processing hyperspectral remote sensing images. Although 2D CNN can extract spatial features, the specific spectral properties are not used effectively. While 3D CNN has the capability for them, but the computational burden increases as stacking layers. To address these issues, we propose a novel HSIC framework based on the residual CNN network by integrating the advantage of 2D and 3D CNN. First, 3D convolutions focus on extracting spectral features with feature recalibration and refinement by channel attention mechanism. The 2D depth-wise separable convolution approach with different size kernels concentrates on obtaining multi-scale spatial features and reducing model parameters. Furthermore, the residual structure optimizes the back-propagation for network training. The results and analysis of extensive HSIC experiments show that the proposed residual 2D-3D CNN network can effectively extract spectral and spatial features and improve classification accuracy.
33
Song, Liangliang, Zhixi Feng, Shuyuan Yang, Xinyu Zhang, and Licheng Jiao. "Self-Supervised Assisted Semi-Supervised Residual Network for Hyperspectral Image Classification." Remote Sensing 14, no. 13 (June 23, 2022): 2997. http://dx.doi.org/10.3390/rs14132997.
Повний текст джерелаАнотація:
Due to the scarcity and high cost of labeled hyperspectral image (HSI) samples, many deep learning methods driven by massive data cannot achieve the intended expectations. Semi-supervised and self-supervised algorithms have advantages in coping with this phenomenon. This paper primarily concentrates on applying self-supervised strategies to make strides in semi-supervised HSI classification. Notably, we design an effective and a unified self-supervised assisted semi-supervised residual network (SSRNet) framework for HSI classification. The SSRNet contains two branches, i.e., a semi-supervised and a self-supervised branch. The semi-supervised branch improves performance by introducing HSI data perturbation via a spectral feature shift. The self-supervised branch characterizes two auxiliary tasks, including masked bands reconstruction and spectral order forecast, to memorize the discriminative features of HSI. SSRNet can better explore unlabeled HSI samples and improve classification performance. Extensive experiments on four benchmarks datasets, including Indian Pines, Pavia University, Salinas, and Houston2013, yield an average overall classification accuracy of 81.65%, 89.38%, 93.47% and 83.93%, which sufficiently demonstrate that SSRNet can exceed expectations compared to state-of-the-art methods.
34
Kit, Eliezer, Oded Gottlieb, and Dov S. Rosen. "EVALUATION OF INCIDENT WAVE ENERGY IN FLUME TEST." Coastal Engineering Proceedings 1, no. 20 (January 29, 1986): 93. http://dx.doi.org/10.9753/icce.v20.93.
Повний текст джерелаАнотація:
A two dimensional model study, carried out for a structure in a flume using irregular waves, presents the problem of determining the relationship between the total incident wave energy attacking the structure and its response to that attack (displacements, forces, etc.) in various sea states, The total incident wave energy can be evaluated indirectly only, because the wave energy measured in the flume contains an extent of residual wave energy in addition to that generated by the wave machine. This residual energy consists of the re-reflected wave energy from the paddle of the wave machine, assuming the existence of quasi-stationary wave conditions in the flume. A method originally presented by Gravesen et al. (1974), was applied in this study to evaluate the total incident wave energy. In view of the results obtained by this method, a physically more sound refinement is proposed for the evaluation of the total incident wave energy (and characteristic wave height). Results of model tests were analyzed by the CAMERI refinement and compared with the Gravesen method and with a cross-spectral least squares method, separating incident and reflected wave spectra from wave spectra measured in the flume, Good agreement was found between the results obtained employing the CAMERI refinement and the cross-spectral least squares method, Advantages and drawbacks of these methods are indicated,
35
Kim, Seon Man. "Wearable Hearing Device Spectral Enhancement Driven by Non-Negative Sparse Coding-Based Residual Noise Reduction." Sensors 20, no. 20 (October 10, 2020): 5751. http://dx.doi.org/10.3390/s20205751.
Повний текст джерелаАнотація:
This paper proposes a novel technique to improve a spectral statistical filter for speech enhancement, to be applied in wearable hearing devices such as hearing aids. The proposed method is implemented considering a 32-channel uniform polyphase discrete Fourier transform filter bank, for which the overall algorithm processing delay is 8 ms in accordance with the hearing device requirements. The proposed speech enhancement technique, which exploits the concepts of both non-negative sparse coding (NNSC) and spectral statistical filtering, provides an online unified framework to overcome the problem of residual noise in spectral statistical filters under noisy environments. First, the spectral gain attenuator of the statistical Wiener filter is obtained using the a priori signal-to-noise ratio (SNR) estimated through a decision-directed approach. Next, the spectrum estimated using the Wiener spectral gain attenuator is decomposed by applying the NNSC technique to the target speech and residual noise components. These components are used to develop an NNSC-based Wiener spectral gain attenuator to achieve enhanced speech. The performance of the proposed NNSC–Wiener filter was evaluated through a perceptual evaluation of the speech quality scores under various noise conditions with SNRs ranging from -5 to 20 dB. The results indicated that the proposed NNSC–Wiener filter can outperform the conventional Wiener filter and NNSC-based speech enhancement methods at all SNRs.
36
Weitzel, Alexander, Claudia Samol, Peter J. Oefner, and Wolfram Gronwald. "Robust Metabolite Quantification from J-Compensated 2D 1H-13C-HSQC Experiments." Metabolites 10, no. 11 (November 7, 2020): 449. http://dx.doi.org/10.3390/metabo10110449.
Повний текст джерелаАнотація:
The spectral resolution of 2D 1H-13C heteronuclear single quantum coherence (1H-13C-HSQC) nuclear magnetic resonance (NMR) spectra facilitates both metabolite identification and quantification in nuclear magnetic resonance-based metabolomics. However, quantification is complicated by variations in magnetization transfer, which among others originate mainly from scalar coupling differences. Methods that compensate for variation in scalar coupling include the generation of calibration factors for individual signals or the use of additional pulse sequence schemes such as quantitative HSQC (Q-HSQC) that suppress the JCH-dependence by modulating the polarization transfer delays of HSQC or, additionally, employ a pure-shift homodecoupling approach in the 1H dimension, such as Quantitative, Perfected and Pure Shifted HSQC (QUIPU-HSQC). To test the quantitative accuracy of these three methods, employing a 600 MHz NMR spectrometer equipped with a helium cooled cryoprobe, a Latin-square design that covered the physiological concentration ranges of 10 metabolites was used. The results show the suitability of all three methods for the quantification of highly abundant metabolites. However, the substantially increased residual water signal observed in QUIPU-HSQC spectra impeded the quantification of low abundant metabolites located near the residual water signal, thus limiting its utility in high-throughput metabolite fingerprinting studies.
37
Kwofie, Francis, Nuwan Undugodage D. Perera, Kaushalya S. Dahal, George P. Affadu-Danful, Koichi Nishikida, and Barry K. Lavine. "Transmission Infrared Microscopy and Machine Learning Applied to the Forensic Examination of Original Automotive Paint." Applied Spectroscopy 76, no. 1 (December 17, 2021): 118–31. http://dx.doi.org/10.1177/00037028211057574.
Повний текст джерелаАнотація:
Alternate least squares (ALS) reconstructions of the infrared (IR) spectra of the individual layers from original automotive paint were analyzed using machine learning methods to improve both the accuracy and speed of a forensic automotive paint examination. Twenty-six original equipment manufacturer (OEM) paints from vehicles sold in North America between 2000 and 2006 served as a test bed to validate the ALS procedure developed in a previous study for the spectral reconstruction of each layer from IR line maps of cross-sectioned OEM paint samples. An examination of the IR spectra from an in-house library (collected with a high-pressure transmission diamond cell) and the ALS reconstructed IR spectra of the same paint samples (obtained at ambient pressure using an IR transmission microscope equipped with a BaF2 cell) showed large peak shifts (approximately 10 cm−1) with some vibrational modes in many samples comprising the cohort. These peak shifts are attributed to differences in the residual polarization of the IR beam of the transmission IR microscope and the IR spectrometer used to collect the in-house IR spectral library. To solve the problem of frequency shifts encountered with some vibrational modes, IR spectra from the in-house spectral library and the IR microscope were transformed using a correction algorithm previously developed by our laboratory to simulate ATR spectra collected on an iS-50 FT-IR spectrometer. Applying this correction algorithm to both the ALS reconstructed spectra and in-house IR library spectra, the large peak shifts previously encountered with some vibrational modes were successfully mitigated. Using machine learning methods to identify the manufacturer and the assembly plant of the vehicle from which the OEM paint sample originated, each of the twenty-six cross-sectioned automotive paint samples was correctly classified as to the “make” and model of the vehicle and was also matched to the correct paint sample in the in-house IR spectral library.
38
Amerian, Y., and B. Voosoghi. "Least Squares Spectral Analysis for Detection of Systematic Behaviour of Digital Level Compensator." Journal of Geodetic Science 1, no. 1 (March 1, 2011): 35–40. http://dx.doi.org/10.2478/v10156-010-0005-4.
Повний текст джерелаАнотація:
Least Squares Spectral Analysis for Detection of Systematic Behaviour of Digital Level CompensatorLevelling is the most precise technique for height difference measurements in geomatics engineering. Various systematic errors affect precise levelling observations and reduce the precision of the observed height differences. This study investigates digital levels residual compensator error and observational method for its elimination. For this purpose the levelling data, which was collected with Zeiss DiNi 12 digital levels, was analysed. There are different statistical and spectral methods that can reveal the presence of systematic errors in levelling results. In this study, the Least Squares Spectral Analysis (LSSA) method is used. The analysis confirmed that using alternating pointing method (BFFB, FBBF) instead of usual observation routine (BFFB) will eliminate the Zeiss DiNi 12 digital levels residual compensator error from section height differences and discrepancies. In this way, it does not matter using different instruments in the forward and backward section runs and the discrepancies can be used to investigate other systematic errors.
39
Jianqiang, Zhang, Liu Weijuan, Zhang Huaihui, Hou Ying, Yang Panpan, Li Changyu, Yang Yanmei, and Li Ming. "Automatic classification of tobacco leaves based on near infrared spectroscopy and nonnegative least squares." Journal of Near Infrared Spectroscopy 26, no. 2 (March 28, 2018): 101–5. http://dx.doi.org/10.1177/0967033518762617.
Повний текст джерелаАнотація:
A nonnegative least squares classifier was proposed in this paper to classify near infrared spectral data. The method used near infrared spectral data of training samples to make up a data dictionary of the sparse representation. By adopting the nonnegative least squares sparse coding algorithm, the near infrared spectral data of test samples would be expressed via the sparsest linear combinations of the dictionary. The regression residual of the test sample of each class was computed, and finally it was assigned to the class with the minimum residual. The method was compared with the other classifying approaches, including the well-performing principal component analysis–linear discriminant analysis and principal component analysis–particle swarm optimization–support vector machine. Experimental results showed that the approach was faster and generally achieved a better prediction performance over compared methods. The method can accurately recognize different classes of tobacco leaves and it provides a new technology for quality evaluation of tobacco leaf in its purchasing activities.
40
Kiessling, Jonas, and Filip Thor. "A Computable Definition of the Spectral Bias." Proceedings of the AAAI Conference on Artificial Intelligence 36, no. 7 (June 28, 2022): 7168–75. http://dx.doi.org/10.1609/aaai.v36i7.20677.
Повний текст джерелаАнотація:
Neural networks have a bias towards low frequency functions. This spectral bias has been the subject of several previous studies, both empirical and theoretical. Here we present a computable definition of the spectral bias based on a decomposition of the reconstruction error into a low and a high frequency component. The distinction between low and high frequencies is made in a way that allows for easy interpretation of the spectral bias. Furthermore, we present two methods for estimating the spectral bias. Method 1 relies on the use of the discrete Fourier transform to explicitly estimate the Fourier spectrum of the prediction residual, and Method 2 uses convolution to extract the low frequency components, where the convolution integral is estimated by Monte Carlo methods. The spectral bias depends on the distribution of the data, which is approximated with kernel density estimation when unknown. We devise a set of numerical experiments that confirm that low frequencies are learned first, a behavior quantified by our definition.
41
Li, Jiaojiao, Chaoxiong Wu, Rui Song, Yunsong Li, and Weiying Xie. "Residual Augmented Attentional U-Shaped Network for Spectral Reconstruction from RGB Images." Remote Sensing 13, no. 1 (December 31, 2020): 115. http://dx.doi.org/10.3390/rs13010115.
Повний текст джерелаАнотація:
Deep convolutional neural networks (CNNs) have been successfully applied to spectral reconstruction (SR) and acquired superior performance. Nevertheless, the existing CNN-based SR approaches integrate hierarchical features from different layers indiscriminately, lacking an investigation of the relationships of intermediate feature maps, which limits the learning power of CNNs. To tackle this problem, we propose a deep residual augmented attentional u-shape network (RA2UN) with several double improved residual blocks (DIRB) instead of paired plain convolutional units. Specifically, a trainable spatial augmented attention (SAA) module is developed to bridge the encoder and decoder to emphasize the features in the informative regions. Furthermore, we present a novel channel augmented attention (CAA) module embedded in the DIRB to rescale adaptively and enhance residual learning by using first-order and second-order statistics for stronger feature representations. Finally, a boundary-aware constraint is employed to focus on the salient edge information and recover more accurate high-frequency details. Experimental results on four benchmark datasets demonstrate that the proposed RA2UN network outperforms the state-of-the-art SR methods under quantitative measurements and perceptual comparison.
42
Petrik, Mikhail, Arkady Chikrii, and Ivan Mudrik. "SIMULATION AND PARAMETERS-IDENTIFICATION METHODS OF HETEROGENEOUS ABNORMAL NEUROLOGICAL MOVEMENTS IN MULTICOMPONENT NEURO-BIOSYSTEMS WITH COGNITIVE FEEDBACK." Journal of Automation and Information sciences 3 (May 1, 2021): 18–33. http://dx.doi.org/10.34229/1028-0979-2021-3-2.
Повний текст джерелаАнотація:
The foundations of mathematical modeling and identification of parameters of heterogeneous abnormal neurological movements (ANM) in multicomponent neuro-biosystems with cognitive feedback have been developed. Based on the methods of integral transformations and spectral analysis developed by the authors for heterogeneous media, a new approach to the construction of hybrid models of wave signal propagation is proposed that describes unwanted tremors of the patient's arm (T-object) as a result of an unconstrained contraction of skeletal muscles due to the cognitive effects of a certain group of neural nodes in the cortex cerebral (CC). A hybrid model of a neuro-biosystem is developed, which describes the state and behavior, namely, the segment-by-segment description of 3D elements of the ANM trajectories of the T-object, taking into account the matrix of cognitive influences of the groups of neuro nodes of the CC. On the basis of hybrid integral Fourier transforms a high-speed analytical vector solution of the model is obtained, which describes the elements of the trajectories on each AND-segment. A new method for calculating of hybrid spectral function, spectral values and matrix of cognitive influences of CC neuronodes is proposed, which determine hybrid integral transformation of solution construction. New non-classical problems of multi-parameter identification of neuro-feedback systems in heterogeneous media based on minimization of the residual functional between observation trajectories and their model analogs are formulated and solved. High-performance algorithms of the amplitude-frequency characteristics identifying of a feedback-system in analytical expressions for the gradients of the residual functional have been constructed, which allow parallel-computations on multicore computers. Computer modeling and identification of ANM trajectories of the studied neuro-feedback-system have been performed.
43
Flasseur, Olivier, Loïc Denis, Éric Thiébaut, and Maud Langlois. "PACO ASDI: an algorithm for exoplanet detection and characterization in direct imaging with integral field spectrographs." Astronomy & Astrophysics 637 (May 2020): A9. http://dx.doi.org/10.1051/0004-6361/201937239.
Повний текст джерелаАнотація:
Context. Exoplanet detection and characterization by direct imaging both rely on sophisticated instruments (adaptive optics and coronagraph) and adequate data processing methods. Angular and spectral differential imaging (ASDI) combines observations at different times and a range of wavelengths in order to separate the residual signal from the host star and the signal of interest corresponding to off-axis sources. Aims. Very high contrast detection is only possible with an accurate modeling of those two components, in particular of the background due to stellar leakages of the host star masked out by the coronagraph. Beyond the detection of point-like sources in the field of view, it is also essential to characterize the detection in terms of statistical significance and astrometry and to estimate the source spectrum. Methods. We extend our recent method PACO, based on local learning of patch covariances, in order to capture the spectral and temporal fluctuations of background structures. From this statistical modeling, we build a detection algorithm and a spectrum estimation method: PACO ASDI. The modeling of spectral correlations proves useful both in reducing detection artifacts and obtaining accurate statistical guarantees (detection thresholds and photometry confidence intervals). Results. An analysis of several ASDI datasets from the VLT/SPHERE-IFS instrument shows that PACO ASDI produces very clean detection maps, for which setting a detection threshold is statistically reliable. Compared to other algorithms used routinely to exploit the scientific results of SPHERE-IFS, sensitivity is improved and many false detections can be avoided. Spectrally smoothed spectra are also produced by PACO ASDI. The analysis of datasets with injected fake planets validates the recovered spectra and the computed confidence intervals. Conclusions. PACO ASDI is a high-contrast processing algorithm accounting for the spatio-spectral correlations of the data to produce statistically-grounded detection maps and reliable spectral estimations. Point source detections, photometric and astrometric characterizations are fully automatized.
44
Ma, Mingming, Yi Niu, Chang Liu, Fu Li, and Guangming Shi. "A Lightweight Multi-Level Information Network for Multispectral and Hyperspectral Image Fusion." Remote Sensing 14, no. 21 (November 6, 2022): 5600. http://dx.doi.org/10.3390/rs14215600.
Повний текст джерелаАнотація:
The process of fusing the rich spectral information of a low spatial resolution hyperspectral image (LR-HSI) with the spatial information of a high spatial resolution multispectral image (HR-MSI) to obtain an HSI with the spatial resolution of an MSI image is called hyperspectral image fusion (HIF). To reconstruct hyperspectral images at video frame rate, we propose a lightweight multi-level information network (MINet) for multispectral and hyperspectral image fusion. Specifically, we develop a novel lightweight feature fusion model, namely residual constraint block based on global variance fine-tuning (GVF-RCB), to complete the feature extraction and fusion of hyperspectral images. Further, we define a residual activity factor to judge the learning ability of the residual module, thereby verifying the effectiveness of GVF-RCB. In addition, we use cascade cross-level fusion to embed the different spectral bands of the upsampled LR-HSI in a progressive manner to compensate for lost spectral information at different levels and to maintain spatial high frequency information at all times. Experiments on different datasets show that our MINet outperforms the state-of-the-art methods in terms of objective metrics, in particular by requiring only 30% of the running time and 20% of the number of parameters.
45
Xue, Yiming, Dan Zeng, Fansheng Chen, Yueming Wang, and Zhijiang Zhang. "A New Dataset and Deep Residual Spectral Spatial Network for Hyperspectral Image Classification." Symmetry 12, no. 4 (April 5, 2020): 561. http://dx.doi.org/10.3390/sym12040561.
Повний текст джерелаАнотація:
Due to the limited varieties and sizes of existing public hyperspectral image (HSI) datasets, the classification accuracies are higher than 99% with convolutional neural networks (CNNs). In this paper, we presented a new HSI dataset named Shandong Feicheng, whose size and pixel quantity are much larger. It also has a larger intra-class variance and a smaller inter-class variance. State-of-the-art methods were compared on it to verify its diversity. Otherwise, to reduce overfitting caused by the imbalance between high dimension and small quantity of labeled HSI data, existing CNNs for HSI classification are relatively shallow and suffer from low capacity of feature learning. To solve this problem, we proposed an HSI classification framework named deep residual spectral spatial setwork (DRSSN). By using shortcut connection structure, which is an asymmetry structure, DRSSN can be deeper to extract features with better discrimination. In addition, to alleviate insufficient training caused by unbalanced sample sizes between easily and hard classified samples, we proposed a novel training loss function named sample balanced loss, which allocated weights to the losses of samples according to their prediction confidence. Experimental results on two popular datasets and our proposed dataset showed that our proposed network could provide competitive results compared with state-of-the-art methods.
46
Li, Xiangli, and Xiao Guo. "Spectral residual methods with two new non-monotone line searches for large-scale nonlinear systems of equations." Applied Mathematics and Computation 269 (October 2015): 59–69. http://dx.doi.org/10.1016/j.amc.2015.07.079.
Повний текст джерела
47
Hou, Z., Y. Chen, K. Tan, and P. Du. "NOVEL HYPERSPECTRAL ANOMALY DETECTION METHODS BASED ON UNSUPERVISED NEAREST REGULARIZED SUBSPACE." ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-3 (April 30, 2018): 539–46. http://dx.doi.org/10.5194/isprs-archives-xlii-3-539-2018.
Повний текст джерелаАнотація:
Anomaly detection has been of great interest in hyperspectral imagery analysis. Most conventional anomaly detectors merely take advantage of spectral and spatial information within neighboring pixels. In this paper, two methods of Unsupervised Nearest Regularized Subspace-based with Outlier Removal Anomaly Detector (UNRSORAD) and Local Summation UNRSORAD (LSUNRSORAD) are proposed, which are based on the concept that each pixel in background can be approximately represented by its spatial neighborhoods, while anomalies cannot. Using a dual window, an approximation of each testing pixel is a representation of surrounding data via a linear combination. The existence of outliers in the dual window will affect detection accuracy. Proposed detectors remove outlier pixels that are significantly different from majority of pixels. In order to make full use of various local spatial distributions information with the neighboring pixels of the pixels under test, we take the local summation dual-window sliding strategy. The residual image is constituted by subtracting the predicted background from the original hyperspectral imagery, and anomalies can be detected in the residual image. Experimental results show that the proposed methods have greatly improved the detection accuracy compared with other traditional detection method.
48
Ding, D. Z., G. M. Li, Y. Y. An, and R. S. Chen. "Application of Hierarchical Two-Level Spectral Preconditioning Method for Electromagnetic Scattering from the Rough Surface." International Journal of Antennas and Propagation 2014 (2014): 1–10. http://dx.doi.org/10.1155/2014/752418.
Повний текст джерелаАнотація:
The higher-order hierarchical Legendre basis functions combining the electrical field integral equations (EFIE) are developed to solve the scattering problems from the rough surface. The hierarchical two-level spectral preconditioning method is developed for the generalized minimal residual iterative method (GMRES). The hierarchical two-level spectral preconditioner is constructed by combining the spectral preconditioner and sparse approximate inverse (SAI) preconditioner to speed up the convergence rate of iterative methods. The multilevel fast multipole method (MLFMM) is employed to reduce memory requirement and computational complexity of the method of moments (MoM) solution. The accuracy and efficiency are confirmed with a couple of numerical examples.
49
Bujanovic, Biljana, Sally A. Ralph, Richard S. Reiner, and Rajai H. Atalla. "Lignin modification in the initial phase of softwood kraft pulp delignification with polyoxometalates (POMs)." Holzforschung 61, no. 5 (August 1, 2007): 492–98. http://dx.doi.org/10.1515/hf.2007.102.
Повний текст джерелаАнотація:
Abstract Commercial softwood kraft pulp with kappa number 30.5 (KP30.5) was delignified with polyoxometalates (POM, Na5(+2)[SiV1(-0.1)MoW10(+0.1)O40]), and POM-treated kraft pulp of kappa number 23.6 was obtained (KPPOM,23.6). Residual lignin from pulps was isolated by mild acid hydrolysis and characterized by analytical and spectral methods to gain insight into lignin reactions taking place during the initial delignification phase. Lignin from POM-delignified pulp was isolated in lower yield. Comparative analysis of residual lignins (RL-KP30.5, RL-KPPOM,23.6) showed that POM leads to products enriched in carbonyl/carboxyl groups and carbohydrates. POM lignins have a lower molecular mass and a lower content of phenolic hydroxyl and methoxyl groups. Based on these results and FTIR spectra, we suggest that aromatic ring cleavage and quinone formation occur during POM delignification. The degree of lignin-cellulose association increases after POM delignification. Lignin-cellulose association was found to be partially unstable under mild alkaline conditions, as residual lignin isolated after alkaline extraction of KPPOM,23.6 pulp (RL-KPPOM/NaOH) exhibited lower glucose content, higher Klason lignin content, and less extraneous material.
50
Бельков, М. В., Д. А. Борисевич, К. Ю. Кацалап та М. А. Ходасевич. "Выбор спектральных переменных в многопараметрической калибровке концентраций C, Mn, Si, Cr, Ni и Cu в низколегированных сталях методами лазерно-искровой эмиссионной спектроскопии". Оптика и спектроскопия 130, № 10 (2022): 1611. http://dx.doi.org/10.21883/os.2022.10.53634.3895-22.
Повний текст джерелаАнотація:
Multivariate calibrations of concentrations of C, Mn, Si, Cr, Ni and Cu have been developed by the partial least squares method for 31 to 39 standard samples of low-alloy steels using low-resolution emission spectra (190-440 nm, resolution 0.4 nm, step 0.1 nm). Three methods of spectral variables selection are considered: a method of ranking spectral variables by their correlation coefficient with the value of the calibrated parameter, a successive projection algorithm and an original modification of searching combination moving windows. The partial least squares model with the spectral variables selection by the method of the searching combination moving windows for C is quantitative: the root mean square error is 0.004%, the residual predictive deviation in the test dataset is 23.4 in the concentration range 0.13 to 0.43 %. Calibrations of Mn (0.04% and 5.2 in the range of 0.47–1.15%), Si (0.003% and 20.7 in the range of 0.15–0.33%), Cr (0.04% and 3.1 in the range of 0.09–0.43%) and Ni (0.01% and 4.8 in the range of 0.05–0.25%) are also quantitative. For Cu in the concentration range of 0.06–0.26%, calibration is qualitative (0.04% and 1.4).