- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR) - Chengdu, China (2018.7.15-2018.7.18)] 2018 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR) - Spectral-Spatial Graph Convolutional Networks for Semel-Supervised Hyperspectral Image Classification
摘要： Collecting label samples is quite costly and time consuming for hyperspectral image (HSI) classification tasks. Semi-supervised learning framework, which combines the intrinsic information of labeled and unlabeled samples can alleviate the deficient labeled samples and increase the accuracy of HSI classification. In this paper, we propose a novel framework for semi-supervised learning on multiple spectral-spatial graphs that is based on graph convolutional networks (SGCN). In the filtering operation on graphs we consider the spatial information and spectral signatures of HSI simultaneously. The experimental results on three real-life HSI data sets, i.e. Botswana Hyperion, Kennedy Space Center, and Indian Pines, show that the proposed SGCN can significantly improve the classification accuracy. For instance, the over accuracy on Indian Pine data is increased from 78% to 93%.
关键词： Hyperspectral image classification,Graph fourier transform,Graph convolutional,Neural networks,Semi-supervised learning
更新于2025-09-23 15:21:01
-
FPGA Implementation of Underwater Image Enhancement using Nonlinear Filtering
摘要： Background/Motivation: Statistical Analysis: they suffer from various adversary underwater conditions and require low power devices such as FPGA. Methods/ Optical image captured by autonomous underwater vehicles need to be preprocessed as Underwater image degradation due to non uniform illumination is corrected by non linear filtering and is implemented in FPGA. Two dimensional Fast Fourier Transform (FFT) and its inverse along with logarithmic computation were implemented in FPGA as a part of filter implementation. One dimensional row wise and column wise FPGA’s can be used for implementing FFT were computed first and then combined to form two dimensional FFT. architectures that require high level parallelism such as image processing algorithms due to its inherent parallelism. Frequency based filtering method which is employed for preprocessing an underwater image successfully provided good histogram compared to the original one. Device utilization for implementing structure was estimated. Novel method of FFT implementation, its inverse and logarithmic computation was used in this work. The method can be applied to digital signal processing applications that can be implemented in FPGA. Automation of image preprocessing is required in unmanned underwater vehicles used for cable detection, navigation etc. These vehicles are to be provided with low power devices such as FPGAs for long run in underwater applications.
关键词： Nonlinear Filtering,Fast Fourier Transform,FPGA,Automated Underwater Vehicles,Underwater Image Enhancement
更新于2025-09-23 15:21:01
-
Detecting volatile compounds in food by open-path Fourier-transform infrared spectroscopy
摘要： We previously found that the brand of a food and spoilage of the food can be identified from the infrared spectra of the volatile compounds released. However, this required pumping the volatile compounds into a gas cell, meaning measurements over large areas could not be made. Gas components can be quantified from a distance of a few metres or kilometres by open-path Fourier-transform infrared spectroscopy (FTIR), and the spatial distributions of gas clouds can even be determined using open-path FTIR and an imaging detection method. In the study described here, we used open-path FTIR to remotely detect volatile compounds in food. Active and passive methods were used to obtain infrared spectra of volatile compounds released from spirits, vinegars, and grapes from a distance of 5 m. The absorption characteristics of ethanol, esters, and unknown volatile compounds were clearly found in the spectra. The brands of the spirits and degree to which the grapes had spoiled were identified by compensating for ethanol in the atmosphere and chemometrics. The results indicate that open-path FTIR can be used to remotely detect volatile compounds released by food and may be able to be used to identify spoiling food in large food warehouses.
关键词： remote detection,volatile compounds,open-path Fourier-transform infrared spectroscopy,food,spoilage
更新于2025-09-23 15:21:01
-
[IEEE 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Chicago, IL, USA (2019.6.16-2019.6.21)] 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) - Stability of Tin-Lead Halide Perovskite Solar Cells
摘要： Envelope representations such as the auditory or traditional spectrogram can be defined by the set of envelopes from the outputs of a filterbank. Common envelope extraction methods discard information regarding the fast fluctuations, or phase, of the signal. Thus, it is difficult to invert, or reconstruct a time-domain signal from, an arbitrary envelope representation. To address this problem, a general optimization approach in the time domain is proposed here, which iteratively minimizes the distance between a target envelope representation and that of a reconstructed time-domain signal. Two implementations of this framework are presented for auditory spectrograms, where the filterbank is based on the behavior of the basilar membrane and envelope extraction is modeled on the response of inner hair cells. One implementation is direct while the other is a two-stage approach that is computationally simpler. While both can accurately invert an auditory spectrogram, the two-stage approach performs better on time-domain metrics. The same framework is applied to traditional spectrograms based on the magnitude of the short-time Fourier transform. Inspired by human perception of loudness, a modification to the framework is proposed, which leads to a more accurate inversion of traditional spectrograms.
关键词： short-time Fourier transform,auditory spectrogram,Spectrogram inversion,gradient methods
更新于2025-09-23 15:19:57
-
Physicala??Chemical Properties of Self-Assembled Structures in Solution of Zinc Phthalocyanine and Bis-3-pentyl-PTCDI Derivative
摘要： For the first time, in this work, we succeed in synthesizing in solution a novel supramolecular self-assembled zinc phthalocyanine (ZnPc) and N,N′-bis(3-pentyl)-perylene-3,4,9,10-bis(dicarboximide) (bis-3-pentyl-PTCDI) system with improved light absorption and phosphorescence lifetime of the charge separated states up to 2.5 ms. Moreover, the structural and optical properties of undoped and doped with iodine ZnPc thin films were investigated by X-ray diffraction, X-ray photoelectron spectroscopy, Fourier transform infrared spectroscopy, Raman analysis, and UV?vis spectroscopy. Analysis of ZnPc:I2:bis-3-pentyl-PTCDI blend in a 2:1 ratio shows that sandwich complexes between them result in improved bulk properties as compared to those of the single-component systems.
关键词： light absorption,Fourier transform infrared spectroscopy,UV?vis spectroscopy,bis-3-pentyl-PTCDI,X-ray diffraction,supramolecular self-assembly,X-ray photoelectron spectroscopy,zinc phthalocyanine,phosphorescence lifetime,Raman analysis
更新于2025-09-23 15:19:57
-
Detecting Laser-Volatilized Salts with a Miniature 100-GHz Spectrometer
摘要： Rotational transitions are unique identi?ers of molecular species, including isotopologues. This article describes the rotational detections of two laser-volatilized salts, NaCl and KCl, made with a miniature Fourier transform millimeter-wave (FTmmW) cavity spectrometer that could one day be used to measure solid composition in the ?eld or in space. The two salts are relevant targets for icy moons in the outer solar system, and in principle, other molecular solids could be analyzed with the FTmmW instrument. By coupling the spectrometer to a collisionally cooling laser ablation source, (a) we demonstrate that the FTmmW instrument is sensitive enough to detect ablation products, and (b) we use the small size of the FTmmW cavity to measure ablation product signal along the carrier gas beam. We ?nd that for 532 nm nanosecond pulses, ablated molecules are widely dispersed in the carrier-gas jet. In addition to the miniature spectrometer results, we present several complementary measurements intended to characterize the laser ablation process. For pulse energies between 10 and 30 mJ, the ablation product yield increases linearly, reaching approximately 1012 salt molecules per 30 mJ pulse. Using mass spectrometry, we observe Li+, Na+, and K+ in the plumes of ablated NaCl, KCl, and LiCl, which implies dissociation of the volatilized material. We do not observe salt ions (e.g., NaCl+). However, with 800 nm femtosecond laser pulses, the triatomic ion clusters Li2Cl+, Na2Cl+, and K2Cl+ are produced. Finally, we observe incomplete volatilization with the nanosecond pulses: some of the ejecta are liquid droplets. The insights about ablation plume physics gleaned from these experiments should guide future implementations of the laser-volatilization technique.
关键词： KCl,Fourier transform millimeter-wave,FTmmW,rotational transitions,NaCl,mass spectrometry,outer solar system,icy moons,laser-volatilized salts,laser ablation
更新于2025-09-23 15:19:57
-
[IEEE TENCON 2019 - 2019 IEEE Region 10 Conference (TENCON) - Kochi, India (2019.10.17-2019.10.20)] TENCON 2019 - 2019 IEEE Region 10 Conference (TENCON) - Low-Frequency Effectiveness of Differential Detection for Electro-optic Sensor System
摘要： This paper describes the effectiveness of differential detection for reducing laser noise in the low-frequency region in an electro-optic (EO) sensor system. Improvement in the signal-to-noise ratio (SNR) is necessary for expanding the applications of EO sensor systems. In a previous work, an alternating current coupling was used at an optical receiver to eliminate a direct current (DC) signal. As a result, the low-frequency effectiveness of the differential detection could not be estimated. Therefore, in this study, we modified a photodiode bias circuit for DC coupling measurement.
关键词： phasor diagram,differential detection,fast Fourier transform,signal-to-ratio,electro-optic sensor
更新于2025-09-23 15:19:57
-
[IEEE 2019 6th International Conference on Systems and Informatics (ICSAI) - Shanghai, China (2019.11.2-2019.11.4)] 2019 6th International Conference on Systems and Informatics (ICSAI) - Laser Strip Center Extraction Methodology for the Detection of Weld Seam
摘要： Envelope representations such as the auditory or traditional spectrogram can be defined by the set of envelopes from the outputs of a filterbank. Common envelope extraction methods discard information regarding the fast fluctuations, or phase, of the signal. Thus, it is difficult to invert, or reconstruct a time-domain signal from, an arbitrary envelope representation. To address this problem, a general optimization approach in the time domain is proposed here, which iteratively minimizes the distance between a target envelope representation and that of a reconstructed time-domain signal. Two implementations of this framework are presented for auditory spectrograms, where the filterbank is based on the behavior of the basilar membrane and envelope extraction is modeled on the response of inner hair cells. One implementation is direct while the other is a two-stage approach that is computationally simpler. While both can accurately invert an auditory spectrogram, the two-stage approach performs better on time-domain metrics. The same framework is applied to traditional spectrograms based on the magnitude of the short-time Fourier transform. Inspired by human perception of loudness, a modification to the framework is proposed, which leads to a more accurate inversion of traditional spectrograms.
关键词： Spectrogram inversion,gradient methods,short-time Fourier transform,auditory spectrogram
更新于2025-09-23 15:19:57
-
Blood-based FTIR-ATR spectroscopy coupled with extreme gradient boosting for the diagnosis of type 2 diabetes
摘要： Timely diagnosis of type 2 diabetes and early intervention and treatment of it are important for controlling metabolic disorders, delaying and reducing complications, reducing mortality, and improving quality of life. Type 2 diabetes was diagnosed by Fourier transform mid-infrared (FTIR) attenuated total reflection (ATR) spectroscopy in combination with extreme gradient boosting (XGBoost). Whole blood FTIR-ATR spectra of 51 clinically diagnosed type 2 diabetes and 55 healthy volunteers were collected. For the complex composition of whole blood and much spectral noise, Savitzky–Golay smoothing was first applied to the FTIR-ATR spectrum. Then PCA was used to eliminate redundant data and got the best number of principle components. Finally, the XGBoost algorithm was used to discriminate the type 2 diabetes from healthy volunteers and the grid search algorithm was used to optimize the relevant parameters of the XGBoost model to improve the robustness and generalization ability of the model. The sensitivity of the optimal XGBoost model was 95.23% (20/21), the specificity was 96.00% (24/25), and the accuracy was 95.65% (44/46). The experimental results show that FTIR-ATR spectroscopy combined with XGBoost algorithm can diagnose type 2 diabetes quickly and accurately without reagents.
关键词： Fourier transform mid-infrared attenuated total reflection spectroscopy,whole blood,extreme gradient boosting,type 2 diabetes
更新于2025-09-23 15:19:57
-
Quantification of calcium in infant formula using laser-induced breakdown spectroscopy (LIBS), Fourier transform mid-infrared (FT-IR) and Raman spectroscopy combined with chemometrics including data fusion
摘要： Laser-induced breakdown spectroscopy (LIBS), Fourier transform mid-infrared (FT-IR) and Raman spectroscopy combined with chemometrics were investigated to quantify calcium (Ca) content in infant formula powder (INF). INF samples (n=51) with calcium content levels (ca. 6.5-30 mg Ca/100kJ) were prepared in accordance with the guidelines of Commission Directive 2006/125/EC. Atomic absorption spectroscopy (AAS) was used as the reference method for Ca content determination. To predict Ca content in INF samples, partial least squares regression (PLSR) models that developed based on LIBS, Raman and FT-IR spectral data, respectively. The model developed using LIBS data achieved the best performance for the quantification of Ca content in INF (R2 (cross-validation (CV))-0.99, RMSECV-0.29 mg/g; R2 (prediction (P))-1, RMSEP-0.63 mg/g). PLSR models that developed based on data fusion of Raman and FT-IR spectral features obtained the second best performance (R2CV-0.97, RMSECV-0.38 mg/g; R2P-0.97, RMSEP-0.36 mg/g). This study demonstrated the potential of LIBS, FT-IR and Raman spectroscopy to accurately quantify Ca content in INF.
关键词： Fourier transform mid-infrared,laser-induced breakdown spectroscopy,chemometrics,Raman spectroscopy,calcium,infant formula
更新于2025-09-23 15:19:57