Combination Of Spectral Features Research Articles

Thermal infrared spectrum variation and waveband features of loaded rocks with different mineral components during the fracture evolution process

Previous studies focused mainly on temporal-spatial analyses of thermal infrared radiant temperature variations based on statistical parameters for rocks. These spectral analyses indicated that the thermal infrared radiance variation during the rock loading and fracturing process was a combined effect of the changes in both temperature and emissivity variation. The spectral features and mechanisms responsible for the thermal infrared spectral radiance changes in the fracturing stage and the relationship to the lithology remain poorly understood. In this paper, three types of typical rocks on the earth ground surface with different mineral components, i.e., sandstone, granite, and marble, were selected to investigate the thermal infrared spectrum (8.0–13.0 µm) waveband regulation during the loading process in the outdoor condition. The experiment yielded the following results: (1) The advantageous waveband corresponded to the reststrahlen feature (RF) waveband of the static emissivity spectrum, which was in the 8.2–9.4 μm, 8.4–10.3 μm and 11.1–11.4 μm range for sandstone, granite, and marble, respectively. There is a significant influence on the thermal infrared radiance variation caused by the waveband range. (2) The emissivity changes during the fracturing stage were the main factors causing the spectral radiance change, with opposite waveband features of the temperature change leading to an overall increase in amplitude. (3) The combination of spectral features in the advantageous waveband was more conducive to the analysis of the temporal-spatial-strength evolution of the thermal infrared radiation (TIR) temperature field on the loading rock surface. These results provide a scientific basis for thermal infrared remote sensing satellite loading design and waveband optimization and combination configuration.

Improving the monitoring of root zone soil salinity under vegetation cover conditions by combining canopy spectral information and crop growth parameters.

Soil salinization is one of the main causes of land degradation in arid and semi-arid areas. Timely and accurate monitoring of soil salinity in different areas is a prerequisite for amelioration. Hyperspectral technology has been widely used in soil salinity monitoring due to its high efficiency and rapidity. However, vegetation cover is an inevitable interference in the direct acquisition of soil spectra during crop growth period, which greatly limits the monitoring of soil salinity by remote sensing. Due to high soil salinity could lead to difficulty in plants' water absorption, and inhibit plant dry matter accumulation, a method for monitoring root zone soil salinity by combining vegetation canopy spectral information and crop aboveground growth parameters was proposed in this study. The canopy spectral information was acquired by a spectroradiometer, and then variable importance in projection (VIP), competitive adaptive reweighted sampling (CARS), and random frog algorithm (RFA) were used to extract the salinity spectral features in cotton canopy spectrum. The extracted features were then used to estimate root zone soil salinity in cotton field by combining with cotton plant height, aboveground biomass, and shoot water content. The results showed that there was a negative correlation between plant height/aboveground biomass/shoot water content and soil salinity in 0-20, 0-40, and 0-60 cm soil layers at different growth stages of cotton. Spectral feature selection by the three methods all improved the prediction accuracy of soil salinity, especially CARS. The prediction accuracy based on the combination of spectral features and cotton growth parameters was significantly higher than that based on only spectral features, with R2 increasing by 10.01%, 18.35%, and 29.90% for the 0-20, 0-40, and 0-60 cm soil layer, respectively. The model constructed based on the first derivative spectral preprocessing, spectral feature selection by CARS, cotton plant height, and shoot water content had the highest accuracy for each soil layer, with R2 of 0.715,0.769, and 0.742 for the 0-20, 0-40, 0-60 cm soil layer, respectively. Therefore, the method by combining cotton canopy hyperspectral data and plant growth parameters could significantly improve the prediction accuracy of root zone soil salinity under vegetation cover conditions. This is of great significance for the amelioration of saline soil in salinized farmlands arid areas.

A Mispronunciation Detection Method of Confusing Vowel Pair for Chinese Students

This paper presents a discriminative features based mispronunciation detection method for confusing vowel pair /i/ vs /I/ that are frequently mispronounced by Chinese learners of English. Firstly, the mean of the 39-dimensional Mel Frequency Cepstral Coefficients (MFCC) feature vector over all the frames of the current phoneme segment is employed as features to characterize the phoneme. Secondly, many specific acoustic features that can effectively capture the crucial properties of the long and short vowels are extracted. Finally, the Support Vector Machine (SVM) classifier is used for discrimination between confusing vowels /i/ and /I/ by using the discriminative features extracted from each phoneme. The experimental results show that, the proposed method can produce higher accuracy than the traditional Automatic Speech Recognition (ASR) based methods. In addition, the combination of spectral features with specific acoustic features can achieve better performance than using individual features.

Identification of Wheat Yellow Rust Using Spectral and Texture Features of Hyperspectral Images

Wheat yellow rust is one of the most destructive diseases in wheat production and significantly affects wheat quality and yield. Accurate and non-destructive identification of yellow rust is critical to wheat production management. Hyperspectral imaging technology has proven to be effective in identifying plant diseases. We investigated the feasibility of identifying yellow rust on wheat leaves using spectral features and textural features (TFs) of hyperspectral images. First, the hyperspectral images were preprocessed, and healthy and yellow rust-infected samples were obtained by creating regions of interest. Second, the extraction of spectral reflectance characteristics and vegetation indices (VIs) were performed from the preprocessed hyperspectral images, and the TFs were extracted using the grey-level co-occurrence matrix from the images transformed by principal component analysis. Third, the successive projections algorithm was employed to choose the optimum wavebands (OWs), and correlation-based feature selection was employed to select the optimal VIs and TFs (those most sensitive to yellow rust and having minimal redundancy between features). Finally, identification models of wheat yellow rust were established using a support vector machine and different features. Six OWs (538, 598, 689, 702, 751, and 895 nm), four VIs (nitrogen reflectance index, photochemical reflectance index, greenness index, and anthocyanin reflectance index), and four TFs (correlation 1, correlation 2, entropy 2, and second moment 3) were selected. The identification models based on the OWs, VIs, and TFs provided overall accuracies of 83.3%, 89.5%, and 86.5%, respectively. The TF results were especially encouraging. The models with the combination of spectral features and TFs exhibited better performance than those using the spectral features or TFs alone. The accuracies of the models with the combined features (OWs and TFs, Vis, and TFs) were 90.6% and 95.8%, respectively. These values were 7.3% and 6.3% higher, respectively, than those of the models using only the OWs or VIs. The model with the combined feature (VIs and TFs) had the highest accuracy (95.8%) and was used to map the yellow rust lesions on wheat leaves with different damage levels. The results showed that the yellow rust lesions on the leaves could be identified accurately. Overall, the combination of spectral features and TFs of hyperspectral images significantly improved the identification accuracy of wheat yellow rust.

Spatial and spectral remote sensing features to detect deforestation in Brazilian Savannas

The Brazilian Savannas have been under increasing anthropic pressure for many years, and land-use/land-cover changes (LULCC) have been largely neglected. Remote sensing provides useful tools to detect changes, but previous studies have not attempted to separate the effects of phenology from deforestation, clearing or fires to improve the accuracy of change detection without a dense time series. The scientific questions addressed in this study were: how well can we differentiate seasonal changes from deforestation processes combining the spatial and spectral information of bi-temporal (normalized difference vegetation index) NDVI images? Which feature best contribute to increase the separability on classification assessment? We applied an object-based remote sensing method that is able to separate seasonal changes due to phenology effects from LULCC by combining spectral and the spatial context using traditional spectral features and semivariogram indices, exploring the full capability of NDVI image difference to train random forest (RF) algorithm. We found that the spatial variability of NDVI values is not affect by vegetation seasonality and, therefore, the combination of spectral features and semivariogram indices provided high global accuracy (97.73%) to separate seasonal changes and deforestation or fires. From the total of 13 features, 6 provided the best combination to increase the separability on classification assessment (4 spatial and 2 spectral features). How to accurately extract LULCC while disregarding the ones caused by phenological differences in Brazilian seasonal biomes undergoing rapid land-cover changes can be achieved by adding semivariogram indices in combination with spectral features as input data to train RF algorithm.

Bagged support vector machines for emotion recognition from speech

Speech emotion recognition, a highly promising and exciting problem in the field of Human Computer Interaction, has been studied and analyzed over several decades. It concerns the task of recognizing a speaker’s emotions from their speech recordings. Recognizing emotions from speech can go a long way in determining a person’s physical and psychological state of well-being. In this work we performed emotion classification on three corpora — the Berlin EmoDB, the Indian Institute of Technology Kharagpur Simulated Emotion Hindi Speech Corpus (IITKGP-SEHSC), and the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS). A combination of spectral features was extracted from them which was further processed and reduced to the required feature set. Ensemble learning has been proven to give superior performance compared to single estimators. We propose a bagged ensemble comprising of support vector machines with a Gaussian kernel as a viable algorithm for the problem at hand. We report the results obtained on the three datasets mentioned above.

Exploring the addition of Landsat 8 thermal band in land-cover mapping

ABSTRACTOne focus of remote-sensing studies is obtaining highly accurate land-cover maps, which is essential for quantifying and monitoring changes in the environment. However, thermal data, which can provide auxiliary information, is often ignored in land-cover classification. In this study we compare the performance of different remote-sensing feature combinations with and without the Landsat 8 thermal band (band 10). The results show that overall the thermal feature had a positive effect on mapping land cover. A combination of spectral features, indices and the thermal feature maximized the improvement in accuracy. The proposed classifier was applied to map land cover in an area in Egypt. The thermal feature significantly reduced the confusion between cropland and wetland. The improvement in accuracy obtained by adding the thermal feature was analysed in a time series spanning 1 year. We found that the thermal feature improved the classification accuracy when temperature variations occurred among the different land-cover types. The effect of the thermal feature was also influenced by the land cover; in cloudless conditions, warmer weather can enhance the accuracy improvement of the thermal feature.

Fourier-transform-infrared-spectroscopy based metabolomic spectral biomarker selection towards optimal diagnostic differentiation of diabetes with and without retinopathy

The purpose of this study is to identify Fourier transform infrared (FTIR) spectroscopy-based serum metabolomic spectral biomarkers using chemometrics for the diagnosis of Diabetic Retinopathy. FTIR spectroscopy was performed on 85 human serum samples [30 type 2 diabetes patients each without retinopathy and with retinopathy along with 25 normal healthy individuals as control]. Difference between mean spectra (DBMS), forward feature selection (FFS), and Mann–Whitney’s U tests were applied for spectral biomarker selection. Classification of disease conditions was achieved using analysis of different combinations of spectral features with linear, quadratic, and cubic Support Vector Machine at 10-fold cross validation. Twelve spectral signatures extracted by FFS could differentiate diabetes and diabetic retinopathy with 90% sensitivity, 92.7% specificity, and 90.5% overall accuracy. Two peaks (1042, 1114.18 cm−1) were associated with carbohydrate and polysaccharide content and five peaks (1114.18, 1165, 1211.18, 1402.70, 1451.14, 1657 cm−1) represented aberrations in total lipid content. Four peaks (1114.18, 1117, 1147, 1165 cm−1) were associated with protein phosphorylation and three peaks (1527, 1544.71, 1591.10 cm−1) with Amide II group. Again, lipidic signatures were strongly corroborated with glycosylated hemoglobin levels in diabetic retinopathy and diabetic subjects. Spectral signatures also revealed an elevated level of β-sheet containing proteins in serum in diabetic retinopathy condition. The method was validated through spectral biomarker selection by the DBMS technique. Thus, this method has the capability of diagnostic cost minimization for detection of diabetic retinopathy by label-free spectral biomarker identification.

Fourier-transform-infrared-spectroscopy based spectral-biomarker selection towards optimum diagnostic differentiation of oral leukoplakia and cancer.

In search of specific label-free biomarkers for differentiation of two oral lesions, namely oral leukoplakia (OLK) and oral squamous-cell carcinoma (OSCC), Fourier-transform infrared (FTIR) spectroscopy was performed on paraffin-embedded tissue sections from 47 human subjects (eight normal (NOM), 16 OLK, and 23 OSCC). Difference between mean spectra (DBMS), Mann-Whitney's U test, and forward feature selection (FFS) techniques were used for optimising spectral-marker selection. Classification of diseases was performed with linear and quadratic support vector machine (SVM) at 10-fold cross-validation, using different combinations of spectral features. It was observed that six features obtained through FFS enabled differentiation of NOM and OSCC tissue (1782, 1713, 1665, 1545, 1409, and 1161cm(-1)) and were most significant, able to classify OLK and OSCC with 81.3% sensitivity, 95.7% specificity, and 89.7% overall accuracy. The 43 spectral markers extracted through Mann-Whitney's U Test were the least significant when quadratic SVM was used. Considering the high sensitivity and specificity of the FFS technique, extracting only six spectral biomarkers was thus most useful for diagnosis of OLK and OSCC, and to overcome inter and intra-observer variability experienced in diagnostic best-practice histopathological procedure. By considering the biochemical assignment of these six spectral signatures, this work also revealed altered glycogen and keratin content in histological sections which could able to discriminate OLK and OSCC. The method was validated through spectral selection by the DBMS technique. Thus this method has potential for diagnostic cost minimisation for oral lesions by label-free biomarker identification.

THE ZURICH ENVIRONMENTAL STUDY OF GALAXIES IN GROUPS ALONG THE COSMIC WEB. III. GALAXY PHOTOMETRIC MEASUREMENTS AND THE SPATIALLY RESOLVED COLOR PROPERTIES OF EARLY- AND LATE-TYPE SATELLITES IN DIVERSE ENVIRONMENTS

We present photometric measurements for the galaxies - and when possible their bulges and disks - in the 0.05<z<0.0585 groups of the Zurich Environmental Study (ZENS); these measurements include (B-I) colors, color gradients and maps, color dispersions, as well as stellar masses and star-formation rates. The ZENS galaxies are classified into quenched, moderately star-forming, and strongly star-forming using a combination of spectral features and FUV-to-optical colors; this approach optimally distinguishes quenched systems from dust-reddened star-forming galaxies. The latter contribute up to 50% to the (B-I) "red sequence" at ~10^10Msun. At fixed morphological or spectral type, we find that galaxy stellar masses are largely independent of environment, and especially of halo mass. As a first utilization of our photometric database, we study, at fixed stellar mass and Hubble type, how (B-I) colors, color gradients and color dispersion of disk satellites depend on group mass (M_GROUP), group-centric distance (R/R_200) and large-scale structure overdensity. The strongest environmental trend is found for disk-dominated satellites with M_GROUP and R/R_200. At M<10^10 Msun, disk-dominated satellites are redder in the inner regions of the groups than in the outer parts. At M>10^10 Msun, these satellites have shallower color gradients in higher mass groups and in the cores of groups compared with lower mass groups and the outskirts of groups. Stellar population analyses and semi-analytic models suggest that disk-dominated satellites undergo quenching of star formation in their outer disks, on timescales ~2 Gyr, as they progressively move inside the group potential.

Spectral evidence for zeolite in the dust on Mars

Spectral features observed in Mars Global Surveyor Thermal Emission Spectrometer data ( ∼ 1670 –220 cm −1) of martian surface dust provide clues to its mineralogy. An emissivity peak at ∼ 1630 cm −1 is consistent with the presence of an H 2O-bearing mineral. This spectral feature can be mapped globally and shows a distribution related to the classical bright regions on Mars that are known to be dust covered. An important spectral feature at ∼ 830 cm −1 present in a newly derived average spectrum of surface dust likely is a transparency feature arising from the fine particulate nature of the dust. Its shape and location are consistent with plagioclase feldspars and also zeolites, which essentially are the hydrous form of feldspar. The generally favored visible/near-infrared spectral analog for martian dust, JSC Mars-1 altered tephra, does not display the ∼ 830 cm −1 feature. Zeolites commonly form from the interaction of low temperature aqueous fluids and volcanic glass in a variety of geologic settings. The combination of spectral features that are consistent with zeolites and the likelihood that Mars has (or had) geologic conditions necessary to produce them makes a strong case for recognizing zeolite minerals as likely components of the martian regolith.

Determination of39Kscattering lengths using photoassociation spectroscopy of the0g−state

Threshold scattering properties of ${}^{39}\mathrm{K}$ atoms are extracted from a line-shape analysis of photoassociation data. Specifically, we consider the rovibrational levels ${(v}^{\ensuremath{'}}=0--6)$ populated in the photoassociation process $\mathrm{K}(4S)+\mathrm{K}(4S)+\ensuremath{\Elzxh}\stackrel{\ensuremath{\rightarrow}}{\ensuremath{\omega}}{\mathrm{K}}_{2}^{*}[{0}_{g}^{\ensuremath{-}}{(v}^{\ensuremath{'}}{,J}^{\ensuremath{'}})].$ The measured spectra exhibit several general trends, including large ${J}^{\ensuremath{'}}=2$ linewidths, large relative ${J}^{\ensuremath{'}}=4$ peak intensities, and odd rotational lines which are weak or missing. The even ${J}^{\ensuremath{'}}$ features signal the presence of a d-wave shape resonance in the ground-state collisions. The combination of spectral features allows us to place the following limits on the ground-state singlet and triplet scattering lengths, ${a}_{s}{=140}_{\ensuremath{-}9}^{+6}$ a.u. and ${a}_{t}=\ensuremath{-}17\ensuremath{-}{0.045(C}_{6}\ensuremath{-}{C}_{6})\ifmmode\pm\else\textpm\fi{}25$ a.u., ${C}_{6}=3800$ a.u.