相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Xiaodong Lv et al.
Summary: This article proposes a remote hand gesture recognition system based on a deep learning framework of multi-attention mechanism convolutional neural network using sEMG energy. The system enhances the recognition accuracy by improving the feature maps and shortcuts, and achieves high accuracy on multiple datasets.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Business
Zeyu Wang et al.
Summary: This study aims to improve the accuracy of risk prediction and credibility detection of network public opinion (NPO) and optimize the network environment using blockchain technology. A smart contract is used to build a risk management system of NPO, enabling the tracing of public opinion through a smart ledger. The study introduces blockchain, NPO, and related theories, discusses the network situation of public opinion dissemination in the blockchain environment, implements a BT-based risk management model for NPO, and conducts experiments to verify its effectiveness. The research also optimizes the theoretical framework of NPO credibility model detection based on the public opinion risk prediction model. The results show that the experimental schemes could reasonably predict and detect the risk and credibility of NPO under the BT, contributing to the optimization of network environment control measures.
TECHNOLOGICAL FORECASTING AND SOCIAL CHANGE
(2023)
Article
Environmental Studies
Zeyu Wang et al.
Summary: Under the rapid economic development trend, exploring resource optimization strategy of cultural and creative enterprises is crucial for sustainable socio-economic development. This study uses a recommendation system and neural network algorithm to model project features, user behavior, and content features in the cultural and creative industry. The constructed model significantly improves recognition accuracy and reduces prediction errors, providing experimental references for subsequent sustainable development and entrepreneurial resource optimization.
Article
Energy & Fuels
Marwa Ben Arab et al.
Summary: This paper proposes a seven-layer smart city energy management strategy to optimize power flow and achieve smooth power demand profiles, reduce electricity bills, and enable free charging of electric vehicles. The strategy is based on the Particle Swarm Optimization algorithm and utilizes a hierarchical structure to transfer energy between smart homes, renewable energy sources, and electric vehicles. Simulation results demonstrate significant efficiency of the proposed strategy.
Article
Green & Sustainable Science & Technology
Mahdi Pouresmaieli et al.
Summary: Sustainable development is crucial in all industries and mining activities. Internet of Things enables automation in mining, leading to increased production efficiency, optimized net present value, improved safety, and reduced pollution. This study provides a comprehensive understanding of IoT applications in mining and sustainable development through a comparative analysis of existing literature.
INTERNATIONAL JOURNAL OF SUSTAINABLE DEVELOPMENT AND WORLD ECOLOGY
(2023)
Article
Materials Science, Multidisciplinary
Huaming Tang et al.
Summary: Identification of rock minerals is made possible through computer vision technologies and deep learning theory. A targeted mineral identification model was developed using polarizing microscope images as the data source. With the capability to extract deep feature information and intelligently classify minerals, this model achieved a high accuracy rate of mineral recognition. The improved mineral identification model accurately depicts mineral species information, providing a convenient and reliable data source for the development of intelligent mineralogy.
Article
Green & Sustainable Science & Technology
Daniel M. Franks et al.
Summary: This article examines the absence of minerals in the Sustainable Development Goals and introduces concepts to better integrate minerals into the development agenda. Minerals are essential for human development, but are often seen as obstacles to sustainability.
NATURE SUSTAINABILITY
(2023)
Article
Plant Sciences
Yimin Hu et al.
Summary: The field of computer vision has great potential for large-scale crop identification using multispectral images, but there is a challenge in finding a balance between accuracy and lightweight design. This paper proposes an improved encoder-decoder framework based on DeepLab v3+ for accurate identification of crops with different planting patterns. The network incorporates ShuffleNet v2 as the backbone and a convolutional block attention mechanism in the decoder module. The results show significant improvements in performance compared to the original DeepLab v3+ on two datasets, DS1 and DS2, representing large-scale and scattered crop planting areas, respectively. The proposed Deep-agriNet is more efficient with fewer parameters and operations, making it an effective tool for crop identification in different regions and countries.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Geochemistry & Geophysics
Shi Li et al.
Summary: In this paper, a new 3D prediction method based on transfer learning is proposed to overcome the problems of limited and scattered data sources in mineral prospecting data. The method utilizes random noise to compensate for the limited sample size and proves to have good accuracy and stability through comparative tests. With the help of the transfer learning algorithm, the 3D CNN prediction method shows advantages in mineral prediction when big data are available.
Article
Geosciences, Multidisciplinary
Meng Gao et al.
Summary: The Weilasituo-bairendaba district in Inner Mongolia, China is an important part of the Cu-Pb-Zn polymetallic metallogenic belt. The district's structure framework consists of regional faults and secondary faults that provide channels for mineralization. Through integrated deposit geology and exploration data, 3D mineral prospectivity modeling using machine learning algorithms such as random forest and XGBoost was conducted. The optimal models showed high accuracy and were validated using ROC curves. The 3D targets identified are beneficial for Cu-Zn exploration in the district.
FRONTIERS IN EARTH SCIENCE
(2023)
Article
Agronomy
Hongjun Ni et al.
Summary: In this study, we propose an image classification model, RepVGG_ECA, which incorporates four image enhancement techniques and the ECA attention mechanism. The classification accuracy of 97.06% achieved by this model outperforms other traditional models. The research results provide a valuable reference for identifying typical rice pests and diseases.
Article
Environmental Studies
Fu Chen et al.
Summary: This study uses machine learning and artificial neural networks to predict the relationship between mineral resource rent and economic performance, greener energy, and environmental policy in China. The results show that mineral resource rent is closely related to dynamic shifts in economic performance and renewable energy use. The accuracy of machine learning experiments is higher compared to artificial neural networks, as evidenced by various error metrics and the coefficient of determination.
Article
Environmental Studies
Chengming Li et al.
Summary: Artificial intelligence (AI) offers businesses a way to save expenses and fundamentally transform innovation tools. This paper constructs a firm-level AI application index using text mining and investigates the impact of AI on corporate innovation efficiency using panel data from 3185 listed companies. The results show that AI significantly improves corporate innovation efficiency. The study also finds that intensified market competition and flattened organizational structure moderate the effects of AI on innovation resource reallocation.
Article
Automation & Control Systems
Hongjuan Yang et al.
Summary: This paper proposes a real-time detection method for rail corrugation based on machine vision and a convolutional neural network, which improves the accuracy and efficiency of rail corrugation detection. A rail surface segmentation method based on the gray maximum value of the sliding window is also proposed. The experimental results show that the improved model has an average detection time of 4.01ms and a detection accuracy 2.78% higher than the unimproved ShuffleNet V2. These research results will contribute to the development of intelligent real-time detection of rail corrugation.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Environmental Sciences
Li Chen et al.
Summary: This study investigates the application of hyperspectral data in CBM enrichment areas and proposes a method for mineral extraction based on spectral feature matching and diagnostic characteristic parameters. The extraction results are verified through X-ray diffraction analysis, and it is found that both ZY-1 02D and Hyperion hyperspectral data yield favorable extraction results for clay and carbonate minerals, with the higher accuracy of ZY-1 02D data.
Article
Environmental Studies
Hui Hu et al.
Summary: Managing shale gas resources can lead to various sustainability issues, which can be addressed through the use of Environment, Society, and Governance (ESG) reports to promote responsible natural resources management. The study examines the financial impact of the first ESG report in China on shale gas companies and finds that the report has heterogeneous effects on the market performance of relevant listed companies. The results suggest that the ESG report provides short-term positive returns to its issuer and potential associates, but negatively affects private and smaller companies.
Article
Business
Zeyu Wang et al.
Summary: The low-carbon city pilot policy in China is a crucial strategy in combating climate change and achieving comprehensive green transformation. This study evaluates the impact of the policy on digital economy growth, finding that it encourages growth and promotes green development. The results are supported by multiple tests and indicate that the policy has a more profound influence on digital economy growth in coastal, non-resource-based, and large-scale cities.
JOURNAL OF INNOVATION & KNOWLEDGE
(2023)
Article
Engineering, Electrical & Electronic
Tao Chen et al.
Summary: In this study, an intelligent energy management method is proposed for a hydrogen-dominant hybrid energy system with low carbon consideration. The method utilizes solid oxide fuel cells and chemical batteries to construct a high-efficient hybrid energy system, taking into account thermodynamics features, accurate battery model characteristics, and low carbon effects. The energy management method based on deep reinforcement learning techniques guides intelligent operation, achieving efficient use of hydrogen and economic as well as low carbon benefits.
Article
Telecommunications
Khursheed Aurangzeb et al.
Summary: This study proposes a framework that integrates energy storage system (ESS) and renewable energy resources (RERs) with smart homes, utilizing a multiheaded convolutional neural network model for accurate prediction of energy produced by RERs. The framework has shown significant results and is suitable for energy management at a community level.
TRANSACTIONS ON EMERGING TELECOMMUNICATIONS TECHNOLOGIES
(2022)
Article
Agriculture, Multidisciplinary
Yun Zhao et al.
Summary: In this paper, a convolutional neural network model based on Inception and residual structure with an embedded modified convolutional block attention module (CBAM) was proposed to enhance classification accuracy of plant leaf diseases. The model achieved a high overall accuracy of 99.55% for the identification of corn, potato, and tomato diseases. Additionally, individual accuracy rates of 98.44%, 99.43%, and 95.20% were achieved for corn, potato, and tomato, respectively. The model demonstrated fewer parameters, shorter training time, and higher recognition accuracy compared to existing image classification models.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Zoology
Ted Cheeseman et al.
Summary: This new convolutional neural network-based photo-identification algorithm is able to extract features and automatically recognize individuals in large humpback whale photo datasets, significantly reducing time and error rates compared to manual matching.
Article
Chemistry, Analytical
Vadim Ziyadinov et al.
Summary: This research proposes a technique for estimating the robustness and stability of convolutional neural networks and examines the impact of noise on recognition accuracy. It demonstrates the existence of an optimal amount of uncertainty in training data and shows that adding an optimal amount of uncertainty can improve the overall recognition quality and noise immunity of the neural networks.
Article
Chemistry, Multidisciplinary
Sania Malik et al.
Summary: Cloud computing has revolutionized computing, but also faces challenges such as power consumption, dynamic resource scaling, and resource provision. This research focuses on multi-resource utilization prediction using FLNN, GA, and PSO, with experimental results showing better accuracy compared to traditional techniques.
APPLIED SCIENCES-BASEL
(2022)
Article
Mathematics
Varun Tripathi et al.
Summary: The present study aims to develop a methodology for cleaner production management using lean and smart manufacturing in Industry 4.0. The results of two case studies show that the developed methodology can achieve a sustainable production system and problem-solving, while enhancing productivity within limited constraints.
Article
Engineering, Environmental
Zhichao Chen et al.
Summary: Garbage classification technology is an important basis for waste treatment and resource recovery. This study proposes a garbage classification system based on deep learning to recognize and recycle domestic garbage. The experimental results show that the system has high accuracy and fast classification performance.
RESOURCES CONSERVATION AND RECYCLING
(2022)
Article
Green & Sustainable Science & Technology
Chenyang Wu et al.
Summary: In recent years, the urbanization rate in China has been increasing, leading to ecological and geological degradation as obstacles to sustainable urban development. This study focuses on the concentrated mineral exploitation area of Tonghua City, evaluating the eco-geological environment quality using game theory and ANP-CV combined with GIS spatial calculation. The results provide a scientific basis for ecological construction and geological environmental protection in the city, with proposed countermeasures to improve the eco-geological environment quality.
Article
Green & Sustainable Science & Technology
Lidija Durdevac Ignjatovic et al.
Summary: This paper presents the results of synthesized and analyzed samples of different compositions of cement paste, which shows that the composition consisting of 5% cement, 24% water, and 71% flotation tailings is the most acceptable for filling excavation spaces and protecting groundwater and the environment.
Article
Engineering, Mechanical
Wen Hou et al.
Summary: This paper proposes a tool condition monitoring (TCM) method based on WGAN-GP and ShuffleNet. By enhancing and balancing the tool monitoring data using WGAN-GP and converting the signal data into grayscale images, a lightweight model is achieved. Experimental results show that the proposed method achieves an accuracy of 99.78% in recognizing the wear state of tools under imbalanced data, demonstrating its superiority.
JOURNAL OF MECHANICAL SCIENCE AND TECHNOLOGY
(2022)
Article
Geochemistry & Geophysics
Yunhui Kong et al.
Summary: This paper focuses on the scientific problem of quantitative mineralization prediction at large depth in the Zaozigou gold deposit, west Qinling, China. Machine learning and Deep learning algorithms are employed for 3D Mineral Prospectivity Mapping (MPM) with a proposed STOAD model, which shows good performance in mineral resources prediction.
Article
Geosciences, Multidisciplinary
Gang Liu et al.
Summary: Three-dimensional modeling of ore deposits is important for the quantitative evaluation of mineral resources. However, the non-stationarity of orebody distribution and uncertainty of data and models make 3D modeling challenging. A new approach called FE-3DRCS is proposed, combining geological exploration data and the MPS method. This approach effectively characterizes and evaluates orebodies in a real application.
NATURAL RESOURCES RESEARCH
(2022)
Article
Astronomy & Astrophysics
Neelam Agrawal et al.
ADVANCES IN SPACE RESEARCH
(2022)
Article
Computer Science, Information Systems
Mwamba Kasongo Dahouda et al.
Summary: This research proposes a deep learning-based feature extraction approach and applies it to the image classification of copper and cobalt raw minerals. The experimental results show that the method can efficiently extract features, and the modular neural network performs well with the boosting-decision tree as a classifier, achieving high accuracy and precision.
Article
Biochemistry & Molecular Biology
Wei-Heng Huang et al.
Summary: This study proposes a new method for phenotype prediction with genome-wide markers, combining deep learning and dimension reduction techniques to address issues such as overfitting and over-parameterization. The method shows significant prediction performance on simulated and real data.
FRONTIERS IN BIOSCIENCE-LANDMARK
(2022)
Article
Neurosciences
Mandong Hu et al.
Summary: The study improved the fuzzy clustering algorithm and designed a brain image processing and brain disease diagnosis prediction model based on fuzzy clustering and HPU-Net. Experimental results show that the improved algorithm has more nodes, lower energy consumption, and greater stability compared to other models under the same conditions.
FRONTIERS IN NEUROSCIENCE
(2021)
Article
Economics
Olga Janikowska et al.
Summary: Sustainable access to raw materials has been a growing concern for EU policy since 2008, with countries like Germany, Finland, United Kingdom, Portugal, and Greece developing their own minerals policy to prioritize sustainability, waste management, and climate issues. The introduction of the 17 Sustainable Development Goals in 2015 and the Paris Agreement targets have also had a significant impact on the mining industry. Analyzing the impact of mineral policy on sustainable mineral supply and CO2 emissions in selected countries can help identify challenges, trends, and successful factors in achieving the SDGs and EU climate targets.