Article
Soil Science
Jiang Liu, Dongxing Zhang, Li Yang, Yuxin Ma, Tao Cui, Xiantao He, Zhaohui Du
Summary: Real-time soil moisture content (SMC) monitoring is important in precision agriculture, and visible and near-infrared (vis-NIR) spectroscopy has been proposed as a promising method for SMC monitoring. However, the prediction models of vis-NIR vary significantly between different soil types. To solve this problem, this study utilizes the External Parameter Orthogonalization (EPO) method to develop a generalized SMC prediction model. Results show the feasibility of using vis-NIR spectroscopy to predict SMC on individual soils and the effectiveness of the EPOPLSR method in developing a generalized prediction model.
Article
Environmental Sciences
Wu Yu, Yongsheng Hong, Songchao Chen, Yiyun Chen, Lianqing Zhou
Summary: Visible and near-infrared (Vis-NIR) spectroscopy can provide a rapid and inexpensive estimation for soil organic carbon (SOC), but external environmental factors can affect the accuracy of the model. The external parameter orthogonalization (EPO) algorithm was investigated to eliminate interference from external parameters, with moisture being the most significant factor. The study showed that different EPO development methods can balance model performance and cost effectively.
Article
Environmental Sciences
Chen Gao, Min Xu, Hanzeyu Xu, Wei Zhou
Summary: This study measured the multi-angle reflectance of soil samples from tidal flats, retrieved soil surface photometric characteristics using the particle swarm optimization algorithm, and calculated soil moisture content by introducing the equivalent water thickness. The results showed a significant correlation between the retrieved equivalent water thickness and soil moisture content, with high precision for estimating soil moisture. The study also found consistency between single scattering albedo values and soil moisture content, and irregular changes in roughness parameter and scattering type when soil samples were soaked with water.
Article
Food Science & Technology
Jian-E Dong, Zhi-Tian Zuo, Ji Zhang, Yuan-Zhong Wang
Summary: The study used a novel digital image method based on 2DCOS and i2DCOS to discriminate the geographical origins of Boletus edulis mushrooms. Results showed that the synchronous 2DCOS spectra model accurately classified samples from different regions with 100% accuracy in both the test and validation sets, demonstrating good discrimination performance. This new analytical method has potential applications in quality control of food, herb, and agricultural products.
Article
Chemistry, Applied
Jiaming Guo, Han Huang, Xiaolong He, Jinwei Cai, Zhixiong Zeng, Chengying Ma, Enli Lue, Qunyu Shen, Yanhua Liu
Summary: This study introduced a method to eliminate the effect of moisture in fresh tea leaves on NIR spectra using external parameter orthogonalization (EPO), and combined it with feature selection algorithm and partial least squares (PLS) prediction model to achieve satisfactory prediction results.
Article
Engineering, Chemical
Haim Kalman
Summary: In this study, the angle of repose, angle of tilting, and Hausner ratio were used to estimate the flowability of particulate materials. The angle of tilting was found to be most appropriate for describing a material's response to moisture content. Three types of materials were identified based on their response to moisture content: solid particles, porous particles, and size-changing particles. A model predicting flowability based on moisture content and Archimedes number was developed, with further experiments needed to relate the model to other materials.
Article
Chemistry, Analytical
Arian Amirvaresi, Hadi Parastar
Summary: A new method based on EPO-SVM was proposed for processing ATR-FT-MIR spectra to solve the authentication problem in saffron. The EPO-SVM approach demonstrated higher classification accuracy in detecting plant adulterants compared to other methods, showing good performance in validation.
ANALYTICA CHIMICA ACTA
(2021)
Article
Geochemistry & Geophysics
Bikram Koirala, Zohreh Zahiri, Paul Scheunders
Summary: This study proposes a supervised methodology to accurately estimate the soil moisture content from spectral reflectance. By using a proxy that is invariant to illumination, viewing angle, and sensor type, the method can normalize the proxy and directly estimate the soil moisture content. Experimental results demonstrate the accuracy of the proposed method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Plant Sciences
Ying Huang
Summary: Accurately predicting soil moisture content in tea plantations is crucial for optimizing irrigation practices and improving crop productivity. Traditional methods are costly and labor-intensive, while machine learning models are limited by insufficient data. To address these challenges, an improved SVM-based model was developed, incorporating novel features and enhancing performance with the BES method for hyper-parameter optimization. Experimental results demonstrated its superior performance compared to traditional SVM approaches and other algorithms.
Article
Environmental Sciences
Hayfa Zayani, Youssef Fouad, Didier Michot, Zeineb Kassouk, Nicolas Baghdadi, Emmanuelle Vaudour, Zohra Lili-Chabaane, Christian Walter
Summary: Understanding the spatial and temporal variability in soil organic carbon (SOC) content is important for assessing soil fertility as well as related parameters. This study evaluated the combination of remote sensing time series with laboratory spectral measurements using machine and deep-learning algorithms to improve predictions of SOC content. Different models and approaches were utilized, and the use of additional information such as soil moisture and laboratory indices improved model performance
Article
Chemistry, Physical
Jinwei Yao, Hui Song, Yizhan Yang
Summary: This work investigated the correlation between the physical and mechanical parameters of concrete in marine environments during corrosion. Concrete specimens with different water-cement ratios were immersed in different simulated seawater solutions, and various physical and mechanical parameters were measured. The results showed a positive correlation between compressive strength, tensile strength, and mass, while the correlation with wave velocity was found to be poor.
Article
Chemistry, Analytical
Mohammed Kamruzzaman, Dipsikha Kalita, Toukir Md. Ahmed, Gamal ElMasry, Yoshio Makino
Summary: Variable selection is a critical step in designing a dedicated multispectral real-time system. This study compares the effects of different algorithms on predicting moisture content in red meat and corn, and finds that the competitive adaptive reweighted sampling-partial least squares regression (CARS-PLSR) model performs the best.
ANALYTICA CHIMICA ACTA
(2022)
Article
Instruments & Instrumentation
Lianxiang Gong, Chenxi Zhu, Yifeng Luo, Xiaping Fu
Summary: This study reconstructed spectra from RGB images taken by smartphones using a BP neural network and pseudo-inverse method, and established a prediction model for chlorophyll content estimation. The results showed a high correlation between predicted and measured values, indicating the potential of real-time measurement of chlorophyll content using spectral reflectance reconstruction based on RGB images.
APPLIED SPECTROSCOPY
(2023)
Article
Engineering, Environmental
Ya-Ping Qi, Ping-Jing He, Dong-Ying Lan, Hao-Yang Xian, Fan Lu, Hua Zhang
Summary: In this study, a rapid and accurate method for predicting the moisture content of solid waste using ATR-FTIR and multiple machine learning methods was developed. By discussing 20 combinations of different spectral preprocessing methods and regression algorithms, a combined model with high accuracy in moisture content prediction was obtained. The experimental results demonstrated that this method can effectively and accurately measure the moisture content of solid waste.
Article
Agricultural Engineering
Yuzhen Wei, Xiaoli Li, Yong He
Summary: This study investigated the generalisation of tea moisture content detection models across different leaf surface orientations and tea varieties. The results showed improved prediction performance through fractional order differential treatments. The study advances tea moisture detection based on VNIR spectra and the ability to generalise across spectra with different characteristics.
BIOSYSTEMS ENGINEERING
(2021)
Article
Environmental Sciences
Jie Liu, Aiai Xu, Changkun Wang, Zhiying Guo, Shiwen Wu, Kai Pan, Fangfang Zhang, Xianzhang Pan
LAND DEGRADATION & DEVELOPMENT
(2020)
Article
Soil Science
Ya Liu, Changkun Wang, Chenchao Xiao, Kun Shang, Yan Zhang, Xianzhang Pan
Summary: In estimating soil fertility properties, using VisNIR spectroscopy alone performed better than combining VisNIR and PXRF, as well as using PXRF alone. VisNIR was found to be the optimal method for rapidly obtaining soil information for the precision application of organic and NPK fertilizers, while PXRF did not perform well on certain soil fertility properties such as SOC.
SOIL SCIENCE SOCIETY OF AMERICA JOURNAL
(2021)
Article
Biotechnology & Applied Microbiology
Aiai Xu, Jie Liu, Zhiying Guo, Changkun Wang, Kai Pan, Fangfang Zhang, Xianzhang Pan
Summary: The study found that various land uses and environmental conditions have different effects on the assembly processes and determinants of soil microbial communities. Predicting the responses of soil microbes to environmental changes requires considering the characteristics of different terrains.
APPLIED AND ENVIRONMENTAL MICROBIOLOGY
(2021)
Article
Microbiology
Jie Liu, Changkun Wang, Zhiying Guo, Ya Liu, Kai Pan, Aiai Xu, Fangfang Zhang, Xianzhang Pan
Summary: Studies on biodiversity-productivity relationships for soil microorganisms using molecular ecology methods have shown that soil bacterial communities are strongly impacted by productivity. Soil bacterial biodiversity increases rapidly with productivity at low levels and then stabilizes, with mechanisms possibly being ecosystem resource availability at large-scale regions and species competition in local regions. These results enhance our understanding of the linkage between belowground microorganisms and aboveground vegetation in arid and semi-arid areas and the value of satellite-derived datasets in research on soil microbial diversity at large spatial scales.
ENVIRONMENTAL MICROBIOLOGY
(2021)
Article
Engineering, Civil
Zhou Zhu, Haifeng Zhao, Fang Hui, Yan Zhang
Summary: This paper addresses the issue of online updating of visual object tracker for car sharing services by adjusting the updating rate adaptively according to the tracking performance of the current frame and assigning weights based on the accuracy and continuity of the tracking results. The proposed method embeds a model update strategy in a lightweight baseline tracker and utilizes an IOU predictor to guide the adjustment of updating weights online, which improves the accuracy of object tracking compared to the baseline tracker, despite the inevitable IOU prediction errors.
JOURNAL OF ADVANCED TRANSPORTATION
(2021)
Article
Soil Science
Fangfang Zhang, Shiwen Wu, Jie Liu, Changkun Wang, Zhiying Guo, Aiai Xu, Kai Pan, Xianzhang Pan
Summary: This study used simulation experiments to acquire hyperspectral data with different vegetation coverages, soil moisture contents, and soil types, and found that the 1D-CNN and LSTM deep learning models performed well in predicting soil moisture content. The results showed that when only bare soil spectra were used, the prediction accuracy was similar for 1D-CNN, LSTM, and PLSR models, indicating that deep learning had no advantage on smaller datasets.
SOIL SCIENCE SOCIETY OF AMERICA JOURNAL
(2021)
Article
Soil Science
Aiai Xu, Jie Liu, Zhiying Guo, Changkun Wang, Kai Pan, Fangfang Zhang, Xianzhang Pan
Summary: The study revealed that in the agro-pastoral ecotone, land-use types have a significant impact on soil microbial communities, especially over longer durations, highlighting the susceptibility of community composition to land-use types under bidirectional conversions.
Article
Environmental Sciences
Fangfang Zhang, Changkun Wang, Kai Pan, Zhiying Guo, Jie Liu, Aiai Xu, Haiyi Ma, Xianzhang Pan
Summary: In this study, a one-dimensional convolutional neural network model was established to simultaneously estimate soil properties and vegetation coverage. The model achieved high prediction accuracy for both soil properties and vegetation coverage. Compared to the traditional partial least-squares regression method, the model showed improved prediction accuracy.
Article
Agronomy
Yanli Li, Hai Zhu, Jifu Li, Huifang Jin, Bilin Lu, Xianzhang Pan
Summary: The study compared five common winter wheat RLD distribution models under soil salt stress and proposed the log-normal model as the most suitable one, which can better fit the observed RLD distribution and have practical application in saline soils.
Article
Environmental Sciences
Jie Liu, Changkun Wang, Zhiying Guo, Aiai Xu, Kai Pan, Xianzhang Pan
Summary: Climate factors have a distinct impact on soil microbial community in arid and semiarid desert areas, but after intensive agriculture, the effects of climate on soil microbial diversity are weakened. The relationships between microbial richness and climate factors differ in different land use types, with fewer significant correlations observed in croplands.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Soil Science
Aiai Xu, Zhiying Guo, Kai Pan, Changkun Wang, Fangfang Zhang, Jie Liu, Xianzhang Pan
Summary: This study examines soil microbial assembly processes and co-occurrence patterns in the agro-pastoral ecotone of northern China. It finds that land-use durations have a significant impact on soil microbial community assembly, with longer durations exhibiting stronger determinism and increased network complexity and stability. These findings improve our understanding of the mechanisms maintaining soil microbial diversity in ecotones.
APPLIED SOIL ECOLOGY
(2023)
Article
Computer Science, Information Systems
Songsong Wu, Xiao-Yuan Jing, Qinghua Zhang, Fei Wu, Haifeng Zhao, Yuning Dong
Article
Spectroscopy
Wu Shi-wen, Wang Chang-kun, Liu Ya, Li Yan-li, Liu Jie, Xu Ai-ai, Pan Kai, Li Yi-chun, Zhang Fang-fang, Pan Xian-zhang
SPECTROSCOPY AND SPECTRAL ANALYSIS
(2019)