Article
Instruments & Instrumentation
Lei Pang, Lianming Wang, Peng Yuan, Lei Yan, Jiang Xiao
Summary: This research successfully predicts the seed viability of Sophora japonica by combining hyperspectral imaging technology with deep learning methods. One-dimensional deep learning methods outperformed machine learning methods in identifying seed viability. By improving the successive projection algorithm and comparing it with other algorithms, the best detection model was found. The vitality prediction model established based on this model had better performance than the detection model.
INFRARED PHYSICS & TECHNOLOGY
(2022)
Article
Geosciences, Multidisciplinary
E. U. Eyo, S. J. Abbey, T. T. Lawrence, F. K. Tetteh
Summary: This research utilizes advanced algorithms and techniques to propose a more accurate method for estimating soil expansion by analyzing the influence of different factors and identifying key features affecting soil behavior. The study shows that meta-heuristic models exhibit the best overall performance.
GEOSCIENCE FRONTIERS
(2022)
Article
Computer Science, Information Systems
Alexey A. Klokov, Di Un Pak, Aleksandr Khorin, Dmitry A. Yudin, Leon Kochiev, Vladimir D. Luchinskiy, Vitaly D. Bezuglyj
Summary: This article introduces a novel approach called DAPS3D to train deep neural networks for 3D semantic segmentation using LiDAR point clouds. The approach is based on a spherical projection of a point cloud and LiDAR-specific masks, enabling the model to adapt to different types of LiDAR. Various high-speed multi-scale spherical projection segmentation models and original augmentations for spherical projections of LiDAR data are proposed. The article also introduces a new method to generate synthetic datasets for domain adaptation problems.
Article
Economics
Honggang Zhang, Zhiyuan Liu, Jian Wang, Yunchi Wu
Summary: This paper proposes the successive over relaxation (SOR) iteration method for flow update in traffic assignment algorithms as a more efficient replacement for the Jacobi and Gauss-Seidel methods. The newly developed GP-SOR method is applied to the user equilibrium traffic assignment problem (UE-TAP) based on the path-based gradient projection (GP) algorithm. Empirical tests demonstrate the need for improvement, which is addressed by employing the Armijo rule to determine the relaxation factor and improving the convergence of the GP-SOR algorithm. Preliminary numerical examples show that GP-SOR algorithm has faster convergence and reduces computing time and number of iterations compared to existing alternatives.
TRANSPORTATION RESEARCH PART E-LOGISTICS AND TRANSPORTATION REVIEW
(2023)
Article
Geochemistry & Geophysics
Xuyang Bai, Shurun Tan
Summary: This article proposes a novel physics-embedded artificial neural network (P-ANN) inversion algorithm for retrieving the vertical distribution of soil moisture and temperature using multichannel passive microwave observations. The P-ANN approach outperforms traditional optimization algorithms and conventional neural network approaches in dealing with layered soil retrieval. The proposed approach holds great potential for remote sensing applications and solving inverse problem challenges.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Miriam Biel-Maeso, Victoria Burke, Janek Greskowiak, Gudrun Massmann, Pablo A. Lara-Martin, Carmen Corada-Fernandez
Summary: In this study, laboratory column experiments were conducted to investigate the transport of pharmaceuticals and artificial sweeteners in two soils under saturated conditions. The results showed different behavior of the compounds, with some reaching maximum concentrations at the outlet and others showing retardation due to sorption. Additionally, vertical concentration profiles varied between soils with different clay content.
SCIENCE OF THE TOTAL ENVIRONMENT
(2021)
Review
Chemistry, Inorganic & Nuclear
Taiwo W. Quadri, Lukman O. Olasunkanmi, Omolola E. Fayemi, Ekemini D. Akpan, Chandrabhan Verma, El-Sayed M. Sherif, Khaled F. Khaled, Eno E. Ebenso
Summary: It has been established that organic corrosion inhibitors form protective films on metal surfaces through coordination bonding, with quantitative structure activity relationship (QSAR) being a recent and reliable computational method to describe these metal-inhibitor interactions. Research interest in the relationship between molecular structure and corrosion inhibition has increased, with advanced techniques like machine learning, artificial neural network (ANN), support vector machine (SVM), and genetic function approximation (GFA) being utilized in this field.
COORDINATION CHEMISTRY REVIEWS
(2021)
Article
Engineering, Environmental
Jiazhi Yang, Xuejun Long, Xiaonan Feng, Jun Wan
Summary: A novel composite material, kaolin/lanthanum carbonate (KLC), was studied for the simultaneous sorption of orthophosphate and phosphonate. It was found that orthophosphate and phosphonate competed for sorption on the surface of KLC. The maximum sorption capacities of both phosphorus species were achieved at pH 4.9, and the impact of ionic strength on sorption capacity was negligible. An artificial neural network (ANN) model was developed and successfully predicted the removal efficiency of orthophosphate and phosphonate.
JOURNAL OF ENVIRONMENTAL CHEMICAL ENGINEERING
(2023)
Article
Engineering, Civil
Guangjian Xiang, Deshun Yin, Chenxi Cao, Lili Yuan
Summary: This study investigated the impact of vibrations on the flow characteristics of soft soil and developed an artificial neural network model to predict soil viscosity. High frequency vibrations and low confining pressure were found to enhance soil flow ability, and the ANN model with genetic algorithm optimization showed high accuracy and reliability in predicting soil viscosity.
GEOMECHANICS AND ENGINEERING
(2021)
Article
Optics
Xuexiao Ma, Jiaqiang Lin, Chuansheng Dai, Jialiang Lv, Peijun Yao, Lixin Xu, Chun Gu
Summary: This paper proposes a new approach to analyze the mode-locked performance of fiber lasers using machine learning technology. By using artificial neural networks to judge pulse evolution and predict pulse shape, and combining the genetic algorithm, the authors demonstrate how to compute the parameters of fiber lasers.
OPTICS AND LASER TECHNOLOGY
(2022)
Review
Physics, Applied
Benjamin Midtvedt, Saga Helgadottir, Aykut Argun, Jesus Pineda, Daniel Midtvedt, Giovanni Volpe
Summary: Deep learning has great potential for quantitative analysis in digital microscopy, but its application is still underutilized due to the steep learning curve involved in developing custom solutions.
APPLIED PHYSICS REVIEWS
(2021)
Review
Geosciences, Multidisciplinary
Abolfazl Baghbani, Tanveer Choudhury, Susanga Costa, Johannes Reiner
Summary: This study reviewed the application of artificial intelligence methods in geotechnical engineering and identified nine prominent areas. Artificial Neural Network (ANN) emerged as the most widely used AI method. The analysis shows that the success and accuracy of AI applications depends on the number and type of datasets and selection of input parameters.
EARTH-SCIENCE REVIEWS
(2022)
Article
Materials Science, Multidisciplinary
Claudia Gorynski, Max Frei, Frank Einar Kruis, Markus Winterer
Summary: Using region-based convolutional neural networks (R-CNNs), this study presents a novel approach to accurately quantify various microstructural characteristics of polycrystalline microstructures, such as Feret diameter, axis length, area, circumference, dihedral angle, and coordination number. A two-step method is employed, involving a semi-automatic annotation tool for generating training data and a fully automated R-CNN for quantitative evaluation of images. The R-CNN performs well in evaluating grain size characteristics, even in images with low contrast, and can differentiate between uni- and bimodal grain size distributions on the sub-micron and nanoscale. The innovative solution makes grain size measuring more accessible, time-effective, less biased, consistent, and statistically more precise.
Article
Environmental Sciences
Xiaokai Chen, Fenling Li, Qingrui Chang
Summary: In this study, a efficient method for estimating winter wheat PNC was established by combining continuous wavelet transform (CWT), successive projection algorithm (SPA), and partial least square (PLS) regression. The results showed that the PNC estimation model based on CWT (CWT-Scale6-SPA-PLS) accurately predicted PNC, providing a new technical method for precision nitrogen management.
Article
Computer Science, Artificial Intelligence
Saban Gulcu
Summary: The training algorithm is a crucial component of artificial neural networks (ANN) that affects their performance. This article presents a new hybrid algorithm called DA-MLP, which uses the dragonfly algorithm to train feed-forward multilayer neural networks (MLP). The experimental study shows that the DA-MLP algorithm is more efficient than other algorithms.
APPLIED SOFT COMPUTING
(2022)
Article
Instruments & Instrumentation
Stefani I. E. Andrade, Roberto K. H. Galvao, Mario C. U. Araujo, Sillas Hadjiloucas
Summary: Real-time observation of water and oil diffusion dynamics using video images, accurate localization of the liquid front achieved through Savitzky-Golay derivative filtering. System identification shows that a fractional-order model with only two parameters can describe the diffusion process dynamics. This method opens the way for video-based cellulose filter paper microfluidics and lateral flow assays, which are cost-effective and have broad diagnostic applications.
REVIEW OF SCIENTIFIC INSTRUMENTS
(2021)
Article
Chemistry, Analytical
Elainy Virginia dos Santos Pereira, David Douglas de Sousa Fernandes, Mario Cesar Ugulino de Araujo, Paulo Henrique Goncalves Dias Diniz, Maria Ines Sucupira Maciel
Summary: This study demonstrated the feasibility of using a miniaturized NIR spectrophotometer in conjunction with the iSPA-PLS-DA method for on-site authentication of goat milk, ultimately proving the high accuracy and cost-effectiveness of this approach in identifying raw goat milk.
MICROCHEMICAL JOURNAL
(2021)
Article
Chemistry, Applied
Valber Elias de Almeida, David Douglas de Sousa Fernandes, Paulo Henrique Goncalves Dias Diniz, Adriano de Araujo Gomes, Germano Veras, Roberto Kawakami Harrop Galvao, Mario Cesar Ugulino Araujo
Summary: This paper introduces a new algorithm, PCA-DP-LDA, for solving classification problems of food data using PCA and LDA. Compared with conventional methods, PCA-DP-LDA achieves more parsimonious and interpretable results, with similar or better classification performance.
Article
Chemistry, Analytical
Railson de Oliveira Ramos, David Douglas de Sousa Fernandes, Valber Elias de Almeida, Paulo Henrique Gonsalves Dias Diniz, Wilton Silva Lopes, Valderi Duarte Leite, Mario Cesar Ugulino de Araujo
Summary: Monitoring the levels of total suspended solids (TSS) and settleable solids (SetS) in wastewater is crucial for maintaining water quality parameters. Traditional reference methods for determining TSS and SetS are time-consuming and expensive. To address this issue, researchers developed SWAMVA, a new analytical instrument equipped with a machine vision-based automatic sampler and software for real-time digital movie capture. SWAMVA significantly reduced the analysis volume and time while providing accurate and precise results.
ANALYTICA CHIMICA ACTA
(2022)
Article
Chemistry, Applied
Ellen Souza de Almeida Duarte, Valber Elias de Almeida, Gean Bezerra da Costa, Mario Cesar Ugulino de Araujo, Germano Veras, Paulo Henrique Goncalves Dias Diniz, David Douglas de Sousa Fernandes
Summary: The study showed that NIR spectroscopy and CACHAS can be used to determine the cassava starch content in wheat flour, with NIR spectroscopy being more suitable for definitive quantitative and qualitative analysis, while CACHAS can only provide an alternative preliminary analysis.
Article
Marine & Freshwater Biology
Maria Virginia da Conceicao Albuquerque, Railson de Oliveira Ramos, Valderi Duarte Leite, Jose Tavares de Sousa, Mario Cesar Ugulino de Araujo, Beatriz Susana Ovruski de Ceballos, Wilton Silva Lopes
Summary: Computational molecular modeling, mass spectrometry and in-vivo tests were used to investigate the liposolubility and ecotoxicity of MC-LR degradation byproducts in Fenton process. One of the byproducts was found to be more fat-soluble and more toxic compared to the original substance.
AQUATIC TOXICOLOGY
(2022)
Article
Food Science & Technology
Anabel S. Lourenco, Tassio A. Nunes, Amanda C. Silva, Williame F. Ribeiro, Mario C. U. Araujo
Summary: In this study, a square wave voltammetric method was developed for simultaneous determination of tryptamine and histamine in wine using a carbon paste electrode modified with nickel phthalocyanine. The proposed sensor showed high selectivity and accuracy, providing recovery values within a range of 82.65-103.02% for the analyzed biogenic amines.
FOOD ANALYTICAL METHODS
(2022)
Article
Chemistry, Applied
Eduardo A. Lima, Francisco A. S. Cunha, Maria J. Oliveira, Wellington S. Lyra, Mario M. S. Junior, Josue C. C. Santos, Sergio L. C. Ferreira, Mario C. U. Araujo, Luciano F. Almeida
Summary: This study developed a fast and sensitive method for the determination of inorganic antimony in grape juice samples. The method allows for quick sample pretreatment and analysis, and provides accurate and reliable results. The method is environmentally friendly and efficient for handling and analyzing a large number of samples.
Article
Chemistry, Analytical
Fernanda S. Soares, Raissa T. E. Ramalho, Amanda C. Silva, Mario C. U. Araujo
Summary: A sensitive, simple, fast, and inexpensive method for determining diuron in water using square wave voltammetry and a glassy carbon electrode is presented. This method was developed to evaluate diuron contamination in river waters close to sugarcane cultivation, located in the metropolitan region of Joao Pessoa, in the state of Paraiba, Brazil. The results showed that the proposed method has a linear response ranging from 38.5-115.0 nmol L-1 (r(2) = 0.993), a LOD equal to 0.2 nmol L-1, and recovery rates from 88.0 to 108.0%. The method does not require any previous chemical treatment of the sample, contributing to green analytical chemistry.
Article
Environmental Sciences
R. O. Ramos, W. S. Lopes, Y. H. D. M. Luna, M. V. da C. Albuquerque, V. D. Leite, J. T. de Sousa, M. C. U. Araujo
Summary: This study aimed to explain the differences in kinetic between reuse water disinfection processes using UV-C radiation and OH center dot radicals through molecular pathways. The results showed that the Fenton reaction was significantly more effective in inactivating thermotolerant coliforms (TTC) compared to H2O2/UV-C or direct UV-C. This superiority was attributed to the powerful action of OH center dot in disassembling the molecular structure of E. coli cell wall and inducing the collapse of the entire cell structure, as demonstrated through molecular mechanisms. Under optimized conditions, treatment with an anaerobic hybrid reactor, intermittent sand filter, and Fenton process achieved high removal rates for TOC, turbidity, and TTC, making the final effluent suitable for direct potable reuse.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY
(2023)
Article
Chemistry, Multidisciplinary
Denise M. Nobrega, Anabel S. Lourenco, David Douglas S. Fernandes, Wellington S. Lyra, Edilene D. T. Moreira, Paulo Henrique G. D. Diniz, Mario C. U. Araujo
Summary: This paper introduces a voltammetric method for determining the coumarin content in guaco leaves, and verifies the reliability of this method. The results show that this method is consistent with high-performance liquid chromatography.
JOURNAL OF THE BRAZILIAN CHEMICAL SOCIETY
(2022)
Article
Chemistry, Analytical
Juliana da Cruz Souza, Sofacles F. C. Soares, Lauro Cassio M. de Paula, Clarimar J. Coelho, Mario Cesar Ugulino de Araujo, Edvan Cirino da Silva
Summary: This paper introduces a bat-inspired algorithm for variable selection in linear discriminant analysis (LDA). The algorithm simulates the echolocation behavior of bats when searching for prey and uses a cost function associated with the average risk of misclassification in LDA. The results show that the algorithm performs similarly to genetic algorithm (GA-LDA) and successive projection algorithm (SPA-LDA) in classifying coffee and vegetable oil samples, and outperforms them in classifying ovarian cancer samples.
MICROCHEMICAL JOURNAL
(2023)
Article
Chemistry, Analytical
Rossana O. Nobrega, Suelly F. da Silva, David D. S. Fernandes, Welligton S. Lyra, Taynna K. L. de Araujo, Paulo H. G. D. Diniz, Mario C. U. Araujo
Summary: In this study, near-infrared spectroscopy and multivariate classification tools were used to develop methods for authenticating and checking the compliance of decaffeinated instant coffees, as well as discriminating between regular instant coffees according to the roasting degree. The combination of these techniques proved to be valuable and fast for checking the quality of instant coffees.
MICROCHEMICAL JOURNAL
(2023)
Article
Chemistry, Multidisciplinary
Kelly D. Silva, Stefani I. E. Andrade, Marcelo B. Lima, Severino S. Monte-Filho, Mario C. U. Araujo, Ricardo A. C. Lima
Summary: In this study, xiquexique cactus was used for the synthesis of highly fluorescent carbon dots, which were then used for the determination of tannins in teas. The proposed method was fast, low-cost, and eco-friendly, and showed no significant difference compared to the reference method.
JOURNAL OF THE BRAZILIAN CHEMICAL SOCIETY
(2023)
Article
Engineering, Chemical
Railson O. Ramos, Maria V. C. Albuquerque, Suelly F. da Silva, Wellington S. Lyra, Mario Cesar U. Araujo, Jose T. de Sousa, Valderi D. Leite, Wilton S. Lopes
Summary: This article presents a detailed study on the oxidation mechanism and kinetics of profenofos by HO center dot radicals. The oxidation reaction showed different behaviors in acidic and basic environments. Various oxidation by-products and reaction intermediates were identified, and the main mechanisms involved in the mineralization process were analyzed.
DESALINATION AND WATER TREATMENT
(2022)