Estimation of nitrogen, phosphorus, and potassium contents in the leaves of different plants using laboratory-based visible and near-infrared reflectance spectroscopy: comparison of partial least-square regression and support vector machine regression methods

The rapid and accurate detection of the moisture content is of great significance to the quality evaluation and oil extraction process of walnut kernel. Near-infrared (NIR) spectroscopy is an ideal method for measuring the moisture content in walnut kernel. In this study, a regression model for moisture content in walnut kernel was developed based on NIR diffuse reflectance spectroscopy using chemometric methods. The different spectral pretreatment methods were adopted to preprocess the original spectral data. The whole spectra band was divided into 5 subbands, 10 subbands, 15 subbands, and 20 subbands to screen specific wavelengths relevant to the walnut kernel moisture content. PLS (partial least square regression), MLR (multivariate linear regression), PCR (principle component regression), and SVR (support vector regression) were used to establish the relationship model between the spectral data and measurement values of the moisture content. In comparison, the optimized modeling conditions were determined as follows: detection wavelength 1349–1490 nm, SNV-FD (standard normal variate transformation and first derivative) preprocessing method, and PLS algorithm. Under these conditions, the square correlation coefficient (R2) and root mean square error of prediction (RMSEP) of the prediction model were 0.9865 and 0.0017, respectively. The results of this study provided a feasible method for the rapid detection of moisture content in walnut kernel. To improve the performance and applicability of the model, it is necessary to continuously expand the size of the sample set.

Download Full-text

Determination of Fat, Protein, and Total Solids in Cheese by Near-Infrared Reflectance Spectroscopy

Journal of AOAC International ◽

10.1093/jaoac/78.3.802 ◽

1995 ◽

Vol 78 (3) ◽

pp. 802-806 ◽

Cited By ~ 46

Author(s):

José Louis Rodriguez-Otero ◽

Maria Hermida ◽

Alberto Cepeda

Keyword(s):

Near Infrared ◽

Independent Set ◽

Nir Spectroscopy ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Sample Treatment ◽

Infrared Reflectance ◽

Near Infrared Reflectance ◽

Total Solids

Abstract Near-infrared reflectance (NIR) spectroscopy was used to analyze fat, protein, and total solids in cheese without any sample treatment. A set of 92 samples of cow’s milk cheese was used for instrument calibration by principal components analysis and modified partial least-square regression. The following statistical values were obtained: standard error of calibration (SEC) = 0.388 and squared correlation coefficient (R2) = 0.99 for fat, SEC = 0.397 and R2 = 0.98 for protein, and SEC = 0.412 and R2 = 0.99 for total solids. To validate the calibration, an independent set of 25 cheese samples of the same type was used. Standard errors of validation were 0.47,0.50, and 0.61 for fat, protein, and total solids, respectively, and hf for the regression of measurements by reference methods versus measurements by NIR spectroscopy was 0.98 for the 3 components.

Download Full-text

Visible and near infrared reflectance spectroscopy for field-scale assessment of Stagnosols properties

Plant Soil and Environment ◽

10.17221/220/2018-pse ◽

2018 ◽

Vol 64 (No. 6) ◽

pp. 276-282 ◽

Cited By ~ 1

Author(s):

Šestak Ivana ◽

Mesić Milan ◽

Zgorelec Željka ◽

Perčin Aleksandra ◽

Stupnišek Ivan

Keyword(s):

Soil Ph ◽

Near Infrared ◽

Soil Depth ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Infrared Reflectance ◽

Field Scale ◽

Near Infrared Reflectance ◽

Vnir Spectroscopy

Spectral data contain information on soil organic and mineral composition, which can be useful for soil quality monitoring. The objective of research was to evaluate hyperspectral visible and near infrared reflectance (VNIR) spectroscopy for field-scale prediction of soil properties and assessment of factors affecting soil spectra. Two hundred soil samples taken from the experiment field (soil depth: 30 cm; sampling grid: 15 × 15 m) were scanned using portable spectroradiometer (350–1050 nm) to identify spectral differences of soil treated with ten different rates of mineral nitrogen (N) fertilizer (0–300 kg N/ha). Principal component analysis revealed distinction between higher- and lower-N level treatments conditioned by differences in soil pH, texture and soil organic matter (SOM) composition. Partial least square regression resulted in very strong correlation and low root mean square error (RMSE) between predicted and measured values for the calibration (C) and validation (V) dataset, respectively (SOM, %: RC2 = 0.75 and RV2 = 0.74; RMSEC = 0.334 and RMSEV = 0.346; soil pH: RC2 = 0.78 and RV2 = 0.62; RMSEC = 0.448 and RMSEV = 0.591). Results indicated that hyperspectral VNIR spectroscopy is an efficient method for measurement of soil functional attributes within precision farming framework.

Download Full-text

Prediction of Soil Texture Distributions by using PLSR and Reflectance Spectroscopy

International Journal of Recent Technology and Engineering - 2 ◽

10.35940/ijrte.c6558.098319 ◽

2019 ◽

Vol 8 (3) ◽

pp. 7876-7881

Keyword(s):

Statistical Analysis ◽

Soil Texture ◽

Near Infrared ◽

Diffuse Reflectance Spectroscopy ◽

Principal Component ◽

Reflectance Spectroscopy ◽

Soil Samples ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression

The texture of soil i.e. Sand, Silt and Clay are the most important physical properties of soil for agricultural management. In the agricultural practices to increase the productivity of soil, moisture-holding capacity, aeration and to support the agronomic decisions the knowledge of soil texture is an essential task. For this purpose, the present research gives better results and fast acquisition of soil information with the use of Visible and Near Infrared (Vis- NIR) Diffuse Reflectance Spectroscopy. A total of 30 soil samples from two different locations from Aurangabad, Maharashtra, India were collected and analyzed for soil texture. To detect the soil texture the Vis-NIR DRS has shown levels of accurate results compared to the traditional laboratory method with less time, cost and effort. To measure the reflectance of soil the ASD FieldSpec4 Spectroradiometer (350-2500nm) was used. By the observation of captured spectra by using Spectroradiometer it showed that on the basis of different textural classes the soil samples could be spectrally separable. For database collection and pre-processing, we have used RS3 and ViewSpec Pro software respectively. The statistical analysis by using the combination of Principal Component Analysis (PCA) and Partial Least Square Regression method gives accurate results. To determine the texture of soil sample thirteen features were calculated. The main goal of this research was to determine the soil texture by using statistical methods and to test the performance of VNIR-SWIR reflectance spectroscopy by using the ASD FieldSpec4 Spectroradiometer for estimation of the texture of the soil. The results showed that R2 = 0.99 gives maximum accuracy for clay content and R2 = 0.988 for silt content and R2 = 0.989 for sand. The Root Mean Square Values (RMSE) for clay, silt, and sand are 0.02392, 0.02399 and 0.02289 respectively. With the use of reflectance spectroscopy and statistical analysis by using regression models we can determine the soil properties accurately in very less time.

Download Full-text

Transferability and scaling of soil total carbon prediction models in Florida

10.7287/peerj.preprints.494v1 ◽

2014 ◽

Author(s):

Sabine Grunwald ◽

Congrong Yu ◽

Xiong Xiong

Keyword(s):

Prediction Error ◽

Near Infrared ◽

Prediction Models ◽

Regional Scale ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Total Carbon ◽

Support Vector ◽

Domain Iii

The applicability, transfer, and scalability of visible/near-infrared (VNIR)-derived soil models are still poorly understood. The objectives of this study in Florida, U.S. were to: (i) compare three methods to predict soil total carbon (TC) using five fields (local scale) and a pooled (regional scale) VNIR spectral dataset, (ii) assess the model’s transferability among fields, and (iii) evaluate the up- and down-scaling behavior of TC prediction models. A total of 560 TC-spectral sets were modeled by Partial Least Square Regression (PLSR), Support Vector Machine (SVM), and Random Forest. The transferability and up- and down-scaling of models were limited by the following factors: (i) the spectral data domain, (ii) soil attribute domain, (iii) methods that describe the internal model structure of VNIR-TC relationships, and (iv) environmental domain space of attributes that control soil carbon dynamics. All soil logTC models showed excellent performance based on all three methods with R2 > 0.86, bias < 0.01%, root mean square prediction error (RMSE) = 0.09%, residual predication deviation (RPD) > 2.70% , and ratio of prediction error to inter-quartile range (RPIQ) > 4.54. PLSR performed substantially better than SVM to scale and transfer models. Upscaled soil TC models performed somewhat better in terms of model fit (R2), RPD, and RPIQ, whereas downscaled models showed less bias and smaller RMSE based on PLSR. Given the many factors that can impinge on empirically derived soil spectral prediction models, as demonstrated by this study, more focus on the applicability and scaling of them is needed.

Download Full-text

Study Hybrid Teknologi Laser – Near Infrared Spectroscopy Untuk Identifikasi Unsur Hara Pada Lahan Gambut

Jurnal Ilmiah Mahasiswa Pertanian ◽

10.17969/jimfp.v4i4.12615 ◽

2019 ◽

Vol 4 (4) ◽

pp. 492-501

Author(s):

Endamin Endamin ◽

Ichwana Ichwana ◽

Agus Arip Munawar

Keyword(s):

Infrared Spectroscopy ◽

Near Infrared ◽

Scatter Correction ◽

Reflectance Spectroscopy ◽

Partial Least Square ◽

Least Square ◽

Near Infrared Reflectance ◽

Acoustic Spectroscopy ◽

Non Destructive ◽

Photo Acoustic Spectroscopy

Abstrak. Lahan gambut terbentuk dari akumulasi sisa-sisa vegetasi yang sudah mengalami humifikasi tetapi belum mengalami mineralisasi. Berdasarkan data Balai Besar Penelitian dan Pengembangan Sumberdaya Lahan Pertanian (2011), lahan gambut di Aceh memiliki luas 216.000 Ha yang tersebar di beberapa kabupaten, salah satunya adalah Kabupaten Aceh Barat. Kondisi lahan gambut di Provinsi Aceh belum dimanfaatkan secara maksimal. Hal ini dikarenakan tidak adanya pengetahuan dan penanganan yang tepat terhadap pengelolaan lahan gambut agar tanaman dapat tumbuh dengan optimal. Semakin berkembangnya ilmu pengetahuan dan teknologi, informasi tentang kandungan unsur hara pada lahan gambut dapat diketahui dengan cepat. Salah satunya adalah pemanfaatan teknologi NDT (Non Destructive Test) yang pengujiannya dapat dilakukan tanpa harus merusak media ataupun objek yang ingin diketahui kandungan unsur haranya. Pengujian dengan NIRS (Near Infrared Reflectance Spectroscopy) merupakan salah satu metode non-destruktif seperti LPAS (Laser Photo Acoustic Spectroscopy) yang dapat digunakan untuk menganalisis dalam berbagai bidang, termasuk unsur hara tanah. Penggunaan laser sebagai sumber pencahayaan berfungsi untuk dapat menembus bahan dengan ketebalan lebih dari 1 cm. Spektrum transmisi laser diakuisisi dengan metode Partial Least Square (PLS). Metode ini menjadi alternatif dalam menganalisis unsur C-Organik, Nitrogen (N) dan Kalium (K) dari segi parameter kimia dalam lahan gambut tersebut sehingga unsur haranya dapat dideteksi dengan cepat dan tepat. Spektrum laser He-Ne untuk tanah didapatkan dengan menggunakan instrumen self developed infrared spectroscopy (FT-NIRS) dengan konfigurasi alur kerja alat (workflow) dibangun dengan menggunakan self modified Thermo Integration®. Spektrum transmisi laser He-Ne diakuisisi dengan metode pulsed excitation dengan wavenumber 5000 – 11000 . Koreksi dan perbaikan spektrum dilakukan dengan tujuan untuk menghilangkan noise pada spektrum akibat interferensi dan scattering photon, serta pengaruh over-heat. Metode yang digunakan adalah Deresolve dan Multiplicative Scatter Correction (MSC). Hybrid Study of Laser Technology - Near Infrared Spectroscopy for Identification of Nutrients in Peatlands Abstract. Peatlands are formed from accumulated remnants of vegetation that have undergone humification but have not experienced mineralization. Based on data from the Center for Agricultural Land Research and Development (2011), peat land in Aceh has an area of 216,000 hectares spread across several districts, one of which is West Aceh District. The condition of peatlands in Aceh Province has not been fully utilized. This is because there is no knowledge and proper handling of peatland management so that the plants can grow optimally. As science and technology develops, information about nutrient content in peatlands can be identified quickly. One of them is the use of NDT (Non Destructive Test) technology, which tests can be carried out without having to damage the media or objects that want to know the elemental content. Testing with NIRS (Near Infrared Reflectance Spectroscopy) is one of the non-destructive methods such as LPAS (Laser Photo Acoustic Spectroscopy) which can be used to analyze in various fields, such as soil nutrients. The use of lasers as a source of lighting functions to be able to penetrate materials with a thickness of more than 1 cm. The laser transmission spectrum was acquired by the Partial Least Square (PLS) method. This method is an alternative in analyzing C-Organic, Nitrogen (N) and Potassium (K) elements in terms of chemical parameters in the peat so that the elements can be detected quickly and precisely. The He-Ne laser spectrum for soil is obtained using the self developed infrared spectroscopy (FT-NIRS) instrument with workflow configurations built using self-modified Thermo Integration®. The He-Ne laser transmission spectrum was acquired by the pulsed excitation method with a wavenumber of 5000 - 11000 cm – 1. Spectrum correction and repairs are carried out with the aim of eliminating noise in the spectrum due to photon interference and scattering, and the effect of overheating. The method used is Deresolve and Multiplicative Scatter Correction (MSC).

Download Full-text

Prediction of Composition of Broiler Chicken Excreta Using Near-infrared Reflectance Spectroscopy

Journal of Agricultural and Marine Sciences [JAMS] ◽

10.24200/jams.vol10iss1pp13-19 ◽

2005 ◽

Vol 10 (1) ◽

pp. 13

Author(s):

I. T. Kadim ◽

W. Al-Marzooqi ◽

O. Mahgoub ◽

K. Annamalai

Keyword(s):

Near Infrared ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Coefficient Of Determination ◽

Infrared Reflectance ◽

Near Infrared Reflectance ◽

Acid Detergent Fibre ◽

Gross Energy ◽

Calibration Equations

Near-infrared reflectance spectroscopic (NIRS) calibrations were developed for the prediction of the content of dry matter (DM); nitrogen (N), ether extract (EE), neutral detergent fibre (NDF), acid detergent fibre (ADF), gross energy (GE), calcium (Ca) and phosphate (P) in broiler excreta samples. The chemical composition of broiler excreta was determined by the conventional chemical analysis methods in the laboratory and compared with NIRS. Excreta samples (n = 72) were oven dried (60 oC) and analyzed for DM, N, EE, NDF, ADF, GE, Ca and P. The determined values (mean ± SD) were as follows: DM: 31.46 ± 7.65 (range:19.14 - 44.51), N: 5.85 ± 2.88 (range: 4.85 -7.00), EE: 1.37 ± 0.25 (range: 0.88-1.99), ADF: 16.71 ± 1.99 (range: 12.11-19.97), NDF: 26.26 ± 1.63 (range: 22.03-30.21), GE: 15.27 ± 0.33 (range: 14.52-16.11), Ca: 2.57 ± 0.22 (range: 2.16-3.01), P: 1.79 ± 0.15 (range: 1.41-2.11). The samples were then scanned in a NIRS model 5000 analyzer and the spectra obtained for each sample. Calibration equations and prediction values were developed for broiler excreta samples. The software used modified partial least square regression statistic, as it is most suitable for natural products. For broiler excreta samples, the coefficient of determination (R2) and the standard error of prediction (SEP) was DM = 0.97, 1.27, N = 0.95, 0.72, EE = 0.92, 0.07, ADF = 0.87, 0.78, NDF = 0.88, 0.72, GE = 0.89; 0.24, Ca = 0.96, 0.06, P = 0.93, 0.09, respectively. The results indicate that it is possible to calibrate NIRS to predict major constituents in broiler excreta samples.

Download Full-text

Prediction of Mineral Composition in Commercial Extruded Dry Dog Food by Near-Infrared Reflectance Spectroscopy

Animals ◽

10.3390/ani9090640 ◽

2019 ◽

Vol 9 (9) ◽

pp. 640 ◽

Cited By ~ 4

Author(s):

Goi ◽

Manuelian ◽

Currò ◽

Marchi

Keyword(s):

Mineral Content ◽

Near Infrared ◽

Prediction Models ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Coefficient Of Determination ◽

Emission Spectrometry ◽

Near Infrared Reflectance ◽

Dog Food

The pet food industry is interested in performing fast analyses to control the nutritional quality of their products. This study assessed the feasibility of near-infrared spectroscopy to predict mineral content in extruded dry dog food. Mineral content in commercial dry dog food samples (n = 119) was quantified by inductively coupled plasma optical emission spectrometry and reflectance spectra (850–2500 nm) captured with FOSS NIRS DS2500 spectrometer. Calibration models were built using modified partial least square regression and leave-one-out cross-validation. The best prediction models were obtained for S (coefficient of determination; R2 = 0.89), K (R2 = 0.85), and Li (R2 = 0.74), followed by P, B, and Sr (R2 = 0.72 each). Only prediction models for S and K were adequate for screening purposes. This study supports that minerals are difficult to determine with NIRS if they are not associated with organic molecules.

Download Full-text

Quantification of potassium concentration with Vis–SWNIR spectroscopy in fresh lettuce

Journal of Innovative Optical Health Sciences ◽

10.1142/s1793545820500297 ◽

2020 ◽

Vol 13 (06) ◽

pp. 2050029

Author(s):

Yating Xiong ◽

Shintaroh Ohashi ◽

Kazuhiro Nakano ◽

Weizhong Jiang ◽

Kenichi Takizawa ◽

...

Keyword(s):

Near Infrared ◽

Potassium Concentration ◽

Spectral Characteristics ◽

Public Health Problem ◽

Short Wave ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Support Vector ◽

Major Public Health Problem

Chronic kidney disease (CKD) is becoming a major public health problem worldwide, and excessive potassium intake is a health threat to patients with CKD. In this study, visible–short-wave near-infrared (Vis–SWNIR) spectroscopy and chemometric algorithms were investigated as nondestructive methods for assessing the potassium concentration in fresh lettuce to benefit the CKD patients’ health. Interactance and transmittance measurements were performed and the competencies were compared based on the multivariate methods of partial least-square regression (PLS) and support vector machine regression (SVR). Meanwhile, several preprocessing methods [first- and second-order derivatives in combination with standard normal variate (SNV)] and wavelength selection method of competitive adaptive reweighted sampling (CARS) were applied to eliminate noise and highlight the spectral characteristics. The PLS models yielded better prediction than the SVR models with higher correlation coefficients ([Formula: see text]) and residual predictive deviation (RPD), and lower root-mean-square error of prediction (RMSEP). Excellent prediction of green leaves was obtained by the interactance measurement with [Formula: see text], [Formula: see text][Formula: see text]mg/100[Formula: see text]g, and [Formula: see text]; while the transmittance spectra of petioles provided optimal prediction with [Formula: see text], [Formula: see text][Formula: see text]mg/100[Formula: see text]g, and RPD[Formula: see text]=[Formula: see text]3.34, respectively. Therefore, the results indicated that Vis–SWNIR spectroscopy is capable of intelligently detecting potassium concentration in fresh lettuce to benefit CKD patients around the world in maintaining and enhancing their health.

Download Full-text

Transferability and scaling of soil total carbon prediction models in Florida

10.7287/peerj.preprints.494 ◽

2014 ◽

Author(s):

Sabine Grunwald ◽

Congrong Yu ◽

Xiong Xiong

Keyword(s):

Prediction Error ◽

Near Infrared ◽

Prediction Models ◽

Regional Scale ◽

Partial Least Square ◽

Least Square ◽

Partial Least Square Regression ◽

Total Carbon ◽

Support Vector ◽

Domain Iii

The applicability, transfer, and scalability of visible/near-infrared (VNIR)-derived soil models are still poorly understood. The objectives of this study in Florida, U.S. were to: (i) compare three methods to predict soil total carbon (TC) using five fields (local scale) and a pooled (regional scale) VNIR spectral dataset, (ii) assess the model’s transferability among fields, and (iii) evaluate the up- and down-scaling behavior of TC prediction models. A total of 560 TC-spectral sets were modeled by Partial Least Square Regression (PLSR), Support Vector Machine (SVM), and Random Forest. The transferability and up- and down-scaling of models were limited by the following factors: (i) the spectral data domain, (ii) soil attribute domain, (iii) methods that describe the internal model structure of VNIR-TC relationships, and (iv) environmental domain space of attributes that control soil carbon dynamics. All soil logTC models showed excellent performance based on all three methods with R2 > 0.86, bias < 0.01%, root mean square prediction error (RMSE) = 0.09%, residual predication deviation (RPD) > 2.70% , and ratio of prediction error to inter-quartile range (RPIQ) > 4.54. PLSR performed substantially better than SVM to scale and transfer models. Upscaled soil TC models performed somewhat better in terms of model fit (R2), RPD, and RPIQ, whereas downscaled models showed less bias and smaller RMSE based on PLSR. Given the many factors that can impinge on empirically derived soil spectral prediction models, as demonstrated by this study, more focus on the applicability and scaling of them is needed.

Download Full-text