Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Debanjan Konar et al.
Summary: This article presents a new shallow 3D self-supervised tensor neural network called 3D quantum-inspired self-supervised tensor neural network (3D-QNet) for volumetric segmentation of medical images. The network consists of input, intermediate, and output layers interconnected using a third-order neighborhood-based topology. Each layer contains quantum neurons designated by qubits or quantum bits. The network incorporates tensor decomposition in quantum formalism for faster convergence and achieves promising results in semantic segmentation.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Jiepeng Yao et al.
Summary: The classification methods of electrical signal can effectively identify salt tolerance in plants. Biotechnology can help researchers discover genes and proteins related to plant electrical signals, but it is time-consuming and expensive. While deep learning can predict protein interaction relationships, the relationships between plant electrical signals and proteins are unclear. To address this, we propose a graph neural network model that integrates plant electrical signal features to discover electrical signal-related proteins.
APPLIED SOFT COMPUTING
(2023)
Review
Chemistry, Multidisciplinary
Connor J. Taylor et al.
Summary: From the beginning of a synthetic chemist's training, experiments are conducted based on recipes from textbooks and manuscripts. However, it has been shown that model-based, algorithm-based, and miniaturized high-throughput techniques outperform human chemical intuition in understanding chemical systems and achieving reaction optimization. Many synthetic chemists are not exposed to these techniques, leading to a disproportionate number of scientists unable to utilize these methodologies. This review serves as a reference for inspired scientists, highlighting the basics and cutting-edge of chemical reaction optimization and its relation to process scale-up.
Article
Computer Science, Artificial Intelligence
Xiaoyu Ma et al.
Summary: This paper investigates the feasibility of incorporating channel-wise attention mechanism in blind image quality assessment (BIQA). An adaptive spatial and channel attention merging Transformer (ASCAM-Former) is proposed to aggregate both spatial-wise and channel-wise attention information. Experimental results show that channel-wise attention mechanism is as competitive as spatial-wise, and the ASCAM-Former yields accurate predictions on image quality datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Computer Science, Artificial Intelligence
Alessia Amelio et al.
Summary: In this paper, a multilayer network approach for representing and compressing ResNet is proposed. It can identify redundant convolutional layers and prune them, resulting in a new compressed ResNet. Experimental results demonstrate the suitability and effectiveness of the proposed approach.
EXPERT SYSTEMS WITH APPLICATIONS
(2023)
Article
Chemistry, Applied
Lixia Luo et al.
Summary: A cholesteric chiral artificial receptor L5 was synthesized and a rapid method for detecting iodide ions was developed. Addition of L5 to Ethanol/Tris solution changed the color of the iodide ion solution from colorless to light yellow, enabling naked-eye identification. The binding of L5 to iodide ions was confirmed by various tests, and it was found that the coordination ratio was 1:1. The hydrogen bonding between L5 and iodide ions was confirmed through multiple analyses. The detection limit of L5 was 0.025 μmol/L, making it suitable for efficient and convenient detection of iodide ions in food samples.
Article
Computer Science, Information Systems
Chu Wang et al.
Summary: Traffic prediction is crucial for urban computing. Existing approaches fail to model the spatial and temporal heterogeneity in traffic data at a fine-grained level, resulting in biased prediction results. To address this, we propose a Trend Graph Attention Network (TGAN) that captures spatial and temporal dependencies for accurate traffic prediction.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Mattia Segu et al.
Summary: Domain generalization aims to train machine learning models that can perform robustly across different domains. This study proposes a new approach that explicitly trains domain-dependent representations and maps domains in a shared latent space, achieving better generalization performance.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Ting Zhang et al.
Summary: This paper proposes a weakly-supervised butterfly detection model based on a saliency map (WBD-SM) to enhance the accuracy of butterfly detection in the ecological environment. Experimental results show that WBD-SM achieves higher recognition accuracy than VGG16 under different division ratios.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Artificial Intelligence
Mengzhu Wang et al.
Summary: This paper introduces a novel deep learning paradigm for enhancing the adaptation ability of unsupervised domain adaptation (UDA). By introducing feature redundancy to reduce feature redundancy and enhance adaptation ability, the proposed method achieves state-of-the-art performance on cross-domain object recognition benchmarks.
PATTERN RECOGNITION
(2023)
Article
Computer Science, Information Systems
Yashna Peerthum et al.
Summary: BatchNorm is shown to improve the training of deep neural networks, especially Convolutional Neural Networks (CNN). The study presents two new optimizers, AffineLayer and BatchNorm-minus, to compare with standard BatchNorm and the case without batch normalization. The research provides empirical evidence that the success of BatchNorm may come primarily from improved weight initialization.
INFORMATION SCIENCES
(2023)
Article
Green & Sustainable Science & Technology
Tingting Wang et al.
Summary: This paper investigates the impact of manufacturing structures on water quality, presenting indicators for clean manufacturing structures and water quality based on extensive data analysis. The findings reveal a significant negative correlation between the cleanliness level of manufacturing structures and chemical oxygen demand concentration. The study highlights the importance of optimizing manufacturing industry structures and reducing the proportion of high-pollution industries to improve water quality.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Computer Science, Interdisciplinary Applications
Feng Guo et al.
Summary: In order to improve traffic management and the performance of the traffic monitoring system at railroad-highway grade crossings, this study proposes a lightweight dense traffic detection network (DTDNet-Lite). The network utilizes an improved path aggregation feature pyramid network (iPAFPN) for multi-scale feature fusion, and employs a lightweight backbone, ResNet18, for efficient and accurate feature extraction. Comprehensive experiments on the VOC 2007 dataset and a customized grade crossing dataset show the superiority of DTDNet-Lite, paving the way for the deployment of efficient embedded AI computing devices for better traffic monitoring at grade crossings.
JOURNAL OF INDUSTRIAL INFORMATION INTEGRATION
(2023)
Article
Computer Science, Information Systems
Karam Park et al.
Summary: In this paper, a dynamic residual self-attention network (DRSAN) is proposed for lightweight single-image super-resolution (SISR) by automatically designing residual connections between building blocks. The DRSAN has dynamic residual connections based on dynamic residual attention (DRA), which adapts its structure according to input statistics. Experimental results show that DRSAN achieves an efficient trade-off between computational complexity and network performance.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Review
Computer Science, Artificial Intelligence
Zanyar Zohourianshahzadi et al.
Summary: This paper reviews literature on attentive deep learning models for image captioning, emphasizing different types of attention mechanisms. The most successful image captioning models follow the encoder-decoder architecture, with the best results currently achieved from variants of multi-head attention with bottom-up attention.
ARTIFICIAL INTELLIGENCE REVIEW
(2022)
Article
Thermodynamics
Haolin Yang et al.
Summary: In this paper, a data-driven deep learning network (GHTnet) was proposed to predict real-time electricity prices, with significant performance improvements achieved through the introduction of a new CNN module and time series summary statistics.
Article
Chemistry, Applied
Krittapas Kaewnu et al.
Summary: A colorimetric indicator cube designed for smart packaging detects ethanol in preserved baby mangoes by indicating color variations. The cube, made of porous melamine foam, shows distinct color changes over a range of ethanol concentrations and maintains good repeatability and storage stability.
Article
Engineering, Civil
Qian Wang et al.
Summary: Automatic crowd behaviour analysis is a crucial task for intelligent transportation systems, and crowd counting plays a key role in this analysis. Recent advancements in deep convolutional neural networks have led to significant progress in crowd counting. This paper evaluates the performance of the baseline Inception-v3 model on commonly used crowd counting datasets, achieving surprisingly good results comparable to or better than existing models. Furthermore, a novel Segmentation Guided Attention Network (SGANet) with Inception-v3 as the backbone and a curriculum loss is proposed, which outperforms prior arts, attaining state-of-the-art performance on multiple datasets.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Environmental Sciences
Andualem Mekonnen Hiruy et al.
Summary: The study revealed multiple bacterial hazards in rivers of the Akaki catchment, mainly originating from untreated urban wastewater, including faecal, antibiotic resistant, and potentially pathogenic bacteria. These hazards were particularly severe in the dry season, affecting the irrigation water quality for vegetable fields that supply markets.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Environmental Sciences
Wenzhong Tang et al.
Summary: Water pollution is a major global environmental problem, especially in developing countries. China has made remarkable achievements in water pollution control in the past twenty years, with wastewater collection and treatment capacity reaching a level similar to developed countries. Although the environmental quality of surface water has continuously improved, water pollution problems persist in the eastern river basins alongside remarkable economic progress. Rapid economic growth, rather than population growth, is the limiting factor for water pollution control in China. Further efforts are needed to enhance wastewater collection and treatment capacity and bridge the gap between effluent discharge limits and surface water quality standards. China's progress in water pollution control provides valuable insights for other developing countries.
Article
Environmental Sciences
Zongqing Lv et al.
Summary: Indicators related to organic matter are crucial for assessing aquatic environment quality. While COD is widely used, it may misrepresent water quality. BOD offers a better indication, but can be time-consuming. Results showed that oxidizing agents rapidly oxidized refractory organic matter, while size-fractional fluorescence and COD measurements could serve as proxies for BOD in monitoring coastal water pollutants.
ENVIRONMENTAL RESEARCH
(2022)
Article
Energy & Fuels
Hayane A. Fernandes et al.
Summary: This study investigated the interference of NaCl, CaCl2, MgCl2, and SrCl2 salt concentrations on the determination of TAN in crude oil using the ASTM D664 method. The experiments demonstrated the significant interference of these salts and identified the composition of the precipitates formed during the titration.
Article
Computer Science, Artificial Intelligence
Zhong Zhang et al.
Summary: The multi-head attention mechanism in Transformer models has been crucial to their success in language modeling. However, the small head size can lead to the low rank bottleneck. To address this issue, a mix-head attention (Mixhead) method is proposed, which combines multiple attention heads to improve the model's expressive power while introducing a negligible number of parameters.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Xiangyu Zhao et al.
Summary: This article introduces a new Prior Attention Network (PANet) for multi-lesion segmentation in medical images, which improves segmentation performance by introducing lesion-related spatial attention mechanism and intermediate supervision strategy.
IEEE TRANSACTIONS ON MEDICAL IMAGING
(2022)
Article
Multidisciplinary Sciences
Xiang Zhang et al.
Summary: In this study, dual-emissive N-GOQDs fluorescent probes were successfully developed for monitoring intracellular pH. These probes possess bright fluorescence, high stability, and good biocompatibility.
Article
Automation & Control Systems
Deepak Kumar Jain et al.
Summary: The fifth industrial revolution, Industry 5.0, integrates humans and machines to meet customization demands using optimized robotized manufacturing processes. This article proposes an AI-based UAV-borne secure communication with classification (AIUAV-SCC) framework for Industry 5.0, addressing security issues in UAV communication and utilizing deep learning for image classification.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Bo Li et al.
Summary: This paper proposes RT-Unet, which combines the advantages of Transformer and Residual network, for accurate medical segmentation. Experimental results demonstrate that RT-Unet outperforms other deep learning-based algorithms in terms of accuracy and efficiency.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Automation & Control Systems
Borja Bovcon et al.
Summary: The study introduces a novel deep encoder-decoder architecture, named WaSR network, specifically designed for obstacle detection in the marine environment. Through the combination of a deep encoder based on ResNet101, a decoder, and inertial information, as well as a novel loss function, the network effectively improves the segmentation accuracy of water components and achieves better segmentation results.
IEEE TRANSACTIONS ON CYBERNETICS
(2022)
Article
Multidisciplinary Sciences
Cem Kalyoncu et al.
Summary: Machine learning approaches have been increasingly used in optical devices and fibers in recent years. This work proposes the use of k-Nearest Neighbor Regression (KNNR) as a non-linear regression method to determine the loss characteristics of a photonic crystal fiber (PCF) based surface plasmon resonance (SPR) sensor in the presence of a bend. Experimental results show that KNNR outperforms Artificial Neural Network (ANN) and Linear Least Square Regression methods, and does not require lengthy training process.
Article
Multidisciplinary Sciences
A. Tarbi et al.
Summary: Researchers evaluate the bandgap energy of GaAsN material using the band anticrossing model and artificial neural networks method, considering the strain caused by lattice mismatch. This method makes GaAsN a potential material for the fabrication of ultrafast optical sensors.
Article
Computer Science, Artificial Intelligence
Zheng Xiao et al.
Summary: This paper proposes an extended attention-based framework for scene text recognition tasks. By introducing the Attention on Attention (AoA) mechanism, the relevance between attention results and queries can be determined, improving the accuracy of recognition. Experimental results show that the proposed method outperforms other benchmarks on multiple datasets.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Engineering, Civil
Tao Ye et al.
Summary: Detection of railway traffic objects is crucial for safe train driving. This study proposes a novel deep learning method, MMFE-Net, to accurately detect railway objects. The network utilizes improved backbone network, spatial feature extraction, and attention fusion enhance module to address challenges in complex railway scenes. Experimental results show that MMFE-Net outperforms other methods on railway traffic dataset and is feasible for practical railway object detection tasks.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Engineering, Civil
Muneeb Ahmed et al.
Summary: Fatigue, drowsiness, and distraction are the main causes of road accidents worldwide. Existing solutions either extract physiological signals of the driver or use computer vision techniques, but they have limited performances. Therefore, this study proposes an ensemble deep learning architecture to determine the state of the driver by incorporating features from the eyes and mouth. The model achieves high accuracy when trained and evaluated on the NTHU-DDD video dataset.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Xinchun Cui et al.
Summary: In this study, a FResnet18 model is proposed to classify MRI images of PD and HC by fusing image texture features with deep features. The results show that the model can successfully differentiate between PD and HC with high accuracy, and it outperforms existing approaches.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Chemistry, Analytical
Zhimin Zhang et al.
Summary: This paper presents a UV-IR dual wavelength COD measurement system based on Chirp modulation, which improves the signal-to-noise ratio of UV signals by simplifying the hardware architecture and avoids cross interference of dual-wavelength. The system is characterized by its low cost, low power consumption, and high precision, making it suitable for large-scale in-situ deployment.
SENSORS AND ACTUATORS B-CHEMICAL
(2022)
Review
Chemistry, Analytical
Piyawan Phansi et al.
Summary: This work aims to revise programs used to obtain accurate equilibrium constants through multiple chained programs. The AutoAnalysis software is valuable for conducting experimental work, including potentiometric titrations. The POTENtit program detects problems in the data obtained with AutoAnalysis during the experimental work, such as the presence of metals in the sample and carbonation of the titrant. It also provides the standard potential of the working electrode and the ionic product of the solvent under the experimental ionic strength. The MINIPOT program refines the results obtained, including electrode calibration and equilibrium constants. These three programs have been updated to run in the Windows 10 environment.
TRAC-TRENDS IN ANALYTICAL CHEMISTRY
(2022)
Article
Urban Studies
Ishaani Priyadarshini et al.
Summary: Water quality is affected by urbanization, and this study uses machine learning techniques to analyze water quality and identify marine litter to address the issue of water pollution and promote sustainable urban development.
Article
Multidisciplinary Sciences
Sunita Kesar et al.
Summary: This study provides detailed information on the usefulness of chlorination in treating sewage for irrigation purposes. Chlorination is a cost-effective method for disinfection. The study optimized the disinfection process using sodium hypochlorite and investigated various process variables. The results showed that certain kinetic models fit well with the experimental data of chlorine disinfection.
Review
Engineering, Environmental
Hao Sun et al.
Summary: This article discusses the importance of quantitative analysis of VFAs and the limitations of commonly used detection techniques. It highlights the advantages of biosensors as an alternative method, with a focus on EAB-based sensors. The article emphasizes the impact of the contact between biofilm and VFA on the performance of the sensor. Additionally, it mentions the applications and challenges of biosensors in different biological processes, and suggests future research directions.
CHEMICAL ENGINEERING JOURNAL
(2022)
Article
Environmental Sciences
Karin Yaniv et al.
Summary: The COVID-19 pandemic has had a global impact on healthcare, economies, and societies, highlighting the need for novel disease monitoring methods. Recent data suggests that wastewater monitoring can be an efficient tool for epidemiological surveillance and early warning of SARS-CoV-2 circulation at the population level. This study showed varying presence of SARS-CoV-2 RNA in wastewater samples from different locations in a city before a clinical resurgence, indicating the potential for early detection of infection outbreaks in populations.
Article
Engineering, Environmental
Urban J. Wunsch et al.
Summary: Dissolved organic matter (DOM) plays a crucial role in the global carbon cycle, and fluorescence spectroscopy combined with parallel factor analysis (PARAFAC) is an efficient method for tracking DOM quality and quantity. A new experimental procedure has been introduced to address the limitations of the PARAFAC method, allowing for mathematical decomposition of small datasets containing highly-correlated fluorescent fractions.
Article
Environmental Sciences
Jakub Nalepa et al.
Summary: Research focuses on developing algorithms for hyperspectral data classification and unmixing, with deep learning techniques proving to be highly effective. However, designing deep models that generalize well remains a practical challenge.
Article
Computer Science, Artificial Intelligence
Farah Saeed et al.
Summary: The study presents an automated crop disease recognition system using PLS regression for feature selection, achieving an accuracy of approximately 90.1%. The proposed method not only improves recognition accuracy but also reduces computational time.
APPLIED SOFT COMPUTING
(2021)
Article
Multidisciplinary Sciences
Khalida Khatoon et al.
Article
Multidisciplinary Sciences
Million Ebba Bote
Summary: Electrocoagulation is an electrochemical method that uses sacrificial electrodes to treat wastewater. The combination of aluminum and iron electrodes influences the removal efficiency of COD, with Al-Fe and Fe-Al combinations showing promising results. Various factors such as pH, reaction time, and current density play a role in determining the overall efficiency of COD removal.
Article
Multidisciplinary Sciences
Nianzhi Jiao et al.
Summary: The study found that chemical oxygen demand (COD) is not an appropriate proxy for assessing the microbial degradability of organic matter in aquatic environments, and recommends the replacement of COD with an optode-based biological oxygen demand method.
Article
Green & Sustainable Science & Technology
Jamie Woodward et al.
Summary: The authors discuss how untreated wastewater containing microplastics and raw sewage is discharged into UK rivers, causing severe contamination of river beds and threatening biodiversity and habitats. Climate change and urbanization will exacerbate the microplastic pollution in river ecosystems, as summer flows decrease and wastewater discharges increase.
NATURE SUSTAINABILITY
(2021)
Article
Chemistry, Analytical
Nan Hao et al.
Summary: A portable self-powered sensor chip was developed to rapidly detect organic pollutants in water bodies by utilizing photocatalytic degradation for COD quantification, providing advantages of speed, sensitivity, environmental friendliness, and ease of use.
ANALYTICAL CHEMISTRY
(2021)
Article
Computer Science, Artificial Intelligence
Le Yan et al.
Summary: This paper proposes a novel method combining deep residual neural network and lower and upper bound estimation for forecasting future flow, which outperforms other deep learning and machine learning models in terms of RMSE, R-2, and CWC based evaluations using spatiotemporal dataset.
APPLIED SOFT COMPUTING
(2021)
Article
Chemistry, Multidisciplinary
Tapendu Samanta et al.
Summary: A compound was converted for detection of fluoride ions, displaying efficient detection with color change and mechanism explained through spectroscopic and NMR studies.
ACS SUSTAINABLE CHEMISTRY & ENGINEERING
(2021)
Article
Computer Science, Information Systems
Gayathri Nagasubramanian et al.
Summary: The IoT in agriculture offers data sharing and automated farming solutions for crops, utilizing machine learning techniques to monitor crop growth and diseases, providing decision-making advice to farmers.
IEEE INTERNET OF THINGS JOURNAL
(2021)
Article
Chemistry, Multidisciplinary
Yukang Gong et al.
Summary: DeepReac+ is an efficient and universal computational framework for the prediction of chemical reaction outcomes and identification of optimal reaction conditions based on deep active learning, achieving state-of-the-art results with a minimum of labeled data on diverse chemical reaction datasets.