Sensors

Research

16 pages, 4578 KB

Open AccessArticle

Soil Organic Carbon Prediction Based on Vis–NIR Spectral Classification Data Using GWPCA–FCM Algorithm

by Yutong Miao, Haoyu Wang, Xiaona Huang, Kexin Liu, Qian Sun, Lingtong Meng and Dongyun Xu

Sensors 2024, 24(15), 4930; https://doi.org/10.3390/s24154930 - 30 Jul 2024

Cited by 1 | Viewed by 2275

Soil visible and near–infrared reflectance spectroscopy is an effective tool for the rapid estimation of soil organic carbon (SOC). The development of spectroscopic technology has increased the application of spectral libraries for SOC research. However, the direct application of spectral libraries for SOC [...] Read more.

Soil visible and near–infrared reflectance spectroscopy is an effective tool for the rapid estimation of soil organic carbon (SOC). The development of spectroscopic technology has increased the application of spectral libraries for SOC research. However, the direct application of spectral libraries for SOC prediction remains challenging due to the high variability in soil types and soil–forming factors. This study aims to address this challenge by improving SOC prediction accuracy through spectral classification. We utilized the European Land Use and Cover Area frame Survey (LUCAS) large–scale spectral library and employed a geographically weighted principal component analysis (GWPCA) combined with a fuzzy c–means (FCM) clustering algorithm to classify the spectra. Subsequently, we used partial least squares regression (PLSR) and the Cubist model for SOC prediction. Additionally, we classified the soil data by land cover types and compared the classification prediction results with those obtained from spectral classification. The results showed that (1) the GWPCA–FCM–Cubist model yielded the best predictions, with an average accuracy of R² = 0.83 and RPIQ = 2.95, representing improvements of 10.33% and 18.00% in R² and RPIQ, respectively, compared to unclassified full sample modeling. (2) The accuracy of spectral classification modeling based on GWPCA–FCM was significantly superior to that of land cover type classification modeling. Specifically, there was a 7.64% and 14.22% improvement in R² and RPIQ, respectively, under PLSR, and a 13.36% and 29.10% improvement in R² and RPIQ, respectively, under Cubist. (3) Overall, the prediction accuracy of Cubist models was better than that of PLSR models. These findings indicate that the application of GWPCA and FCM clustering in conjunction with the Cubist modeling technique can significantly enhance the prediction accuracy of SOC from large–scale spectral libraries. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

11 pages, 2735 KB

Open AccessArticle

Developing a Portable Autofluorescence Detection System and Its Application in Biological Samples

by Jiaxing Zhou, Yunfei Li, Jinfeng Zhang and Fuhong Cai

Sensors 2024, 24(11), 3351; https://doi.org/10.3390/s24113351 - 23 May 2024

Cited by 2 | Viewed by 2252

Abstract

Advanced glycation end-products (AGEs) are complex compounds closely associated with several chronic diseases, especially diabetes mellitus (DM). Current methods for detecting AGEs are not suitable for screening large populations, or for long-term monitoring. This paper introduces a portable autofluorescence detection system that measures [...] Read more.

Advanced glycation end-products (AGEs) are complex compounds closely associated with several chronic diseases, especially diabetes mellitus (DM). Current methods for detecting AGEs are not suitable for screening large populations, or for long-term monitoring. This paper introduces a portable autofluorescence detection system that measures the concentration of AGEs in the skin based on the fluorescence characteristics of AGEs in biological tissues. The system employs a 395 nm laser LED to excite the fluorescence of AGEs, and uses a photodetector to capture the fluorescence intensity. A model correlating fluorescence intensity with AGEs concentration facilitates the detection of AGEs levels. To account for the variation in optical properties of different individuals’ skin, the system includes a 520 nm light source for calibration. The system features a compact design, measuring only 60 mm × 50 mm × 20 mm, and is equipped with a miniature STM32 module for control and a battery for extended operation, making it easy for subjects to wear. To validate the system’s effectiveness, it was tested on 14 volunteers to examine the correlation between AGEs and glycated hemoglobin, revealing a correlation coefficient of 0.49. Additionally, long-term monitoring of AGEs’ fluorescence and blood sugar levels showed a correlation trend exceeding 0.95, indicating that AGEs reflect changes in blood sugar levels to some extent. Further, by constructing a multivariate predictive model, the study also found that AGEs levels are correlated with age, BMI, gender, and a physical activity index, providing new insights for predicting AGEs content and blood sugar levels. This research supports the early diagnosis and treatment of chronic diseases such as diabetes, and offers a potentially useful tool for future clinical applications. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

14 pages, 2740 KB

Open AccessArticle

Raman Spectroscopy of Disperse Systems with Varying Particle Sizes and Correction of Signal Losses

by Erik Spoor, Viktoria Oerke, Matthias Rädle and Jens-Uwe Repke

Sensors 2024, 24(10), 3132; https://doi.org/10.3390/s24103132 - 15 May 2024

Cited by 2 | Viewed by 1630

Abstract

In this paper, a dispersion of glass beads of different sizes in an ammonium nitrate solution is investigated with the aid of Raman spectroscopy. The signal losses caused by the dispersion are quantified by an additional scattered light measurement and used to correct [...] Read more.

In this paper, a dispersion of glass beads of different sizes in an ammonium nitrate solution is investigated with the aid of Raman spectroscopy. The signal losses caused by the dispersion are quantified by an additional scattered light measurement and used to correct the measured ammonium nitrate concentration. Each individual glass bead represents an interface at which the excitation laser is deflected from its direction causing distortion in the received Raman signal. It is shown that the scattering losses measured with the scattered light probe correlate with the loss of the Raman signal, which means that the data obtained can be used to correct the measured values. The resulting correction function considers different particle sizes in the range of 2–99 µm as well as ammonium nitrate concentrations of 0–20 wt% and delivers an RMSEP of 1.952 wt%. This correction provides easier process access to dispersions that were previously difficult or impossible to measure. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

16 pages, 9595 KB

Open AccessArticle

Impact of Reducing Statistically Small Population Sampling on Threshold Detection in FBG Optical Sensing

by Gabriel Cibira, Ivan Glesk, Jozef Dubovan and Daniel Benedikovič

Sensors 2024, 24(7), 2285; https://doi.org/10.3390/s24072285 - 3 Apr 2024

Cited by 3 | Viewed by 1505

Abstract

Many techniques have been studied for recovering information from shared media such as optical fiber that carries different types of communication, sensing, and data streaming. This article focuses on a simple method for retrieving the targeted information with the least necessary number of [...] Read more.

Many techniques have been studied for recovering information from shared media such as optical fiber that carries different types of communication, sensing, and data streaming. This article focuses on a simple method for retrieving the targeted information with the least necessary number of significant samples when using statistical population sampling. Here, the focus is on the statistical denoising and detection of the fiber Bragg grating (FBG) power spectra. The impact of the two-sided and one-sided sliding window technique is investigated. The size of the window is varied up to one-half of the symmetrical FBG power spectra bandwidth. Both, two- and one-sided small population sampling techniques were experimentally investigated. We found that the shorter sliding window delivered less processing latency, which would benefit real-time applications. The calculated detection thresholds were used for in-depth analysis of the data we obtained. It was found that the normality three-sigma rule does not need to be followed when a small population sampling is used. Experimental demonstrations and analyses also showed that novel denoising and statistical threshold detection do not depend on prior knowledge of the probability distribution functions that describe the FBG power spectra peaks and background noise. We have demonstrated that the detection thresholds’ adaptability strongly depends on the mean and standard deviation values of the small population sampling. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

22 pages, 15334 KB

Open AccessArticle

Raw Spectral Filter Array Imaging for Scene Recognition

by Hassan Askary, Jon Yngve Hardeberg and Jean-Baptiste Thomas

Sensors 2024, 24(6), 1961; https://doi.org/10.3390/s24061961 - 19 Mar 2024

Cited by 1 | Viewed by 1535

Abstract

Scene recognition is the task of identifying the environment shown in an image. Spectral filter array cameras allow for fast capture of multispectral images. Scene recognition in multispectral images is usually performed after demosaicing the raw image. Along with adding latency, this makes [...] Read more.

Scene recognition is the task of identifying the environment shown in an image. Spectral filter array cameras allow for fast capture of multispectral images. Scene recognition in multispectral images is usually performed after demosaicing the raw image. Along with adding latency, this makes the classification algorithm limited by the artifacts produced by the demosaicing process. This work explores scene recognition performed on raw spectral filter array images using convolutional neural networks. For this purpose, a new raw image dataset is collected for scene recognition with a spectral filter array camera. The classification is performed using a model constructed based on the pretrained Places-CNN. This model utilizes all nine channels of spectral information in the images. A label mapping scheme is also applied to classify the new dataset. Experiments are conducted with different pre-processing steps applied on the raw images and the results are compared. Higher-resolution images are found to perform better even if they contain mosaic patterns. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

13 pages, 4532 KB

Open AccessArticle

A Microlens Array Grating for Miniature Multi-Channel Spectrometers

by Shuonan Shan, Jingwen Li, Peiyuan Liu, Qiaolin Li, Xiaohao Wang and Xinghui Li

Sensors 2023, 23(20), 8381; https://doi.org/10.3390/s23208381 - 11 Oct 2023

Cited by 10 | Viewed by 3562

Abstract

Most existing multi-channel spectrometers are constructed by physically stacking single-channel spectrometers, resulting in their large size, high weight, and limited number of channels. Therefore, their miniaturization is urgently needed. In this paper, a microlens array grating is designed for miniature multi-channel spectrometers. A [...] Read more.

Most existing multi-channel spectrometers are constructed by physically stacking single-channel spectrometers, resulting in their large size, high weight, and limited number of channels. Therefore, their miniaturization is urgently needed. In this paper, a microlens array grating is designed for miniature multi-channel spectrometers. A transmissive element integrating microlens arrays and gratings, the MLAG, enables simultaneous focusing and dispersion. Using soft lithography, the MLAG was fabricated with a deviation of less than 2.2%. The dimensions are 10 mm × 10 mm × 4 mm with over 2000 available units. The MLAG spectrometer operates in the 400–700 nm wavelength range with a resolution of 6 nm. Additionally, the designed MLAG multi-channel spectrometer is experimentally verified to have independently valid cells that can be used in multichannel spectrometers. The wavelength position repeatability deviation of each cell is about 0.5 nm, and the repeatability of displacement measurements by the chromatic confocal sensor with the designed MLAG multi-channel spectrometer is less than 0.5 μm. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

12 pages, 18467 KB

Open AccessArticle

Gaussian Decomposition vs. Semiclassical Quantum Simulation: Obtaining the High-Order Derivatives of a Spectrum in the Case of Photosynthetic Pigment Optical Properties Studying

by Andrei P. Razjivin, Vladimir S. Kozlovsky, Aleksandr A. Ashikhmin and Roman Y. Pishchalnikov

Sensors 2023, 23(19), 8248; https://doi.org/10.3390/s23198248 - 5 Oct 2023

Cited by 1 | Viewed by 1486

Abstract

In this paper, a procedure for obtaining undistorted high derivatives (up to the eighth order) of the optical absorption spectra of biomolecule pigments has been developed. To assess the effectiveness of the procedure, the theoretical spectra of bacteriochlorophyll a, chlorophyll a, [...] Read more.

In this paper, a procedure for obtaining undistorted high derivatives (up to the eighth order) of the optical absorption spectra of biomolecule pigments has been developed. To assess the effectiveness of the procedure, the theoretical spectra of bacteriochlorophyll a, chlorophyll a, spheroidene, and spheroidenone were simulated by fitting the experimental spectra using the differential evolution algorithm. The experimental spectra were also approximated using sets of Gaussians to calculate the model absorption spectra. Theoretical and model spectra can be differentiated without smoothing (high-frequency noise filtering) to obtain high derivatives. Superimposition of the noise track on the model spectra allows us to obtain test spectra similar to the experimental ones. Comparison of the high derivatives of the model spectra with those of the test spectra allows us to find the optimal parameters of the filter, the application of which leads to minimal differences between the high derivatives of the model and test spectra. For all four studied pigments, it was shown that smoothing the experimental spectra with optimal filters makes it possible to obtain the eighth derivatives of the experimental spectra, which were close to the eighth derivatives of their theoretical spectra. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

23 pages, 11783 KB

Open AccessArticle

A Spectral Encoding Simulator for Broadband Active Illumination and Reconstruction-Based Spectral Measurement

by Peng Jiang, Xiaoxu Wang, Zihui Zhang, Guochao Gu, Jifeng Li, Heng Wu, Limin He and Guanyu Lin

Sensors 2023, 23(10), 4608; https://doi.org/10.3390/s23104608 - 10 May 2023

Viewed by 2625

Abstract

Spectral reflectance or transmittance measurements provide intrinsic information on the material of an object and are widely used in remote sensing, agriculture, diagnostic medicine, etc. Most reconstruction-based spectral reflectance or transmittance measurement methods based on broadband active illumination use narrow-band LEDs or lamps [...] Read more.

Spectral reflectance or transmittance measurements provide intrinsic information on the material of an object and are widely used in remote sensing, agriculture, diagnostic medicine, etc. Most reconstruction-based spectral reflectance or transmittance measurement methods based on broadband active illumination use narrow-band LEDs or lamps combined with specific filters as spectral encoding light sources. These light sources cannot achieve the designed spectral encoding with a high resolution and accuracy due to their low degree of freedom for adjustment, leading to inaccurate spectral measurements. To address this issue, we designed a spectral encoding simulator for active illumination. The simulator is composed of a prismatic spectral imaging system and a digital micromirror device. The spectral wavelengths and intensity are adjusted by switching the micromirrors. We used it to simulate spectral encodings according to the spectral distribution on micromirrors and solved the DMD patterns corresponding to the spectral encodings with a convex optimization algorithm. To verify the applicability of the simulator for spectral measurements based on active illumination, we used it to numerically simulate existing spectral encodings. We also numerically simulated a high-resolution Gaussian random measurement encoding for compressed sensing and measured the spectral reflectance of one vegetation type and two minerals through numerical simulations. We reconstructed the spectral transmittance of a calibrated filter through an experiment. The results show that the simulator can measure the spectral reflectance or transmittance with a high resolution and accuracy. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

11 pages, 2135 KB

Open AccessArticle

Performance-Enhanced Static Modulated Fourier Transform Spectrometer with a Spectral Reconstruction

by Ju Yong Cho, Seunghoon Lee and Won Kweon Jang

Sensors 2023, 23(5), 2603; https://doi.org/10.3390/s23052603 - 27 Feb 2023

Cited by 1 | Viewed by 2200

Abstract

A static modulated Fourier transform spectrometer has been noted to be a compact and fast evaluation tool for spectroscopic inspection, and many novel structures have been reported to support its performance. However, it still suffers from poor spectral resolution due to the limited [...] Read more.

A static modulated Fourier transform spectrometer has been noted to be a compact and fast evaluation tool for spectroscopic inspection, and many novel structures have been reported to support its performance. However, it still suffers from poor spectral resolution due to the limited sampling data points, which marks its intrinsic drawback. In this paper, we outline the enhanced performance of a static modulated Fourier transform spectrometer with a spectral reconstruction method that can compensate for the insufficient data points. An enhanced spectrum can be reconstructed by applying a linear regression method to a measured interferogram. We obtain the transfer function of a spectrometer by analyzing what interferogram can be detected with different values of parameters such as focal length of the Fourier lens, mirror displacement, and wavenumber range, instead of direct measurement of the transfer function. Additionally, the optimal experimental conditions for the narrowest spectral width are investigated. Application of the spectral reconstruction method achieves an improved spectral resolution from 74 cm⁻¹ when spectral reconstruction is not applied to 8.9 cm⁻¹, and a narrowed spectral width from 414 cm⁻¹ to 371 cm⁻¹, which are close to the values of the spectral reference. In conclusion, the spectral reconstruction method in a compact static modulated Fourier transform spectrometer effectively enhances its performance without any additional optic in the structure. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

18 pages, 6151 KB

Open AccessArticle

Design and Research of Chromatic Confocal System for Parallel Non-Coaxial Illumination Based on Optical Fiber Bundle

by Yali Zhang, Qing Yu, Chong Wang, Yaozu Zhang, Fang Cheng, Yin Wang, Tianliang Lin, Ting Liu and Lin Xi

Sensors 2022, 22(24), 9596; https://doi.org/10.3390/s22249596 - 7 Dec 2022

Cited by 4 | Viewed by 4095

Abstract

Conventional chromatic confocal systems are mostly single-point coaxial illumination systems with a low signal-to-noise ratio, light energy utility and measurement efficiency. To overcome the above shortcomings, we propose a parallel non-coaxial-illumination chromatic-confocal-measurement system based on an optical fiber bundle. Based on the existing [...] Read more.

Conventional chromatic confocal systems are mostly single-point coaxial illumination systems with a low signal-to-noise ratio, light energy utility and measurement efficiency. To overcome the above shortcomings, we propose a parallel non-coaxial-illumination chromatic-confocal-measurement system based on an optical fiber bundle. Based on the existing single-point non-coaxial-illumination system, the optical fiber bundle is used as the optical beam splitter to achieve parallel measurements. Thus, the system can yield measurements through line scanning, which greatly improves measurement efficiency. To verify the measurement performance of the system, based on the calibration experiment, the system realizes the measurement of the height of the step, the thickness of the transparent specimen and the reconstruction of the three-dimensional topography of the surface of the step and coin. The experimental results show that the measuring range of the system is 200 μm. The measurement accurcy can reach micron level, and the system can realize a good three-dimensional topography reconstruction effect. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

25 pages, 9952 KB

Open AccessArticle

Characterisation and Quenching Correction for an Al₂O₃:C Optical Fibre Real Time System in Therapeutic Proton, Helium, and Carbon-Charged Beams

by Luana de Freitas Nascimento, Paul Leblans, Brent van der Heyden, Mark Akselrod, Jo Goossens, Luis Enrique Correa Rocha, Ana Vaniqui and Dirk Verellen

Sensors 2022, 22(23), 9178; https://doi.org/10.3390/s22239178 - 25 Nov 2022

Cited by 10 | Viewed by 2100

Abstract

Real time radioluminescence fibre-based detectors were investigated for application in proton, helium, and carbon therapy dosimetry. The Al₂O₃:C probes are made of one single crystal (1 mm) and two droplets of micro powder in two sizes (38 μm and [...] Read more.

Real time radioluminescence fibre-based detectors were investigated for application in proton, helium, and carbon therapy dosimetry. The Al₂O₃:C probes are made of one single crystal (1 mm) and two droplets of micro powder in two sizes (38 μm and 4 μm) mixed with a water-equivalent binder. The fibres were irradiated behind different thicknesses of solid slabs, and the Bragg curves presented a quenching effect attributed to the nonlinear response of the radioluminescence (RL) signal as a function of linear energy transfer (LET). Experimental data and Monte Carlo simulations were utilised to acquire a quenching correction method, adapted from Birks’ formulation, to restore the linear dose–response for particle therapy beams. The method for quenching correction was applied and yielded the best results for the ‘4 μm’ optical fibre probe, with an agreement at the Bragg peak of 1.4% (160 MeV), and 1.5% (230 MeV) for proton-charged particles; 2.4% (150 MeV/u) for helium-charged particles and of 4.8% (290 MeV/u) and 2.9% (400 MeV/u) for the carbon-charged particles. The most substantial deviations for the ‘4 μm’ optical fibre probe were found at the falloff regions, with ~3% (protons), ~5% (helium) and 6% (carbon). Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

13 pages, 5764 KB

Open AccessArticle

Location of Latent Forensic Traces Using Multispectral Bands

by Samuel Miralles-Mosquera, Bernardo Alarcos and Alfredo Gardel

Sensors 2022, 22(23), 9142; https://doi.org/10.3390/s22239142 - 25 Nov 2022

Cited by 2 | Viewed by 3192

Abstract

In this paper, a conventional camera modified to capture multispectral images, has been used to locate latent forensic traces with a smart combination of wavelength filters, capturing angle, and illumination sources. There are commercial multispectral capture devices adapted to the specific tasks of [...] Read more.

In this paper, a conventional camera modified to capture multispectral images, has been used to locate latent forensic traces with a smart combination of wavelength filters, capturing angle, and illumination sources. There are commercial multispectral capture devices adapted to the specific tasks of the police, but due to their high cost and operation not well adapted to the field work in a crime scene, they are not currently used by forensic units. In our work, we have used a digital SLR camera modified to obtain a nominal sensitivity beyond the visible spectrum. The goal is to obtain forensic evidences from a crime scene using the multispectral camera by an expert in the field knowing which wavelength filters and correct illumination sources should be used, making visible latent evidences hidden from the human-eye. In this paper, we show a procedure to retrieve from latent forensic traces, showing the validity of the system in different real cases (blood stains, hidden/erased tattoos, unlocking patterns on mobile devices). This work opens the possibility of applying multispectral inspections in the forensic field specially for operational units for the location of latent through non-invasive optical procedures. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

14 pages, 4718 KB

Open AccessArticle

Development and Validation of a Tunable Diode Laser Absorption Spectroscopy System for Hot Gas Flow and Small-Scale Flame Measurement

by Ran Tu, Junqing Gu, Yi Zeng, Xuejin Zhou, Kai Yang, Jiaojiao Jing, Zhihong Miao and Jianhong Yang

Sensors 2022, 22(17), 6707; https://doi.org/10.3390/s22176707 - 5 Sep 2022

Cited by 12 | Viewed by 4713

Abstract

TDLAS (tunable diode laser absorption spectroscopy) is an important gas analysis method that can be employed to obtain characteristic parameters non-invasively by the infrared absorption spectra of tracer molecules such as CH₄, H₂O and O₂. In this [...] Read more.

TDLAS (tunable diode laser absorption spectroscopy) is an important gas analysis method that can be employed to obtain characteristic parameters non-invasively by the infrared absorption spectra of tracer molecules such as CH₄, H₂O and O₂. In this study, a portable H₂O-based TDLAS system with a dual optical path was developed with the aim of assessing the combustion characteristics of flammable gases. Firstly, a calculation method of gas characteristics including temperature and velocity combining absorption spectra and a HITRAN database was provided. Secondly, to calibrate and validate this TDLAS system precisely, a pressure vessel and a shock tube were introduced innovatively to generate static or steady flow fields with preset constant temperatures, pressures, or velocities. Static tests within environment pressures up to 2 MPa and steady flow field tests with temperatures up to 1600 K and flow velocities up to 950 m/s were performed for verification. It was proved that this system can provide an accurate values for high temperature and velocity gas flows. Finally, an experimental investigation of CH₄/air flames was conducted to test the effectiveness of the system when applied to small diffusion flames. This TDLAS system gave satisfactory flame temperature and velocity data owing to the dual optical path design and high frequency scanning, which compensated for scale effects and pulsation of the flame. This work demonstrates a valuable new approach to thermal hazard analysis in specific environments. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

18 pages, 4513 KB

Open AccessArticle

Detection of Water pH Using Visible Near-Infrared Spectroscopy and One-Dimensional Convolutional Neural Network

by Dengshan Li and Lina Li

Sensors 2022, 22(15), 5809; https://doi.org/10.3390/s22155809 - 3 Aug 2022

Cited by 23 | Viewed by 4270

Abstract

pH is an important parameter for water quality detection. This study proposed a novel calibration regression strategy based on a one-dimensional convolutional neural network (1D-CNN) for water pH detection using visible near-infrared (Vis-NIR) spectroscopy. Two groups of Vis-NIR spectral analysis experiments of water [...] Read more.

pH is an important parameter for water quality detection. This study proposed a novel calibration regression strategy based on a one-dimensional convolutional neural network (1D-CNN) for water pH detection using visible near-infrared (Vis-NIR) spectroscopy. Two groups of Vis-NIR spectral analysis experiments of water pH detection were employed to evaluate the performance of 1D-CNN. Two conventional multivariate regression calibration methods, including partial least squares (PLS) and least squares support vector machine (LS-SVM), were introduced for comparative analysis with 1D-CNN. The successive projections algorithm (SPA) was adopted to select the feature variables. In addition, the learning mechanism of 1D-CNN was interpreted through visual feature maps by convolutional layers. The results showed that the 1D-CNN models obtained the highest prediction accuracy based on full spectra for the two experiments. For the spectrophotometer experiment, the root mean square error of prediction (RMSEP) was 0.7925, and the determination coefficient of prediction (

R_{p}^{2}

) was 0.8515. For the grating spectrograph experiment, the RMSEP was 0.5128 and the

R_{p}^{2}

was 0.9273. The convolutional layers could automatically preprocess the spectra and effectively extract the spectra features. Compared with the traditional regression methods, 1D-CNN does not need complex spectra pretreatment and variable selection. Therefore, 1D-CNN is a promising regression approach, with higher prediction accuracy and better modeling convenience for rapid water pH detection using Vis-NIR spectroscopy. Full article

(This article belongs to the Special Issue Spectral Detection Technology, Sensors and Instruments)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Spectral Detection Technology, Sensors and Instruments

Share This Special Issue

Special Issue Editors

Special Issue Information

Benefits of Publishing in a Special Issue

Related Special Issue

Published Papers (14 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI