Swarm intelligence algorithms own superior performance in solving high-dimensional and multi-objective optimization problems.The application of the swarm intelligence algorithms to visible and near-infrared(VIS-NIR)sp...Swarm intelligence algorithms own superior performance in solving high-dimensional and multi-objective optimization problems.The application of the swarm intelligence algorithms to visible and near-infrared(VIS-NIR)spectral analysis of soil moisture can contribute to the optimization of the soil moisture prediction model and the development of the real-time soil moisture sensor.In this study,a high-resolution spectrometer was used to obtain spectral data of different levels of soil moisture which were manually configured.Isolation Forest algorithm(iForest)was used to eliminate outliers from the data.Based on the root mean square error of prediction RMSEP of Back Propagation Neural Network(BPNN)model results,a series of new swarm intelligence algorithms,including Manta Ray Foraging Optimization(MRFO),Slime Mould Algorithm(SMA),etc.,were used to select the characteristic wavelengths of soil moisture.The analysis results showed that MRFO owned the best performance if only from the predictive capability perspective and SMA had a better performance when considering the proportion of the selecting wavelengths and the results of the model prediction.By comparing and analyzing the modeling results of traditional intelligence algorithms Genetic Algorithm(GA)and Particle Swarm Optimization(PSO),it was found that the new swarm intelligence had a better performance in selecting the characteristic wavelengths of soil moisture.Integrating the results of all intelligence algorithms used,soil moisture sensitive wavelengths were selected as 490 nm,513 nm,543 nm,900 nm and 926 nm,which provide the basis for the design of real-time soil moisture sensor based on VIS-NIR.展开更多
The aim of this study was in-line,rapid,and non-destructive detection for soluble solid content(SSC)in pomelos using visible and near-infrared spectroscopy(Vis-NIRS).However,the large size and thick rind of pomelo aff...The aim of this study was in-line,rapid,and non-destructive detection for soluble solid content(SSC)in pomelos using visible and near-infrared spectroscopy(Vis-NIRS).However,the large size and thick rind of pomelo affect the stability of spectral acquisition and the biological variabilities affect the robustness of models.Given these issues,in this study,an efficient prototype in-line detection system in transmittance mode was designed and evaluated in comparison with an off-line detection system.Data from the years 2019 and 2020 were used for modeling and the external validation data were obtained by the inline detection system in 2021.The wavelength selection methods of changeable size moving window(CSMW),random frog(RF),and competitive adaptive reweighted sampling(CARS)were used to improve the prediction accuracy of partial least squares regression(PLSR)models.The best performance of internal prediction was obtained by CARS-PLSR and the determination coefficient of prediction(),root mean square error of prediction(RMSEP),and residual predictive deviation(RPD)were 0.958,0.204%,and 4.821,respectively.However,all models obtained large prediction biases in external validation.The latent variable updating(LVU)method was proposed to update models and improve the performance in external validation.Ten samples from the external validation set were randomly selected to update the models.Compared with the recalibration method,LVU could effectively modify the original models which matched the SSC range of the external validation set.The CSMW-PLSR models were more robust in external validations.The off-line model with LVU performed best with a root mean square error of validation(RMSEV)of 0.599%and the in-line model with recalibration obtained RMSEV of 0.864%.These results demonstrated the application potential of the transmittance Vis-NIRS for in-line rapid prediction of SSC in pomelos and the modeling and updating methods could be applied to samples with biological variabilities.展开更多
Powdery mildew is a disease that threatens wheat production and causes severe economic losses worldwide. Its timely diagnosis is imperative for preventing and controlling its spread. In this study, the multiangle cano...Powdery mildew is a disease that threatens wheat production and causes severe economic losses worldwide. Its timely diagnosis is imperative for preventing and controlling its spread. In this study, the multiangle canopy spectra and disease severity of wheat were investigated at several developmental stages and degrees of disease severity. Four wavelength variable-selected algorithms: successive projection(SPA), competitive adaptive reweighted sampling(CARS), feature selection learning(Relief-F), and genetic algorithm(GA), were used to identify bands sensitive to powdery mildew. The wavelength variables selected were used as input variables for partial least squares(PLS), extreme learning machine(ELM), random forest(RF), and support vector machine(SVM) algorithms, to construct a suitable prediction model for powdery mildew. Spectral reflectance and conventional vegetation indices(VIs) displayed angle effects under several disease severity indices(DIs). The CARS method selected relatively few wavelength variables and showed a relatively homogeneous distribution across the 13 viewing zenith angles.Overall accuracies of the four modeling algorithms were ranked as follows: ELM(0.70–0.82) > PLS(0.63–0.79) > SVM(0.49–0.69) > RF(0.43–0.69). Combinations of features and algorithms generated varied accuracies, with coefficients of determination(R^(2)) single-peaked at different observation angles. The constructed CARS-ELM model extracted a predictable bivariate relationship between the multi-angle canopy spectrum and disease severity, yielding an R^(2)> 0.8 at each measured angle. Especially for larger angles,monitoring accuracies were increased relative to the optimal VI model(40% at-60°, 33% at +60°), indicating that the CARS-ELM model is suitable for extreme angles of-60° and +60°. The results are proposed to provide a technical basis for rapid and large-scale monitoring of wheat powdery mildew.展开更多
This study was undertaken to investigate the feasibility of near-infrared(NIR) hyperspectral imaging(1 000–2 500 nm) for non-destructive and quantitative prediction of protein content in peanut kernels. Partial least...This study was undertaken to investigate the feasibility of near-infrared(NIR) hyperspectral imaging(1 000–2 500 nm) for non-destructive and quantitative prediction of protein content in peanut kernels. Partial least squares regression(PLSR) calibration model was established between the spectral data extracted from the hyperspectral images and the reference measured protein content values, with the coefficient of determination of prediction(R_P^2) of 0.885 and root mean square error of prediction(RMSEP) of 0.465%.Regression coefficients(RC) from PLSR analysis were used to identify the most essential wavelengths that had the greatest influence on changes in the protein content. Eight optimal wavelengths were selected by RC and its corresponding simplified RC-PLSR prediction model was also obtained, showing better performance with a higher R_P^2 of 0.870 and a lower RMSEP of 0.494%. The results indicate that hyperspectral imaging with PLSR analysis can be used as a rapid and non-destructive method for predicting protein content in peanut.展开更多
We propose and experimentally demonstrate compact on-chip 1×2 wavelength selective switches(WSSs) based on silicon microring resonators(MRRs) with nested pairs of subrings(NPSs). Owing to the resonance splitting ...We propose and experimentally demonstrate compact on-chip 1×2 wavelength selective switches(WSSs) based on silicon microring resonators(MRRs) with nested pairs of subrings(NPSs). Owing to the resonance splitting induced by the inner NPSs, the proposed devices are capable of performing selective channel routing at certain resonance wavelengths of the outer MRRs. System demonstration of dynamic channel routing using fabricated devices with one and two NPSs is carried out for 10 Gb∕s non-return-to-zero signal. The experimental results verify the effectiveness of the fabricated devices as compact on-chip WSSs.展开更多
A novel wavelength selectable fiber ring laser was proposed. Superimposed fiber Bragg gratings with uniform reflectivity were made to serve as comb-like filter. Optical tunable filter combined with the comb filter rea...A novel wavelength selectable fiber ring laser was proposed. Superimposed fiber Bragg gratings with uniform reflectivity were made to serve as comb-like filter. Optical tunable filter combined with the comb filter realized fast and accurately wavelength selectivity. The SMSR at each wavelength is more than 50dB.展开更多
We propose a wavelength selective diffraction using reflectors placed on three-dimensional grid cross points. Different wavelengths are separated into spots distributed in two-dimensional plane. Compact device with hi...We propose a wavelength selective diffraction using reflectors placed on three-dimensional grid cross points. Different wavelengths are separated into spots distributed in two-dimensional plane. Compact device with high port counts is attainable.展开更多
This study was carried out to investigate the feasibility of using visible and near infrared hyperspectral imaging for the variety classification of mung beans.Raw hyperspectral images of mung beans were acquired in t...This study was carried out to investigate the feasibility of using visible and near infrared hyperspectral imaging for the variety classification of mung beans.Raw hyperspectral images of mung beans were acquired in the wavelengths of 380-1023 nm,and all images were calibrated by the white and dark reference images.The spectral reflectance values were extracted from the region of interest(ROI)of each calibrated hyperspectral image,and then they were treated as the independent variables.The dependent variables of four varieties of mung beans were set as 1,2,3 and 4,respectively.The extreme learning machine(ELM)model was established using full spectral wavelengths for classification.Modified gram-schmidt(MGS)method was used to identify effective wavelengths.Based on the selected wavelengths,the ELM and linear discriminant analysis(LDA)models were built.All models performed excellently with the correct classification rates(CCRs)covering 99.17%-99.58% in the training sets and 99.17%-100%in the testing sets.Fifteen wavelengths(432 nm,455 nm,468 nm,560 nm,705 nm,736 nm,760 nm,841 nm,861 nm,921 nm,930 nm,937 nm,938 nm,959 nm and 965 nm)were recommended by MGS.The results demonstrated that hyperspectral imaging could be used as a non-destructive method to classify mung bean varieties,and MGS was an effective wavelength selection method.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.32071915)China Agriculture Research System of MOF and MARA-Food Legumes(CARS-08).
文摘Swarm intelligence algorithms own superior performance in solving high-dimensional and multi-objective optimization problems.The application of the swarm intelligence algorithms to visible and near-infrared(VIS-NIR)spectral analysis of soil moisture can contribute to the optimization of the soil moisture prediction model and the development of the real-time soil moisture sensor.In this study,a high-resolution spectrometer was used to obtain spectral data of different levels of soil moisture which were manually configured.Isolation Forest algorithm(iForest)was used to eliminate outliers from the data.Based on the root mean square error of prediction RMSEP of Back Propagation Neural Network(BPNN)model results,a series of new swarm intelligence algorithms,including Manta Ray Foraging Optimization(MRFO),Slime Mould Algorithm(SMA),etc.,were used to select the characteristic wavelengths of soil moisture.The analysis results showed that MRFO owned the best performance if only from the predictive capability perspective and SMA had a better performance when considering the proportion of the selecting wavelengths and the results of the model prediction.By comparing and analyzing the modeling results of traditional intelligence algorithms Genetic Algorithm(GA)and Particle Swarm Optimization(PSO),it was found that the new swarm intelligence had a better performance in selecting the characteristic wavelengths of soil moisture.Integrating the results of all intelligence algorithms used,soil moisture sensitive wavelengths were selected as 490 nm,513 nm,543 nm,900 nm and 926 nm,which provide the basis for the design of real-time soil moisture sensor based on VIS-NIR.
基金the key research and development projects of Zhejiang province(Grant No.2022C02021).
文摘The aim of this study was in-line,rapid,and non-destructive detection for soluble solid content(SSC)in pomelos using visible and near-infrared spectroscopy(Vis-NIRS).However,the large size and thick rind of pomelo affect the stability of spectral acquisition and the biological variabilities affect the robustness of models.Given these issues,in this study,an efficient prototype in-line detection system in transmittance mode was designed and evaluated in comparison with an off-line detection system.Data from the years 2019 and 2020 were used for modeling and the external validation data were obtained by the inline detection system in 2021.The wavelength selection methods of changeable size moving window(CSMW),random frog(RF),and competitive adaptive reweighted sampling(CARS)were used to improve the prediction accuracy of partial least squares regression(PLSR)models.The best performance of internal prediction was obtained by CARS-PLSR and the determination coefficient of prediction(),root mean square error of prediction(RMSEP),and residual predictive deviation(RPD)were 0.958,0.204%,and 4.821,respectively.However,all models obtained large prediction biases in external validation.The latent variable updating(LVU)method was proposed to update models and improve the performance in external validation.Ten samples from the external validation set were randomly selected to update the models.Compared with the recalibration method,LVU could effectively modify the original models which matched the SSC range of the external validation set.The CSMW-PLSR models were more robust in external validations.The off-line model with LVU performed best with a root mean square error of validation(RMSEV)of 0.599%and the in-line model with recalibration obtained RMSEV of 0.864%.These results demonstrated the application potential of the transmittance Vis-NIRS for in-line rapid prediction of SSC in pomelos and the modeling and updating methods could be applied to samples with biological variabilities.
基金supported by the National Natural Science Foundation of China (31971791)the National Key Research and Development Program of China (2017YFD0300204)。
文摘Powdery mildew is a disease that threatens wheat production and causes severe economic losses worldwide. Its timely diagnosis is imperative for preventing and controlling its spread. In this study, the multiangle canopy spectra and disease severity of wheat were investigated at several developmental stages and degrees of disease severity. Four wavelength variable-selected algorithms: successive projection(SPA), competitive adaptive reweighted sampling(CARS), feature selection learning(Relief-F), and genetic algorithm(GA), were used to identify bands sensitive to powdery mildew. The wavelength variables selected were used as input variables for partial least squares(PLS), extreme learning machine(ELM), random forest(RF), and support vector machine(SVM) algorithms, to construct a suitable prediction model for powdery mildew. Spectral reflectance and conventional vegetation indices(VIs) displayed angle effects under several disease severity indices(DIs). The CARS method selected relatively few wavelength variables and showed a relatively homogeneous distribution across the 13 viewing zenith angles.Overall accuracies of the four modeling algorithms were ranked as follows: ELM(0.70–0.82) > PLS(0.63–0.79) > SVM(0.49–0.69) > RF(0.43–0.69). Combinations of features and algorithms generated varied accuracies, with coefficients of determination(R^(2)) single-peaked at different observation angles. The constructed CARS-ELM model extracted a predictable bivariate relationship between the multi-angle canopy spectrum and disease severity, yielding an R^(2)> 0.8 at each measured angle. Especially for larger angles,monitoring accuracies were increased relative to the optimal VI model(40% at-60°, 33% at +60°), indicating that the CARS-ELM model is suitable for extreme angles of-60° and +60°. The results are proposed to provide a technical basis for rapid and large-scale monitoring of wheat powdery mildew.
基金Supported by the Natural Science Foundation of Guangdong Province(2017A030310558)China Postdoctoral Science Foundation(2017M612672)Fundamental Research Funds for the Central Universities(2017MS067)
文摘This study was undertaken to investigate the feasibility of near-infrared(NIR) hyperspectral imaging(1 000–2 500 nm) for non-destructive and quantitative prediction of protein content in peanut kernels. Partial least squares regression(PLSR) calibration model was established between the spectral data extracted from the hyperspectral images and the reference measured protein content values, with the coefficient of determination of prediction(R_P^2) of 0.885 and root mean square error of prediction(RMSEP) of 0.465%.Regression coefficients(RC) from PLSR analysis were used to identify the most essential wavelengths that had the greatest influence on changes in the protein content. Eight optimal wavelengths were selected by RC and its corresponding simplified RC-PLSR prediction model was also obtained, showing better performance with a higher R_P^2 of 0.870 and a lower RMSEP of 0.494%. The results indicate that hyperspectral imaging with PLSR analysis can be used as a rapid and non-destructive method for predicting protein content in peanut.
基金supported in part by the National Natural Science Foundation of China under Grant 61125504/61235007in part by the 863 High-Tech Program under Grant 2013AA013402
文摘We propose and experimentally demonstrate compact on-chip 1×2 wavelength selective switches(WSSs) based on silicon microring resonators(MRRs) with nested pairs of subrings(NPSs). Owing to the resonance splitting induced by the inner NPSs, the proposed devices are capable of performing selective channel routing at certain resonance wavelengths of the outer MRRs. System demonstration of dynamic channel routing using fabricated devices with one and two NPSs is carried out for 10 Gb∕s non-return-to-zero signal. The experimental results verify the effectiveness of the fabricated devices as compact on-chip WSSs.
文摘A novel wavelength selectable fiber ring laser was proposed. Superimposed fiber Bragg gratings with uniform reflectivity were made to serve as comb-like filter. Optical tunable filter combined with the comb filter realized fast and accurately wavelength selectivity. The SMSR at each wavelength is more than 50dB.
文摘We propose a wavelength selective diffraction using reflectors placed on three-dimensional grid cross points. Different wavelengths are separated into spots distributed in two-dimensional plane. Compact device with high port counts is attainable.
基金This work was supported by the National Key Scientific Instrument and Equipment Development Projects(2014YQ470377)the Scientific Research Foundation for Returned Overseas Students and the Fundamental Research Funds for the Central Universities of China(2012FZA6005,2013QNA6011).
文摘This study was carried out to investigate the feasibility of using visible and near infrared hyperspectral imaging for the variety classification of mung beans.Raw hyperspectral images of mung beans were acquired in the wavelengths of 380-1023 nm,and all images were calibrated by the white and dark reference images.The spectral reflectance values were extracted from the region of interest(ROI)of each calibrated hyperspectral image,and then they were treated as the independent variables.The dependent variables of four varieties of mung beans were set as 1,2,3 and 4,respectively.The extreme learning machine(ELM)model was established using full spectral wavelengths for classification.Modified gram-schmidt(MGS)method was used to identify effective wavelengths.Based on the selected wavelengths,the ELM and linear discriminant analysis(LDA)models were built.All models performed excellently with the correct classification rates(CCRs)covering 99.17%-99.58% in the training sets and 99.17%-100%in the testing sets.Fifteen wavelengths(432 nm,455 nm,468 nm,560 nm,705 nm,736 nm,760 nm,841 nm,861 nm,921 nm,930 nm,937 nm,938 nm,959 nm and 965 nm)were recommended by MGS.The results demonstrated that hyperspectral imaging could be used as a non-destructive method to classify mung bean varieties,and MGS was an effective wavelength selection method.