Evaluation of effective spectral features for glacial lake mapping by using Landsat-8 OLI imagery

Glacial lake mapping provides the most feasible way for investigating the water resources and monitoring the flood outburst hazards in High Mountain Region.However,various types of glacial lakes with different properties bring a constraint to the rapid and accurate glacial lake mapping over a large scale.Existing spectral features to map glacial lakes are diverse but some are generally limited to the specific glaciated regions or lake types,some have unclear applicability,which hamper their application for the large areas.To this end,this study provides a solution for evaluating the most effective spectral features in glacial lake mapping using Landsat-8 imagery.The 23 frequently-used lake mapping spectral features,including single band reflectance features,Water Index features and image transformation features were selected,then the insignificant features were filtered out based on scoring calculated from two classical feature selection methods-random forest and decision tree algorithm.The result shows that the three most prominent spectral features(SF)with high scores are NDWI1,EWI,and NDWI3(renamed as SF8,SF19 and SF12 respectively).Accuracy assessment of glacial lake mapping results in five different test sites demonstrate that the selected features performed well and robustly in classifying different types of glacial lakes without any influence from the mountain shadows.SF8 and SF19 are superior for the detection of large amount of small glacial lakes,while some lake areas extracted by SF12 are incomplete.Moreover,SF8 achieved better accuracy than the other two features in terms of both Kappa Coefficient(0.8812)and Prediction(0.9025),which further indicates that SF8 has great potential for large scale glacial lake mapping in high mountainous area.

Stacked spectral feature space patch: An advanced spectral representation for precise crop classification based on convolutional neural network

Spectral and spatial features in remotely sensed data play an irreplaceable role in classifying crop types for precision agriculture. Despite the thriving establishment of the handcrafted features, designing or selecting such features valid for specific crop types requires prior knowledge and thus remains an open challenge. Convolutional neural networks(CNNs) can effectively overcome this issue with their advanced ability to generate high-level features automatically but are still inadequate in mining spectral features compared to mining spatial features. This study proposed an enhanced spectral feature called Stacked Spectral Feature Space Patch(SSFSP) for CNN-based crop classification. SSFSP is a stack of twodimensional(2 D) gridded spectral feature images that record various crop types’ spatial and intensity distribution characteristics in a 2 D feature space consisting of two spectral bands. SSFSP can be input into2 D-CNNs to support the simultaneous mining of spectral and spatial features, as the spectral features are successfully converted to 2 D images that can be processed by CNN. We tested the performance of SSFSP by using it as the input to seven CNN models and one multilayer perceptron model for crop type classification compared to using conventional spectral features as input. Using high spatial resolution hyperspectral datasets at three sites, the comparative study demonstrated that SSFSP outperforms conventional spectral features regarding classification accuracy, robustness, and training efficiency. The theoretical analysis summarizes three reasons for its excellent performance. First, SSFSP mines the spectral interrelationship with feature generality, which reduces the required number of training samples.Second, the intra-class variance can be largely reduced by grid partitioning. Third, SSFSP is a highly sparse feature, which reduces the dependence on the CNN model structure and enables early and fast convergence in model training. In conclusion, SSFSP has great potential for practical crop classification in precision agriculture.

Spectral feature characterization and nitrogen content prediction in soils with different particle sizes and moisture contents

The objective of this research is to analyze the influences of light source incidence angle,fiber height,moisture content,and particle size on loamy mixed soil spectra.Nitrogen(N)content calibration and cross-validation models at different moisture contents and particle sizes were obtained using partial least squares(PLS)analysis.Spectral data were collected using a spectrophotometer.Fiber height of 100 mm and light source angle at 45°were chosen to obtain the sharpest spectra without apparent scattering effect.The results show that moisture content and particle size strongly influenced the absorbance of the spectra,and a better N prediction model was obtained when the particle sizes were in the ranges of 0.5-1.0,1.0-2.0 and 2.0-5.0 mm,with the correlation coefficients(r)of 0.819,0.815 and 0.818,and standard errors of prediction(SEP)of 2.29,2.41 and 2.42 mg/kg,respectively.Poor N prediction model was obtained when the soil was kept in its natural moisture content with r of 0.575 and SEP of 3.275 mg/kg,compared to the performance of dried soil samples with r of 0.815 and SEP of 2.425 mg/kg.

A double-layer model for improving the estimation of wheat canopy nitrogen content from unmanned aerial vehicle multispectral imagery

The accurate and rapid estimation of canopy nitrogen content(CNC)in crops is the key to optimizing in-season nitrogen fertilizer application in precision agriculture.However,the determination of CNC from field sampling data for leaf area index(LAI),canopy photosynthetic pigments(CPP;including chlorophyll a,chlorophyll b and carotenoids)and leaf nitrogen concentration(LNC)can be time-consuming and costly.Here we evaluated the use of high-precision unmanned aerial vehicle(UAV)multispectral imagery for estimating the LAI,CPP and CNC of winter wheat over the whole growth period.A total of 23 spectral features(SFs;five original spectrum bands,17 vegetation indices and the gray scale of the RGB image)and eight texture features(TFs;contrast,entropy,variance,mean,homogeneity,dissimilarity,second moment,and correlation)were selected as inputs for the models.Six machine learning methods,i.e.,multiple stepwise regression(MSR),support vector regression(SVR),gradient boosting decision tree(GBDT),Gaussian process regression(GPR),back propagation neural network(BPNN)and radial basis function neural network(RBFNN),were compared for the retrieval of winter wheat LAI,CPP and CNC values,and a double-layer model was proposed for estimating CNC based on LAI and CPP.The results showed that the inversion of winter wheat LAI,CPP and CNC by the combination of SFs+TFs greatly improved the estimation accuracy compared with that by using only the SFs.The RBFNN and BPNN models outperformed the other machine learning models in estimating winter wheat LAI,CPP and CNC.The proposed double-layer models(R^(2)=0.67-0.89,RMSE=13.63-23.71 mg g^(-1),MAE=10.75-17.59 mg g^(-1))performed better than the direct inversion models(R^(2)=0.61-0.80,RMSE=18.01-25.12 mg g^(-1),MAE=12.96-18.88 mg g^(-1))in estimating winter wheat CNC.The best winter wheat CNC accuracy was obtained by the double-layer RBFNN model with SFs+TFs as inputs(R^(2)=0.89,RMSE=13.63 mg g^(-1),MAE=10.75 mg g^(-1)).The results of this study can provide guidance for the accurate and rapid determination of winter wheat canopy nitrogen content in the field.

Sanxingdui Cultural Relics Recognition Algorithm Based on Hyperspectral Multi-Network Fusion

Sanxingdui cultural relics are the precious cultural heritage of humanity with high values of history,science,culture,art and research.However,mainstream analytical methods are contacting and detrimental,which is unfavorable to the protection of cultural relics.This paper improves the accuracy of the extraction,location,and analysis of artifacts using hyperspectral methods.To improve the accuracy of cultural relic mining,positioning,and analysis,the segmentation algorithm of Sanxingdui cultural relics based on the spatial spectrum integrated network is proposed with the support of hyperspectral techniques.Firstly,region stitching algorithm based on the relative position of hyper spectrally collected data is proposed to improve stitching efficiency.Secondly,given the prominence of traditional HRNet(High-Resolution Net)models in high-resolution data processing,the spatial attention mechanism is put forward to obtain spatial dimension information.Thirdly,in view of the prominence of 3D networks in spectral information acquisition,the pyramid 3D residual network model is proposed to obtain internal spectral dimensional information.Fourthly,four kinds of fusion methods at the level of data and decision are presented to achieve cultural relic labeling.As shown by the experiment results,the proposed network adopts an integrated method of data-level and decision-level,which achieves the optimal average accuracy of identification 0.84,realizes shallow coverage of cultural relics labeling,and effectively supports the mining and protection of cultural relics.

A Study of Estimation Model for the Chlorophyll Content of Wheat Leaf Based on Hyperspectral Imaging

In order to explore the spectral features and sensitive wave band of wheat leaf,we establish a quantitative relationship model between wheat chlorophyll content and spectral features to promote the application of hyperspectral technology in precise wheat fertilization and fast,non-destructive growth monitoring.Using the relational analysis,we analyze the relationship between chlorophyll content and spectral reflectance or the first derivative,and establish the chlorophyll content monitoring model.By selection and verification,the best estimation models for wheat chlorophyll content are as follows:SPAD = 36.75 + 188.168R387,SPAD =2094.242R'7153+ 112646.744 R'7152-1.561E7 R'715+42.991.The two models can well estimate the SPAD value of wheat leaf,and comparatively speaking,the SPAD estimation model based on wave band R387 has greater accuracy.

Exploring the potential of Sentinel-2A satellite data for aboveground biomass estimation in fragmented Himalayan subtropical pine forest

The Sentinel-2 A satellite having embedded advantage of red edge spectral bands offers multispectral imageries with improved spatial,spectral and temporal resolutions as compared to the other contemporary satellites providing medium resolution data.Our study was aimed at exploring the potential of Sentinel-2 A imagery to estimate Above Ground Biomass(AGB) of Subtropical Pine Forest in Pakistan administered Kashmir.We developed an AGB predictive model using field inventory and Sentinel 2 A based spectral and textural parameters along with topographic features derived from ALOS Digital Elevation Model(DEM).Field inventory data was collected from 108 randomly distributed plots(0.1 ha each) across the study area.The stepwise linear regression method was employed to investigate the potential relationship between field data and corresponding satellite data.Biomass and carbon mapping of the study area was carried out through established AGB estimation model with R(o.86),R2(0.74),adjusted R2(0.72) and RMSE value of 33 t/ha.Our results showed that first order textures(mean,standard deviation and variance) significantly contributed in AGB predictive modeling while only one spectral band ratio made contribution from spectral domain.Our study leads to the conclusion that Sentinel-2 A optical data is a potential source for AGB estimation in subtropical pine forest of the area of interest with added benefit of its free of cost availability,higher quality data and long-term continuity that can be utilized for biomass carbon distribution mapping in the resource constraint study area for sustainable forest management.

A Rub-Impact Recognition Method Based on Improved Convolutional Neural Network

Based on the theory of modal acoustic emission(AE),when the convolutional neural network(CNN)is used to identify rotor rub-impact faults,the training data has a small sample size,and the AE sound segment belongs to a single channel signal with less pixel-level information and strong local correlation.Due to the convolutional pooling operations of CNN,coarse-grained and edge information are lost,and the top-level information dimension in CNN network is low,which can easily lead to overfitting.To solve the above problems,we first propose the use of sound spectrograms and their differential features to construct multi-channel image input features suitable for CNN and fully exploit the intrinsic characteristics of the sound spectra.Then,the traditional CNN network structure is improved,and the outputs of all convolutional layers are connected as one layer constitutes a fused feature that contains information at each layer,and is input into the network’s fully connected layer for classification and identification.Experiments indicate that the improved CNN recognition algorithm has significantly improved recognition rate compared with CNN and dynamical neural network(DNN)algorithms.

Weighted spectral correlation angle target detection method for land‑based hyperspectral imaging

In land-based spectral imaging,the spectra of ground objects are inevitably afected by the imaging conditions(weather conditions,atmospheric conditions,light conditions,zenith and azimuth angle conditions)and spatial distribution of targets,leading to uncertainties featured by“same object diferent spectrum”.That is,the spectrum of a ground object may change within a certain range under diferent imaging conditions.Traditional target detection(TD)methods are mainly based on similarity measurements and do not fully account for the spectral uncertainties.These detection methods are prone to false detections or missed detections.Therefore,reducing the impact of spectral uncertainties on TD is an important research topic in hyperspectral imaging.In this paper,we frst review traditional TD methods and compare their principles and characteristics.It is found that the spectral correlation angle(SCA)method has good adaptability in land-based imaging.The shortcoming of the SCA method that it cannot refect the local spectrum characteristics,is also analyzed.As the efect of spectral uncertainties cannot be completely overcome by the SCA method,a new similarity measurement method,the weighted spectral correlation angle(WSCA)method,is proposed.It can reduce the infuence of spectral uncertainties on TD by increasing the weight of particular bands.Finally,we use two sets of experiments to analyze the efect of the WSCA method on TD.Its performance in overcoming spectral uncertainties caused by variations in imaging conditions or uneven spatial distributions of targets is tested.The results show that the WSCA method can efectively reduce the infuence of spectral uncertainties and obtain a good detection result.

Determination of soil physicochemical attributes in farming sites through visible,near-infrared diffuse reflectance spectroscopy and PLSR modeling

Introduction:An accurate and reliable detection of soil physicochemical attributes(SPAs)is a difficult and complicated issue in soil science.The SPA may be varied spatially and temporally with the complexity of nature.In the past,SPA detection has been obtained through routine soil chemical and physical laboratory analysis.However,these laboratory methods do not fulfill the rapid requirements.Accordingly,diffuse reflectance spectroscopy(DRS)can be used to nondestructively detect and characterize soil attributes with superior solution.In the present article,we report a study done through spectral curves in the visible(350–700 nm)and near-infrared(700–2500 nm)(VNIR)region of 74 soil specimens which were agglomerated by farming sectors of Phulambri Tehsil of the Aurangabad region of Maharashtra,India.The quantitative analysis of VNIR spectrum was done.Results:The spectra of agglomerated farming soils were acquired by the Analytical Spectral Device(ASD)Field spec 4 spectroradiometer.The soil spectra of the VNIR region were preprocessed to get pure spectra which were the input for regression modeling.The partial least squares regression(PLSR)model was computed to construct the calibration models,which were individually validated for the prediction of SPA from the soil spectrum.The computed model was based on a correlation study between reflected spectra and detected SPA.The detected SPAs were soil organic carbon(SOC),nitrogen(N),soil organic matter(SOM),pH values,electrical conductivity(EC),phosphorus(P),potassium(K),iron(Fe),sand,silt,and clay.The accuracy of the PLSR model-validated determinant(R^(2))values were SOC 0.89,N 0.68,SOM 0.93,pH values 0.82,EC 0.89,P 0.98,K 0.82,Fe 0.94,sand 0.98,silt 0.90,and clay 0.69 with root mean square error of prediction(RMSEP)3.51,4.34,2.66,2.12,4.11,1.41,4.22,1.56,1.89,1.97,and 9.91,respectively.According to the experimental results,the VNIR-DRS was better for detection of SPA and produced more accurate predictions for SPA.Conclusions:In conclusion,the methods examined here offered rapid and novel detection of SPA from reflectance spectroscopy.The outcome of the present research will be apt for precision farming and decision-making.

Unsupervised GRNN flood mapping approach combined with uncertainty analysis using bi-temporal Sentinel-2 MSI imageries

Floods occur frequently worldwide.The timely,accurate mapping of the flooded areas is an important task.Therefore,an unsupervised approach is proposed for automated flooded area mapping from bitemporal Sentinel-2 multispectral images in this paper.First,spatial–spectral features of the images before and after the flood are extracted to construct the change magnitude image(CMI).Then,the certain flood pixels and non-flood pixels are obtained by performing uncertainty analysis on the CMI,which are considered reliable classification samples.Next,Generalized Regression Neural Network(GRNN)is used as the core classifier to generate the initial flood map.Finally,an easy-toimplement two-stage post-processing is proposed to reduce the mapping error of the initial flood map,and generate the final flood map.Different from other methods based on machine learning,GRNN is used as the classifier,but the proposed approach is automated and unsupervised because it uses samples automatically generated in uncertainty analysis for model training.Results of comparative experiments in the three sub-regions of the Poyang Lake Basin demonstrate the effectiveness and superiority of the proposed approach.Moreover,its superiority in dealing with uncertain pixels is further proven by comparing the classification accuracy of different methods on uncertain pixels.