Method: Use Multiple Scattering Correction to eliminate the interference of scattering on spectrum in the process of field measurement so as to improve the accuracy of prediction model of tree canopy nitrogen content....Method: Use Multiple Scattering Correction to eliminate the interference of scattering on spectrum in the process of field measurement so as to improve the accuracy of prediction model of tree canopy nitrogen content. Apple trees in Qixia of Yantai City were taken as the test material. The spectral reflectivity of apple tree canopy went through the First Derivative (FD) and Multiple Scattering Correction (MSC) plus first derivative, respectively. The correlation coefficients were calculated between spectral reflectivity and nitrogen content. The Support Vector Machine (SVM) method was used to establish the prediction model. The result indicates that the MSC pre-processing can improve the correlation between spectral reflectivity and nitrogen content. The SVM model with MSC + FD pre-processing was a good way to predict the nitrogen content. The calibration R<sup>2</sup> of the model was 0.746;the validation R2 was 0.720;and its RMSE was 0.452 g·kgˉ<sup>1</sup>. MSC can commendably eliminate scattering error to improve the prediction accuracy of prediction model.展开更多
From the perspective of village classification, a set of methods for accurately measuring the potential of rural settlement consolidation are proposed. Taking Feicheng in Shandong Province as the research area, combin...From the perspective of village classification, a set of methods for accurately measuring the potential of rural settlement consolidation are proposed. Taking Feicheng in Shandong Province as the research area, combined with the corresponding planning, a classification and evaluation system for rural residential areas was constructed to classify rural residential areas from the four levels of natural resources, economy and society, supporting facilities and livability. The theoretical and practical potential of one type of rural settlements (relocating and merger village) is mainly calculated, and the potential scale, potential level and potential spatial distribution of different regions are analyzed. (Level and spatial distribution are analyzed. The results showed that the 621 rural settlements can be divided into five types: 148 urban-rural integration villages (URIV), 41 historical and cultural villages (HCV), 56 cluster developing villages (CDV), 155 keeping and limiting villages (KLV) and 221 relocating and merger villages (RMV). According to calculations, the theoretical potential of relocating and merger village areas was 1971.47 hm2, accounting for 19.52% of the total scale of Feicheng residential areas;the comprehensive correction coefficient of each region was 0.48 - 0.70, and the revised actual potential is 1082.68 hm2, accounting for the theoretical potential 54.92% of the total. There were large spatial differences in the scale of potential in different regions. In general, the theoretical and practical potential of the central and southern regions was higher than that of the northern regions.展开更多
This paper is aimed to monitor the soil nitrate nitrogen content in the apple orchards rapidly, accurately and in real time by making full use of the effective information of soil spectra. The 96 air-dried soil sample...This paper is aimed to monitor the soil nitrate nitrogen content in the apple orchards rapidly, accurately and in real time by making full use of the effective information of soil spectra. The 96 air-dried soil samples of the apple orchards in Qixia county, Yantai city, Shandong province were used as the data source. Spectral measurements of soil samples were carried out by ASD Fieldspec 3 in the darkroom, and the content of the soil nitrate nitrogen was determined by chemical method. Then the hyperspectral reflectance of soil samples were preprocessed by Multivariate Scatter Correction (MSC) and First Derivative (FD), the correlation analysis was carried out with the soil nitrate nitrogen content. The sensitive wavelength of soil nitrate nitrogen was screened. Finally, the Support Vector Machine (SVM) model for the soil nitrate nitrogen content was established. The results showed that the selected sensitive wavelength were 617 nm, 760 nm, 1239 nm, 1442 nm, 1535 nm, 1695 nm, 1776 nm, 1907 nm and 2088 nm. Hyperspectral monitoring model was established by SVM, in which the prediction set R2 was 0.959, RMSE was 0.281, RPD was 3.835;the correction set R2 was 0.822, RMSE was 0.392, RPD was 2.037. The SVM model could be used to monitor the soil nitrate content accurately.展开更多
The best hyperspectral estimation model of soil total nitrogen (TN) was established, which provided the basis for rapid and accurate estimation of soil total nitrogen content, scientific and rational fertilization and...The best hyperspectral estimation model of soil total nitrogen (TN) was established, which provided the basis for rapid and accurate estimation of soil total nitrogen content, scientific and rational fertilization and soil informatization management. A total of 92 brown soil samples were collected from the orchard of Qixia County, Yantai City, Shandong Province. After drying and grinding, the hyperspectrum of the soil was measured in the laboratory using ASD FieldSpec3. The TN contents of brown soil were measured by Kjeldahl method. The sensitive wavelengths were selected by multiple linear stepwise regression method. The hyperspectral estimation model of TN was established by Random Forest (RF) and Support Vector Machines (SVM). The models were validated by independent samples. The best estimation model was obtained. The sensitive wavelengths were 956 nm, 995 nm, 1020 nm, 1410 nm, 1659 nm and 2020 nm. The coefficients of determination (R2) of the two estimation models were 0.8011 and 0.8283, the root mean square errors (RMSE) were 0.022 and 0.025, and relative errors (RE) were 0.1422 and 0.1639, respectively. Random Forest model and Support Vector Machines model are feasible in estimating TN contents, but the Support Vector Machines model is better.展开更多
Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and...Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and classification of ground objects at home and abroad. The research results of identification and classification of forest tree species, grassland and urban land features were summarized. Then the researches of classification methods were summarized. Finally the prospects of hyperspectral remote sensing in ground object identification and classification were prospected.展开更多
We took distribution visualization of chlorophyll content in apple leaves to estimate the nutrient content and growth levels of apple leaves. 130 mature and non-destructive apple leaves were collected, and imaging spe...We took distribution visualization of chlorophyll content in apple leaves to estimate the nutrient content and growth levels of apple leaves. 130 mature and non-destructive apple leaves were collected, and imaging spectroscopy data were collected by SOC710VP hyperspectral imager. The chlorophyll content of the leaves was determined on the spectral information of the leaves. After pre-processing, we took linear wavelength stepwise regression method to choose the sensitive wavelength of chlorophyll content. And then we established partial least squares, principal component analysis and stepwise regression model. Finally, the chlorophyll content distribution visualization was realized. The results showed that the sensitive wavelengths of the chlorophyll content were 712.50 nm, 509.95 nm, 561.22 nm, 840.62 nm, 696.67 nm and 987.91 nm. The R2, RMSE, RE of the optical chlorophyll content estimation model, and the principal component analysis regression model, were 0.800, 0.319 and 26.4%. The chlorophyll content of each pixel on the hyperspectral image of apple leaves was calculated by the best estimation model and we completed the visualization distribution of chlorophyll content, which provided a technical support for the rapid detection of nutrient distribution.展开更多
Estimating an apple tree canopy nitrogen contents using hyperspectral techniques is important in theoretical and realistic significance for fertilization and management.Taking 80 Fuji apple trees at the early fruiting...Estimating an apple tree canopy nitrogen contents using hyperspectral techniques is important in theoretical and realistic significance for fertilization and management.Taking 80 Fuji apple trees at the early fruiting stage as the research objects,the hyperspectral characteristics of the apple canopy were analyzed systematically.The apple canopy hyperspectral and the canopy nitrogen contents were measured respectively.The canopy hyperspectral characteristics under different nitrogen contents were analyzed and selected the sensitive wave bands.The apple canopy nitrogen content monitoring models were established by using multiple regression method,robust regression and BP neural network method.The results showed that the canopy hyperspectral reflectance had obvious differences under different nitrogen contents.The sensitive bands concentrate on 724~1136 nm.Estimation models based on hyperspectral indices are not ideal.Models based on robust regression(M regression)and BP neural network are better than multiple statistical model,and the accuracy of the BP neural network monitoring model is the best.The results of the study provide a certain reference for estimating apple nutrition using hyperspectral technology.展开更多
The aim of this study is to establish the estimation model of potassium content in apple leaves by using vegetation index. A total of 96 fresh apple leaves were collected from 24 orchards in Qixia County, Shandong Pro...The aim of this study is to establish the estimation model of potassium content in apple leaves by using vegetation index. A total of 96 fresh apple leaves were collected from 24 orchards in Qixia County, Shandong Province. The spectral reflectance of the leaves was measured by ASD FieldSpec4. The difference vegetation index (DVI), ratio vegetation index (RVI) and normalized vegetation index (NDVI) were used to make the contour map through Matlab platform, and the combination of high correlation wavelength was selected to establish the random forest (RF) regression model of potassium content. The hyperspectral reflectance increased with the increase of leaf potassium content. The correlation between DVI and the content of potassium is higher than NDVI and RVI. The optimal vegetation index was DVI (364,740), the correlation coefficient was 0.5355. The random forest regression model established with DVI selected vegetation index was the best. R2 was 0.8995, RMSE and RE% were 0.0791 and 0.0617 respectively. Using DVI to establish the random forest regression model to reverse the potassium content of apple leaves has achieved good results. It is important to determine the growth status of apple in hyperspectral and to determine the potash fertilizer of apple trees.展开更多
The aims of this study are to explore the effect of different scales in the high spectral data on the estimation of chlorophyll content of apple leaves, to find out the optimal differential window scale and to establi...The aims of this study are to explore the effect of different scales in the high spectral data on the estimation of chlorophyll content of apple leaves, to find out the optimal differential window scale and to establish a model for estimating the chlorophyll content of apple leaves. Taking the apple leaves as the research object, the actual spectral reflectance of the leaves was determined by the ASD Field Spec 3 spectrometer and the chlorophyll contents of the leaves were measured in the laboratory. Firstly, the differential transformations from 1 to 30 window scales were done for actual spectral data respectively, and correlation analyses were done between apple leaf chlorophyll content and differential data, then two sensitive wavelengths were chosen under each window. Secondly, taking five consecutive differential windows as a group, the best differential window was selected in each group. Lastly, after the conversion of two sensitive wavelengths in six differential windows, relationship analyses between chlorophyll content of apple leaves and two sensitive wavelengths were done, then two new parameters with the largest correlation coefficient were chosen to establish estimation model. Results showed that with increasing differential window, the determination coefficient (R2) of estimation model decreased after an initial increase, the tipping point was at the 13th differential window scale. Testing the partial least squares (PLS) model and the stepwise regression (SR) model established under differential window scale of the 13th, the R2 of the SR model was higher than that of the PLS model. The RMSE and RE% of the SR model were lower than that of the PLS model, which showed that SR model was more suitable to estimate chlorophyll content.展开更多
Using the PROSAIL radiation transfer model and HJ-1A-HSI data to simulate the canopy reflectivity of apple trees, this study lays the foundation for the inversion of canopy parameters. Taking Qixia City of Yantai City...Using the PROSAIL radiation transfer model and HJ-1A-HSI data to simulate the canopy reflectivity of apple trees, this study lays the foundation for the inversion of canopy parameters. Taking Qixia City of Yantai City, Shandong Province as the research area, the apple tree was taken as the research object, and the hyperspectral reflectance, LAI and sample GPS of apple canopy were measured in the field. The parameters required for the PROSAIL model were obtained by experimental methods. The model simulates the reflectivity;the HSI image data is preprocessed, and the canopy reflectivity is extracted by GPS coordinates. The PROSAIL model and the HSI image simulated reflectance were fitted to the measured apple canopy reflectivity. The decisive factor (R2) of the simulated reflectance and the measured reflectance of the PROSAIL model was 0.9944, and the relative error (RE%)was 0.1845. The HSI data simulated reflectance and measured reflectance. The coefficient of determination is 0.9714 and the relative error is 0.6202. Both have achieved good fitting effects and can be used for inversion studies of apple canopy parameters.展开更多
The apple orchard in Qixia City, Yantai City, Shandong Province was used as the research area. The nitrogen content inversion of apple canopy was studied by using the satellite remote sensing images of GF-1. On the ba...The apple orchard in Qixia City, Yantai City, Shandong Province was used as the research area. The nitrogen content inversion of apple canopy was studied by using the satellite remote sensing images of GF-1. On the basis of GF-1 satellite multispectral image preprocessing, vegetation index was extracted by band math. The nitrogen sensitive vegetation index of apple canopy was selected by correlation analysis of nitrogen content in apple canopy. The best inversion model for the nitrogen content of apple canopy was selected by establishing the regression model of univariate and multivariate factors. The nitrogen content of the canopy of apple orchard in the study area was inverted in space. The results showed that the 6 vegetation indices of RVI, NDVI, EVI, VARI, NPCI and NRI were better correlated with nitrogen content in the vegetation index based on GF-1 satellite multispectral imaging. The best inversion model of nitrogen content in apple canopy layer is the multivariate stepwise regression (MSR) model: Nc = 35.74– 41.978^*NPCI-10.78^*NDVI. The R^2 and RMSE of the model was 0.69 and 1.07. The spatial inversion of nitrogen content in apple orchard canopy was obtained. This study provided theoretical basis and technical support for large-area rapid monitoring of regional fruit tree nutrients.展开更多
The nutrient inversion model of apple leaves was established by spectral analysis technology to provide technical support for the fine management of apple trees.In Shuangquan Town,Changqing District,Jinan City,Shandon...The nutrient inversion model of apple leaves was established by spectral analysis technology to provide technical support for the fine management of apple trees.In Shuangquan Town,Changqing District,Jinan City,Shandong Province,the Fuji apple trees with stopping period of spring shoots were taken as research objects.The spectral reflectance and nitrogen content of apple leaves were measured by ASD Field Spec 4 portable ground object spectrometer.Analyzed the correlation between leaf nitrogen content and spectral reflectance.The sensitive wavelengths with high correlation coefficient were select by fractional differential algorithm,and the optimal vegetation index was constructed and screened out.Partial Least Square Regression(PLSR),Support Vector Machine(SVM)and Random Forests(RF)method were used to construct an inversion model of leaf nitrogen content.The results show that the RF model based on fractional differential second-order treatment is the best inversion model for the nitrogen content of leaves during stopping period of spring shoots.The modeling accuracy determination coefficient R2 reached 0.891,RMSE was 0.0841,and RPD was 2.1396.The determination coefficient R2 of the fitting results of the verification set was 0.617,RMSE was 0.1251,and RPD was 1.7105.The inversion model established by RF method is effective in monitoring the nitrogen content in apple leaves,which provides a theoretical basis for monitoring the growth of apple by hyperspectral technology.展开更多
Imaging spectrometer was used to measure the spectral data of apple leaves.The spectral reflectance of apple leaves was extracted.The nitrogen content of apple leaves was correlated with the spectral reflectance after...Imaging spectrometer was used to measure the spectral data of apple leaves.The spectral reflectance of apple leaves was extracted.The nitrogen content of apple leaves was correlated with the spectral reflectance after SG smoothing first-order differential treatment.The sensitive wavelengths were selected and nitrogen content prediction models were founded.The results showed that the spectral of apple leaves with different concentration gradients were obvious.The higher nitrogen content was,the lower spectral reflectance was.Established estimation models by using the selected SG smooth first-order differential spectral sensitive wavelengths SG-FDR403,SG-FDR469,SG-FDR525,SG-FDR566,SG-FDR650,SG-FDR696,SG-FDR781,SG-FDR851,SG-FDR933.The determined coefficient(R^2)of the partial least squares model was 0.5202.The root mean square error(RMSE)of that was 2.19 and the relative error(RE)of that was 5.89%.The R^2 of the support vector machine(SVM)model was 0.724.The RMSE of that was 1.94,and the RE of that was 5.13%.It is indicated that the SVM model can estimate the nitrogen content of apple leaves effectively.展开更多
A prediction model of apple leaf nitrogen content based on ground imaging spectroscopy was established to rapidly and nondestructively detect nitrogen content in apple leaves.SOC710VP hyperspectral imager was used to ...A prediction model of apple leaf nitrogen content based on ground imaging spectroscopy was established to rapidly and nondestructively detect nitrogen content in apple leaves.SOC710VP hyperspectral imager was used to obtain the imaging spectral information of apple leaves,and the average spectral curve of interest region was extracted.The study is to analyze the characteristics of imaging spectral curves of apple leaves with different nitrogen content.On the basis of the SG smoothing and first derivative pretreatment of the spectral curve,the maximum sensitive band with nitrogen content is screened and the spectral parameters are constructed.Three modeling methods of BP,SVM and RF were used to establish the prediction model of nitrogen content in apple leaves.The results showed that in the visible range,the nitrogen content of apple leaves was negatively correlated with the reflectance of the spectral curve,and was most obvious in the green range.The R2 of BP,SVM and RF of apple leaf nitrogen content prediction model were 0.7283,0.8128,0.9086,RMSE were 0.9359,0.7365,0.5368,the R2 of test model were 0.6260,0.7294,0.6512,RMSE were 0.9460,0.7350,0.9024.Comparing the prediction results of the three models,the optimal prediction model is SVM model,which can well predict the nitrogen content of apple leaves.展开更多
文摘Method: Use Multiple Scattering Correction to eliminate the interference of scattering on spectrum in the process of field measurement so as to improve the accuracy of prediction model of tree canopy nitrogen content. Apple trees in Qixia of Yantai City were taken as the test material. The spectral reflectivity of apple tree canopy went through the First Derivative (FD) and Multiple Scattering Correction (MSC) plus first derivative, respectively. The correlation coefficients were calculated between spectral reflectivity and nitrogen content. The Support Vector Machine (SVM) method was used to establish the prediction model. The result indicates that the MSC pre-processing can improve the correlation between spectral reflectivity and nitrogen content. The SVM model with MSC + FD pre-processing was a good way to predict the nitrogen content. The calibration R<sup>2</sup> of the model was 0.746;the validation R2 was 0.720;and its RMSE was 0.452 g·kgˉ<sup>1</sup>. MSC can commendably eliminate scattering error to improve the prediction accuracy of prediction model.
文摘From the perspective of village classification, a set of methods for accurately measuring the potential of rural settlement consolidation are proposed. Taking Feicheng in Shandong Province as the research area, combined with the corresponding planning, a classification and evaluation system for rural residential areas was constructed to classify rural residential areas from the four levels of natural resources, economy and society, supporting facilities and livability. The theoretical and practical potential of one type of rural settlements (relocating and merger village) is mainly calculated, and the potential scale, potential level and potential spatial distribution of different regions are analyzed. (Level and spatial distribution are analyzed. The results showed that the 621 rural settlements can be divided into five types: 148 urban-rural integration villages (URIV), 41 historical and cultural villages (HCV), 56 cluster developing villages (CDV), 155 keeping and limiting villages (KLV) and 221 relocating and merger villages (RMV). According to calculations, the theoretical potential of relocating and merger village areas was 1971.47 hm2, accounting for 19.52% of the total scale of Feicheng residential areas;the comprehensive correction coefficient of each region was 0.48 - 0.70, and the revised actual potential is 1082.68 hm2, accounting for the theoretical potential 54.92% of the total. There were large spatial differences in the scale of potential in different regions. In general, the theoretical and practical potential of the central and southern regions was higher than that of the northern regions.
文摘This paper is aimed to monitor the soil nitrate nitrogen content in the apple orchards rapidly, accurately and in real time by making full use of the effective information of soil spectra. The 96 air-dried soil samples of the apple orchards in Qixia county, Yantai city, Shandong province were used as the data source. Spectral measurements of soil samples were carried out by ASD Fieldspec 3 in the darkroom, and the content of the soil nitrate nitrogen was determined by chemical method. Then the hyperspectral reflectance of soil samples were preprocessed by Multivariate Scatter Correction (MSC) and First Derivative (FD), the correlation analysis was carried out with the soil nitrate nitrogen content. The sensitive wavelength of soil nitrate nitrogen was screened. Finally, the Support Vector Machine (SVM) model for the soil nitrate nitrogen content was established. The results showed that the selected sensitive wavelength were 617 nm, 760 nm, 1239 nm, 1442 nm, 1535 nm, 1695 nm, 1776 nm, 1907 nm and 2088 nm. Hyperspectral monitoring model was established by SVM, in which the prediction set R2 was 0.959, RMSE was 0.281, RPD was 3.835;the correction set R2 was 0.822, RMSE was 0.392, RPD was 2.037. The SVM model could be used to monitor the soil nitrate content accurately.
文摘The best hyperspectral estimation model of soil total nitrogen (TN) was established, which provided the basis for rapid and accurate estimation of soil total nitrogen content, scientific and rational fertilization and soil informatization management. A total of 92 brown soil samples were collected from the orchard of Qixia County, Yantai City, Shandong Province. After drying and grinding, the hyperspectrum of the soil was measured in the laboratory using ASD FieldSpec3. The TN contents of brown soil were measured by Kjeldahl method. The sensitive wavelengths were selected by multiple linear stepwise regression method. The hyperspectral estimation model of TN was established by Random Forest (RF) and Support Vector Machines (SVM). The models were validated by independent samples. The best estimation model was obtained. The sensitive wavelengths were 956 nm, 995 nm, 1020 nm, 1410 nm, 1659 nm and 2020 nm. The coefficients of determination (R2) of the two estimation models were 0.8011 and 0.8283, the root mean square errors (RMSE) were 0.022 and 0.025, and relative errors (RE) were 0.1422 and 0.1639, respectively. Random Forest model and Support Vector Machines model are feasible in estimating TN contents, but the Support Vector Machines model is better.
文摘Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and classification of ground objects at home and abroad. The research results of identification and classification of forest tree species, grassland and urban land features were summarized. Then the researches of classification methods were summarized. Finally the prospects of hyperspectral remote sensing in ground object identification and classification were prospected.
文摘We took distribution visualization of chlorophyll content in apple leaves to estimate the nutrient content and growth levels of apple leaves. 130 mature and non-destructive apple leaves were collected, and imaging spectroscopy data were collected by SOC710VP hyperspectral imager. The chlorophyll content of the leaves was determined on the spectral information of the leaves. After pre-processing, we took linear wavelength stepwise regression method to choose the sensitive wavelength of chlorophyll content. And then we established partial least squares, principal component analysis and stepwise regression model. Finally, the chlorophyll content distribution visualization was realized. The results showed that the sensitive wavelengths of the chlorophyll content were 712.50 nm, 509.95 nm, 561.22 nm, 840.62 nm, 696.67 nm and 987.91 nm. The R2, RMSE, RE of the optical chlorophyll content estimation model, and the principal component analysis regression model, were 0.800, 0.319 and 26.4%. The chlorophyll content of each pixel on the hyperspectral image of apple leaves was calculated by the best estimation model and we completed the visualization distribution of chlorophyll content, which provided a technical support for the rapid detection of nutrient distribution.
文摘Estimating an apple tree canopy nitrogen contents using hyperspectral techniques is important in theoretical and realistic significance for fertilization and management.Taking 80 Fuji apple trees at the early fruiting stage as the research objects,the hyperspectral characteristics of the apple canopy were analyzed systematically.The apple canopy hyperspectral and the canopy nitrogen contents were measured respectively.The canopy hyperspectral characteristics under different nitrogen contents were analyzed and selected the sensitive wave bands.The apple canopy nitrogen content monitoring models were established by using multiple regression method,robust regression and BP neural network method.The results showed that the canopy hyperspectral reflectance had obvious differences under different nitrogen contents.The sensitive bands concentrate on 724~1136 nm.Estimation models based on hyperspectral indices are not ideal.Models based on robust regression(M regression)and BP neural network are better than multiple statistical model,and the accuracy of the BP neural network monitoring model is the best.The results of the study provide a certain reference for estimating apple nutrition using hyperspectral technology.
基金ACKNOWLEDGMENT This paper was supported by Shandong Province Natural Science Fund (ZR2012DM007), the National Nature Science Foundation of China (41271369) and Youth science and technology innovation fund of Shandong Agricultural University (23731).
文摘The aim of this study is to establish the estimation model of potassium content in apple leaves by using vegetation index. A total of 96 fresh apple leaves were collected from 24 orchards in Qixia County, Shandong Province. The spectral reflectance of the leaves was measured by ASD FieldSpec4. The difference vegetation index (DVI), ratio vegetation index (RVI) and normalized vegetation index (NDVI) were used to make the contour map through Matlab platform, and the combination of high correlation wavelength was selected to establish the random forest (RF) regression model of potassium content. The hyperspectral reflectance increased with the increase of leaf potassium content. The correlation between DVI and the content of potassium is higher than NDVI and RVI. The optimal vegetation index was DVI (364,740), the correlation coefficient was 0.5355. The random forest regression model established with DVI selected vegetation index was the best. R2 was 0.8995, RMSE and RE% were 0.0791 and 0.0617 respectively. Using DVI to establish the random forest regression model to reverse the potassium content of apple leaves has achieved good results. It is important to determine the growth status of apple in hyperspectral and to determine the potash fertilizer of apple trees.
文摘The aims of this study are to explore the effect of different scales in the high spectral data on the estimation of chlorophyll content of apple leaves, to find out the optimal differential window scale and to establish a model for estimating the chlorophyll content of apple leaves. Taking the apple leaves as the research object, the actual spectral reflectance of the leaves was determined by the ASD Field Spec 3 spectrometer and the chlorophyll contents of the leaves were measured in the laboratory. Firstly, the differential transformations from 1 to 30 window scales were done for actual spectral data respectively, and correlation analyses were done between apple leaf chlorophyll content and differential data, then two sensitive wavelengths were chosen under each window. Secondly, taking five consecutive differential windows as a group, the best differential window was selected in each group. Lastly, after the conversion of two sensitive wavelengths in six differential windows, relationship analyses between chlorophyll content of apple leaves and two sensitive wavelengths were done, then two new parameters with the largest correlation coefficient were chosen to establish estimation model. Results showed that with increasing differential window, the determination coefficient (R2) of estimation model decreased after an initial increase, the tipping point was at the 13th differential window scale. Testing the partial least squares (PLS) model and the stepwise regression (SR) model established under differential window scale of the 13th, the R2 of the SR model was higher than that of the PLS model. The RMSE and RE% of the SR model were lower than that of the PLS model, which showed that SR model was more suitable to estimate chlorophyll content.
基金the National Natural Science Foundation of China(41671346)National Key Research and Development Program of China (2017YFE0122500)+2 种基金the Taishan Scholar Assistance Program from Shandong Provincial GovernmentFunds of Shandong “Double Tops” Program(SYL2017XTTD02)Shandong major scientific and technological innovation project: Research demonstration and extension of orchard irrigation and fertilization in accurate management(2018CXGC0209).
文摘Using the PROSAIL radiation transfer model and HJ-1A-HSI data to simulate the canopy reflectivity of apple trees, this study lays the foundation for the inversion of canopy parameters. Taking Qixia City of Yantai City, Shandong Province as the research area, the apple tree was taken as the research object, and the hyperspectral reflectance, LAI and sample GPS of apple canopy were measured in the field. The parameters required for the PROSAIL model were obtained by experimental methods. The model simulates the reflectivity;the HSI image data is preprocessed, and the canopy reflectivity is extracted by GPS coordinates. The PROSAIL model and the HSI image simulated reflectance were fitted to the measured apple canopy reflectivity. The decisive factor (R2) of the simulated reflectance and the measured reflectance of the PROSAIL model was 0.9944, and the relative error (RE%)was 0.1845. The HSI data simulated reflectance and measured reflectance. The coefficient of determination is 0.9714 and the relative error is 0.6202. Both have achieved good fitting effects and can be used for inversion studies of apple canopy parameters.
基金the National Natural Science Foundation of China(41671346)National Key Research and Development Program of China (2017YFE0122500)+2 种基金the Taishan Scholar Assistance Program from Shandong Provincial GovernmentFunds of Shandong “Double Tops” Program(SYL2017XTTD02)Shandong major scientific and technological innovation project: Research demonstration and extension of orchard irrigation and fertilization in accurate management(2018CXGC0209).
文摘The apple orchard in Qixia City, Yantai City, Shandong Province was used as the research area. The nitrogen content inversion of apple canopy was studied by using the satellite remote sensing images of GF-1. On the basis of GF-1 satellite multispectral image preprocessing, vegetation index was extracted by band math. The nitrogen sensitive vegetation index of apple canopy was selected by correlation analysis of nitrogen content in apple canopy. The best inversion model for the nitrogen content of apple canopy was selected by establishing the regression model of univariate and multivariate factors. The nitrogen content of the canopy of apple orchard in the study area was inverted in space. The results showed that the 6 vegetation indices of RVI, NDVI, EVI, VARI, NPCI and NRI were better correlated with nitrogen content in the vegetation index based on GF-1 satellite multispectral imaging. The best inversion model of nitrogen content in apple canopy layer is the multivariate stepwise regression (MSR) model: Nc = 35.74– 41.978^*NPCI-10.78^*NDVI. The R^2 and RMSE of the model was 0.69 and 1.07. The spatial inversion of nitrogen content in apple orchard canopy was obtained. This study provided theoretical basis and technical support for large-area rapid monitoring of regional fruit tree nutrients.
基金This paper was supported by the National Natural Science Foundation of China(41671346)the National Key Research and Development Program of China(2017YFE0122500)+1 种基金Shandong Major Scientific and Technological Innovation Project(2018CXGC0209)the Taishan Scholar Assistance Program from Shandong Provincial Government,Funds of Shandong“Double Tops”Program(SYL2017XTTD02).
文摘The nutrient inversion model of apple leaves was established by spectral analysis technology to provide technical support for the fine management of apple trees.In Shuangquan Town,Changqing District,Jinan City,Shandong Province,the Fuji apple trees with stopping period of spring shoots were taken as research objects.The spectral reflectance and nitrogen content of apple leaves were measured by ASD Field Spec 4 portable ground object spectrometer.Analyzed the correlation between leaf nitrogen content and spectral reflectance.The sensitive wavelengths with high correlation coefficient were select by fractional differential algorithm,and the optimal vegetation index was constructed and screened out.Partial Least Square Regression(PLSR),Support Vector Machine(SVM)and Random Forests(RF)method were used to construct an inversion model of leaf nitrogen content.The results show that the RF model based on fractional differential second-order treatment is the best inversion model for the nitrogen content of leaves during stopping period of spring shoots.The modeling accuracy determination coefficient R2 reached 0.891,RMSE was 0.0841,and RPD was 2.1396.The determination coefficient R2 of the fitting results of the verification set was 0.617,RMSE was 0.1251,and RPD was 1.7105.The inversion model established by RF method is effective in monitoring the nitrogen content in apple leaves,which provides a theoretical basis for monitoring the growth of apple by hyperspectral technology.
基金the National Natural Science Foundation of China(41671346)Funds of Shandong“Double Tops”Program(SYL2017XTTD02)Shandong major scientific and technological innovation project:Research demonstration and extension of orchard irrigation and fertilization in accurate management(2018CXGC0209).
文摘Imaging spectrometer was used to measure the spectral data of apple leaves.The spectral reflectance of apple leaves was extracted.The nitrogen content of apple leaves was correlated with the spectral reflectance after SG smoothing first-order differential treatment.The sensitive wavelengths were selected and nitrogen content prediction models were founded.The results showed that the spectral of apple leaves with different concentration gradients were obvious.The higher nitrogen content was,the lower spectral reflectance was.Established estimation models by using the selected SG smooth first-order differential spectral sensitive wavelengths SG-FDR403,SG-FDR469,SG-FDR525,SG-FDR566,SG-FDR650,SG-FDR696,SG-FDR781,SG-FDR851,SG-FDR933.The determined coefficient(R^2)of the partial least squares model was 0.5202.The root mean square error(RMSE)of that was 2.19 and the relative error(RE)of that was 5.89%.The R^2 of the support vector machine(SVM)model was 0.724.The RMSE of that was 1.94,and the RE of that was 5.13%.It is indicated that the SVM model can estimate the nitrogen content of apple leaves effectively.
基金the National Natural Science Foundation of China(41671346)Funds of Shandong“Double Tops”Program(SYL2017XTTD02).
文摘A prediction model of apple leaf nitrogen content based on ground imaging spectroscopy was established to rapidly and nondestructively detect nitrogen content in apple leaves.SOC710VP hyperspectral imager was used to obtain the imaging spectral information of apple leaves,and the average spectral curve of interest region was extracted.The study is to analyze the characteristics of imaging spectral curves of apple leaves with different nitrogen content.On the basis of the SG smoothing and first derivative pretreatment of the spectral curve,the maximum sensitive band with nitrogen content is screened and the spectral parameters are constructed.Three modeling methods of BP,SVM and RF were used to establish the prediction model of nitrogen content in apple leaves.The results showed that in the visible range,the nitrogen content of apple leaves was negatively correlated with the reflectance of the spectral curve,and was most obvious in the green range.The R2 of BP,SVM and RF of apple leaf nitrogen content prediction model were 0.7283,0.8128,0.9086,RMSE were 0.9359,0.7365,0.5368,the R2 of test model were 0.6260,0.7294,0.6512,RMSE were 0.9460,0.7350,0.9024.Comparing the prediction results of the three models,the optimal prediction model is SVM model,which can well predict the nitrogen content of apple leaves.