Search

In this paper, 1,538 papers retrieved with the keywords "sports artificial intelligence (AI)" on the Web of Science database since 2007 were taken as the data source, and the Cite Space V software was used to visualize and analyze them. A visual knowledge graph was used to streamline the countries, institutions and authors conducting sports AI research, discipline distribution, research hotspots and development trends in the past 15 years. Subsequently, its development direction and research progress were discussed. Sports AI was widely distributed, with the US, China and the UK leading the way. The most prolific authors and teams in research on sports AI were concentrated in American universities. Their main research direction is to develop and improve smart wearable devices based on machine learning and deep learning technologies for different groups of people. Research on sports AI involved multiple disciplines, which mainly applied and referred to research methodologies and theories on engineering, computer science and sports science. It could be seen from the frequency and centrality of keywords that in the current field of sports AI, machine learning is the main direction, artificial neural networks is the main algorithm, and practical and empirical research based on data mining is the focus. The research hotspots were divided into three major clusters: physical health promotion, sports injury prevention and control, and athletic performance enhancement. How to introduce intelligent technology into sports for a perfect integration still has an arduous and long way to go. Future development requires joint efforts and participation of scientific researchers, professionals and common people. [ABSTRACT FROM AUTHOR]

The problem of global warming is getting more serious, and vehicle emission is the main cause. In recent years, the number of locomotives in China has been increasing at a rate of more than 20% per year, and the problem of automobile pollution is becoming more serious. The transportation industry is the main source of fossil fuel combustion and environmental pollution. Therefore, in this paper, we propose a multi-view deep neural network (MVDNN) to analyze the key factors affecting the fuel consumption of automobiles. The experiments show that the introduction of human input improves the prediction accuracy and the root mean square error (RMSE) achieves 0.993. In addition, this paper also finds that for drivers, driving habits, driving frequency, and safety awareness are the most important factors affecting the fuel consumption of vehicles by combining Lasso regression with MVDNN. Finally, by comparing the prediction accuracy of different experiments, relevant policy suggestions are made. [ABSTRACT FROM AUTHOR]

As the fastest-growing crowdfunding model, equity crowdfunding (ECF) brings high returns and uncertainty. In this context, it is crucial to understand these crowdfunding projects' actual performance. Since ECF is currently in the early stage of integration, there are still a lot of risk issues, such as the uncertainty of equity structure, capital supervision, or project management. Therefore, this paper develops a new profitability indicator, "return on registered capital," to test its impact on the ECF project's actual return. This paper studies which factors affect the financial performance of ECF projects through the traditional statistical model and a deep neural network (DNN) model. There is evidence that return on registered capital affects the actual return of the project. At the same time, the company's operating time and the number of employees had an unexpected effect on project performance. In addition, the recognition accuracy of the DNN model in this study exceeds 97%, which affirms the applicability of the DNN model in the analysis of ECF success factors. This paper also uses tenfold cross-validation to prove that deep learning has certain advantages in this topic's accuracy and generalization error. This study explores whether company representatives' gender and knowledge level affect project performance. The results will be described in detail in the paper. [ABSTRACT FROM AUTHOR]

As China's economic level and industrial volume continue to develop and expand, the traditional methods of business management and resource scheduling demonstrate significant limitations and misalignments with China's daily economic activities. Traditional industrial and commercial management resource scheduling is highly reliant on manual labor. When confronted with the requirements of large-scale project management, the pure manual mode cannot track the real-time progress of the project flexibly, effectively, and in a timely manner, resulting in the inability to complete the corresponding resource-scheduling work efficiently, which will affect the overall operation progress of the project. Deep neural networks have a high-application potential for solving extremely complex and highly nonlinear optimization problems. Artificial intelligence and deep learning research have advanced at a rapid pace over the past few decades, thanks to the efforts of numerous researchers. Combined with GPU technology, the deep learning framework can provide an extremely complex optimization problem with a practical and feasible optimization scheme and corresponding solution path in a very short amount of time. Therefore, this paper explores the potential application of deep learning technology to industrial and commercial resource-scheduling management. By analyzing the benefits of deep learning technology and the bottleneck issues of existing industrial and commercial resource scheduling, a real-time optimized industrial and commercial resource-scheduling model based on deep learning technology is developed. The model is evaluated using the respective data set. The test results demonstrate that the resource-scheduling model proposed in this paper has strong real-time and high-operation efficiency and can assist engineers in completing the corresponding resource-scheduling tasks. [ABSTRACT FROM AUTHOR]

The classification that distinguishes whether machines are driving on roads or working in fields based on their global navigation satellite system (GNSS) trajectories is essential for effective management of cross-regional agricultural machinery services in China. In this paper, a novel field–road classification method utilizing multiple deep neural networks (MultiDNN) is proposed to enhance the accuracy of field and road point classification. The MultiDNN model incorporates a bi-directional long short-term memory network (BiLSTM), a topology adaptive graph convolution network (TAG), and a self-attention network (ATT) to effectively extract spatio-temporal features for field–road classification. The BiLSTM is used to capture temporal relationships along the time axis of a trajectory, providing global contextual information for each point. Then, the TAG network is used to obtain the spatio-temporal relationships between adjacent points in a trajectory, offering local contextual information for each point. Finally, the ATT network assigns varying weights to features to emphasize important characteristics. The performance of the MultiDNN model was evaluated using a wheat harvesting trajectory dataset, and the results showed that it achieved a high degree of accuracy, up to 89.75%, outperforming the best baseline method (GCN) by 2.79%. [ABSTRACT FROM AUTHOR]

Background: Lung cancer is one of the cancers with the highest mortality rate in China. With the rapid development of high-throughput sequencing technology and the research and application of deep learning methods in recent years, deep neural networks based on gene expression have become a hot research direction in lung cancer diagnosis in recent years, which provide an effective way of early diagnosis for lung cancer. Thus, building a deep neural network model is of great significance for the early diagnosis of lung cancer. However, the main challenges in mining gene expression datasets are the curse of dimensionality and imbalanced data. The existing methods proposed by some researchers can't address the problems of high-dimensionality and imbalanced data, because of the overwhelming number of variables measured (genes) versus the small number of samples, which result in poor performance in early diagnosis for lung cancer. Method: Given the disadvantages of gene expression data sets with small datasets, high-dimensionality and imbalanced data, this paper proposes a gene selection method based on KL divergence, which selects some genes with higher KL divergence as model features. Then build a deep neural network model using Focal Loss as loss function, at the same time, we use k-fold cross validation method to verify and select the best model, we set the value of k is five in this paper. Result: The deep learning model method based on KL divergence gene selection proposed in this paper has an AUC of 0.99 on the validation set. The generalization performance of model is high. Conclusion: The deep neural network model based on KL divergence gene selection proposed in this paper is proved to be an accurate and effective method for lung cancer prediction. [ABSTRACT FROM AUTHOR]

In recent years, China's engineering construction management level has been greatly improved, but compared with other industries, the construction industry still has low production efficiency, serious waste, and low level of information problems, and especially in the process of engineering management practice, schedule delay has become the focus of engineering management problems. With the continuous development of science and technology, computer and information technology have been continuously applied in engineering, among which deep neural network (DNN) technology, lean management, information visualization technology, and other technologies have become the hot spot of industry research, and the application of these emerging technologies to improve the level of project schedule control has become an urgent demand of the industry. Therefore, on the basis of deep learning, this paper analyzes the principle and application of object detection and feature extraction constructed by neural network and combines text feature extraction and image feature extraction methods. This application provides a new idea for the development of the construction industry. [ABSTRACT FROM AUTHOR]

Soil salinization has become one of the major environmental problems threatening food security. The identification and knowledge of the spatial distributions of soil salinization are key in addressing this problem. This study assumes that a good saline land identification effect can be obtained with the help of deep learning methods. Therefore, this study used the OLI sensor data from the Landsat-8, based on the U2-Network, and proposes a method to extract saline land from remote sensing images. The study also adds different salinity indices (SI, SI1, and SI2) to explore its impact on classification accuracy. Through our method, accurate saline soil distribution information were obtained, and several verification indicators (the Intersection-over-Union (IoU), recall, precision, and F1-score) were all measured above 0.8. In addition, compared with the multi-spectral training results, the classification accuracy increased after adding a specific salinity index, and most of the accuracy indices increased by about 2% (the IoU increased by 3.70%, recall increased by 1.50%, precision increased by 2.81%, and F1-score increased by 2.13%). In addition, we also included a case study based on our methodology to analyze the distribution characteristics and changes of saline soil in the Zhenlai area of Northeast China from 2016 to 2020. We found that the area of saline land in the Zhenlai area has reduced, which shows that the extraction method proposed in this study is feasible. Overall, this paper indicates that deep learning-based methods can efficiently extract the salinity of soil and enhance the mapping of its spatial distribution. The study has the broad impact of supplementing satellite imagery for salinity modeling and helping to guide agricultural land management practices for northeastern China and other salinized regions. [ABSTRACT FROM AUTHOR]

The unidirectional consolidation theory of soils is widely used in certain conditions and approximate calculations. The multidirectional theory of soil consolidation is more reasonable than the unidirectional theory in practical applications but is much more complicated in terms of index determination and solution. To address the above problem, in this paper, we propose a deep learning method using physics-informed neural networks (PINN) to predict the excess pore water pressure of two-dimensional soil consolidation. In the proposed method, (1) a fully connected neural network is constructed; (2) the computational domain, partial differential equation (PDE), and constraints are defined to generate data for model training; and (3) the PDE of two-dimensional soil consolidation and the model of the neural network are connected to reduce the loss of the model. The effectiveness of the proposed method is verified by comparison with the numerical solution of PDE for two-dimensional consolidation. Moreover, the FEM and the proposed PINN-based method are applied to predict the consolidation of foundation soils in a real case of Sichuan Railway in China, and the results are quite consistent. The proposed deep learning approach can be used to investigate large and complex multidirectional soil consolidation. [ABSTRACT FROM AUTHOR]

Annual monitoring of the spatial distribution of cultivated land is important for maintaining the ecological environment, achieving a status quo of land resource management, and guaranteeing agricultural production. With the gradual development of remote sensing technology, it has become a common practice to obtain cultivated land boundary information on a large scale with the help of satellite Earth observation images. Traditional land use classification methods are affected by multiple types of land cover, which leads to a decrease in the accuracy of cultivated land mapping. In contrast, although the current advanced methods (such as deep learning) can obtain more accurate cultivated land mapping results than traditional methods, such methods often require the use of a massive amount of training samples, large computing power, and highly complex model tuning processes, increasing the cost of mapping and requiring the involvement of more professionals. This has hindered the promotion of related methods in mapping institutions. This paper proposes a method based on time series vector features (MTVF), which uses vector thinking to establish the features. The advantage of this method is that the introduction of vector features enlarges the differences between the different land cover types, which overcomes the loss of mapping accuracy caused by the influences of the spectra of different ground objects and ensures the calculation efficiency. Moreover, the MTVF uses a traditional method (random forest) as the classification core, which makes the MTVF less demanding than advanced methods in terms of the number of training samples. Sentinel-2 satellite images were used to carry out cultivated land mapping for 2020 in northern Henan Province, China. The results show that the MTVF has the potential to accurately identify cultivated land. Furthermore, the overall accuracy, producer accuracy, and user accuracy of the overall study area and four sub-study areas were all greater than 90%. In addition, the cultivated land mapping accuracy of the MTVF is significantly better than that of the maximum likelihood, support vector machine, and artificial neural network methods. [ABSTRACT FROM AUTHOR]

Timely and accurate survey of construction and demolition waste (C&DW) distribution is of significance for managing C&DW and enhancing the quality of the urban environment, especially in rapidly urbanizing country of China. Automatic C&DW recognition using high-resolution remote sensing images is an extremely important surveying method. Existing C&DW recognition methods can only obtain a rough spatial distribution of C&DW. Mapping precise C&DW distribution on large-scale remote sensing images is still challenging. In this paper, to solve the problem of precise C&DW recognition, a novel deep learning algorithm based on high-resolution remote sensing images is proposed, which we refer to as the cross-channel multi-scale gated fusion network (CCMGNet). CCMGNet extracts depth features of C&DW from RGB and NIR bands by using two independent encoding streams. A gated fusion layer and a multi-scale attention module are designed to effectively fuse the features from two encoding streams. We manually labelled the sample set used for training and testing on 26 remote sensing images captured by the Gaofen-2 (GF-2) satellite. These images were taken in nine typical cities in China over the time period from 2019 to 2021. The results showed that the proposed method is effective in recognizing C&DW, realizing 89.42% and 84.52% in precision and IoU, and is superior to other state-of-the-art deep learning algorithms and existing C&DW recognition methods. The effectiveness of all significant components in CCMGNet was confirmed by ablation experiments. The proposed method was applied to extract C&DW regions in large-scale remote sensing images, with an example of a GF-2 image of Beijing. The high extraction efficiency and satisfactory visual effect demonstrate the potential of the proposed method for surveying the spatial distribution of C&DW. The novel method will play a crucial role in automatically surveying the spatial distribution and size of C&DW across the country. [ABSTRACT FROM AUTHOR]

With the downward pressure of China's economy and the impact of the epidemic, the accumulated market risk has increased the liquidity pressure of the banking industry, and the mismatch between deposit maturity and loan maturity is the main cause for the increase of liquidity risk. The twenty-first century is the era of rapid and in-depth development of data management technology. The explosive growth of massive financial data makes the information data related to the liquidity risk of commercial banks present the characteristics of complexity, diversity, and heterogeneity. The traditional risk early warning model cannot deal with the influence between a large number of influencing factors and the nonlinear factors of commercial bank liquidity risk. Based on this transformation, the circular neural network model is introduced into the field of liquidity risk early warning of commercial banks from the perspective of the mismatch risk of financing maturity of commercial banks, and the driving factors and risk warning signs of liquidity risk of commercial banks are further analyzed from the institutional level, policy level, industry level, and micro commercial bank level. This paper uses network crawler technology, text analysis, and grounded analysis technology to intelligently identify the liquidity risk of commercial banks and establishes an early warning index system based on the influencing factors of commercial banks and internal liquidity risk. Also, it constructs an intelligent early warning model of commercial bank liquidity risk based on deep learning and uses the data of commercial banks from 2000 to 2020 for early warning. The results show that the constructed model has high accuracy, which can provide support for banks and relevant government departments to formulate and resolve bank liquidity risk. [ABSTRACT FROM AUTHOR]

Different from traditional algorithms and model, machine learning is a systematic and comprehensive application of computer algorithms and statistical models, and it has been widely used in many fields. In the field of finance, machine learning is mainly used to study the future trend of capital market price. In this paper, to predict the time-series data of stock, we applied the traditional models and machine learning models for forecasting the linear and non-linear problem, respectively. First, stock samples that occurred from year 2010 to 2019 at the New York Stock Exchange are collected. Next, the ARIMA (autoregressive integrated moving average model) model and LSTM (long short-term memory) neural network model are applied to train and predict stock price and stock price subcorrelation. Finally, we evaluate the proposed model by several indicators, and the experiment results show that: (1) Stock price and stock price correlation are accurately predicted by the ARIMA model and LSTM model; (2) compared with ARIMA, the LSTM model performance better in prediction; and (3) the ensemble model of ARIMA-LSTM significantly outperforms other benchmark methods. Therefore, our proposed method provides theoretical support and method reference for investors about stock trading in China stock market. [ABSTRACT FROM AUTHOR]

As a result of the fast rise of globalization, people in China are learning English at a rapid pace. However, there is a severe shortage of English teachers in the region, which is a major hindrance. To address these concerns, a deep learning-based algorithm is proposed that can not only check English pronunciation but also help learners distinguish between phonemic and quality phonemic while listening and differentiating, as well as correct phonemic errors, thereby increasing their language learning capacity. In order to study the application of nonlinear network identification technology in English learning, this paper evaluates the English pronunciation quality through the deep learning algorithm of deep learning combined with the related contents of neural network data model, and the experimental results of speech recognition structure are analyzed and discussed in detail. The concordance between machine and manual intonation evaluation is 80%, the concordance rate of adjacent intonation evaluation is 98.33%, and the Pearson correlation coefficient is 0.627 that shows the technique is reliable. The method of English pronunciation and speech identification model is sensible and dependable, which can give beginners a punctual, exact and impartial judgment and response guidance, assist learners to get on the differences between their phonemic and standard phonemic, and correct phonemic mistakes, in order to enhance the ability of oral English learning. [ABSTRACT FROM AUTHOR]

This paper attempts to apply recurrent neural networks (RNN) to price forecasts and financial trading. Compared with previous neural networks models, the recurrent neural network can better use the previous information to infer subsequent events, which is more suitable for price time series analysis. Long Short-Term Memory (LSTM) has made structural changes to the RNN to avoid long-term dependency problems. The empirical research uses the 2010–2017 price panel data of four kinds of soybean futures in China's futures market, and confirms the model's improved predictive ability through statistical tests. The empirical analysis of futures trading verifies the practice of these model strategies in terms of risked return. This paper improves and expands the application of recurrent neural networks model, and provides a new idea for applying artificial neural network algorithm to futures trading. [ABSTRACT FROM AUTHOR]

Ice clouds are mostly composed of different ice crystal habits. It is of great importance to classify ice crystal habits seeing as they could greatly impact single‐scattering properties of ice crystal particles. The single‐scattering properties play an important role in the study of cloud remote sensing and the Earth's atmospheric radiation budget. However, there are countless ice crystals with different shapes in ice clouds, and the task of empirical classification based on naked‐eye observations is unreliable, time consuming and subjective, which leads to classification results having obvious uncertainties and biases. In this paper, the images of ice crystals observed from airborne Cloud Particle Imager in China are used to establish an ice crystal data set called Ice Crystals Database in China, which consists of 10 habit categories containing over 7,000 images. We propose an automatic classification model of ice crystal habits, called TL‐ResNet152, which is a deep convolutional neural network based on the newly developed method of transfer learning. The results show that the TL‐ResNet152 model could achieve reliable performance in ice crystal habits classification with the accuracy of 96%, which is far more accurate than traditional classification methods. Achieving high‐precision automatic classification of ice crystal habits will help us better understand the radiation characteristics of ice clouds. Plain Language Summary: In recent years, Convolutional Neural Networks (CNNs), as one of the representative algorithms used in deep learning, have been widely applied in the field of image classification and have achieved remarkable results. However, to the best of our knowledge, there are few studies regarding the application of deep CNNs to ice crystal image classification. In this paper, we propose an automatic classification model of ice crystal habits called TL‐ResNet152, which is a deep CNN based on a newly developed method of transfer learning. Using the TL‐ResNet152 model, high precision and automatic classification of ice crystal habits could be achieved, furthering our understanding of radiation characteristics of ice clouds. We have set up an ice crystal data set, called Ice Crystals Database in China, which consists of 10 habit categories with 7,282 images. As far as we know, it is the first publicly available ice crystals data set in China, pertaining to ice crystals observed in natural ice clouds. The publication of this database will promote more and more researches into understanding the physical process of ice clouds based on ice crystal habits classification. Key Points: An ice crystal data set called Ice Crystals Database in China (ICDC) containing over 7,000 images has been first establishedWe propose an automatic classification model of ice crystal habits, TL‐ResNet152, which is a convolutional neural network created using transfer learningThe high‐precision automatic classification of ice crystal habits is achieved by using this classification model [ABSTRACT FROM AUTHOR]

Optimal operation of hydropower reservoir systems is a classical optimization problem of high dimensionality and stochastic nature. A key challenge lies in improving the interpretability of operation strategies, i.e., the cause–effect relationship between system outputs (or actions) and contributing variables such as states and inputs. This paper reports for the first time a new deep reinforcement learning (DRL) framework for optimal operation of reservoir systems based on deep Q-networks (DQNs), which provides a significant advance in understanding the performance of optimal operations. DQN combines Q-learning and two deep artificial neural networks (ANNs), and acts as the agent to interact with the reservoir system through learning its states and providing actions. Three knowledge forms of learning considering the states, actions, and rewards were constructed to improve the interpretability of operation strategies. The impacts of these knowledge forms and DRL learning parameters on operation performance were analyzed. The DRL framework was tested on the Huanren hydropower system in China, using 400-year synthetic flow data for training and 30-year observed flow data for verification. The discretization levels of reservoir water level and energy output yield contrasting effects: finer discretization of water level improved performance in terms of annual hydropower generated and hydropower production reliability; however, finer discretization of hydropower production can reduce search efficiency, and thus the resulting DRL performance. Compared with benchmark algorithms including dynamic programming, stochastic dynamic programming, and decision tree, the proposed DRL approach can effectively factor in future inflow uncertainties when determining optimal operations and can generate markedly higher hydropower. This study provides new knowledge of the performance of DRL in the context of hydropower system characteristics and data input features, and shows promise for potentially being implemented in practice to derive operation policies that can be updated automatically by learning from new data. [ABSTRACT FROM AUTHOR]

• We construct a fine-grained classification dataset of wet Age-related macular degeneration (AMD). • A knowledge-driven deep learning model, KFWC, is proposed to improve accuracy in fine-grained disease classification with insufficient data. • KFWC can detect 10 signs of lesions on medical images and can diagnose two subtypes of wet-AMD. • Multiple experiments demonstrate the effectiveness and interpretability of the proposed KFWC. Automated diagnosis using deep neural networks can help ophthalmologists detect the blinding eye disease wet Age-related Macular Degeneration (AMD). Wet-AMD has two similar subtypes, Neovascular AMD and Polypoidal Choroidal Vasculopathy (PCV). However, due to the difficulty in data collection and the similarity between images, most studies have only achieved the coarse-grained classification of wet-AMD rather than a fine-grained one of wet-AMD subtypes. Therefore, designing and building a deep learning model to diagnose neovascular AMD and PCV is a great challenge. To solve this issue, in this paper, we propose a Knowledge-driven Fine-grained Wet-AMD Classification Model (KFWC) to enhance the model's accuracy in the fine-grained disease classification with insufficient data. We innovatively introduced a two-stage method. In the first stage, we present prior knowledge of 10 lesion signs through pre-training; in the second stage, the model implements the classification task with the help of human knowledge. With the pre-training of priori knowledge of 10 lesion signs from input images, KFWC locates the powerful image features in the fine-grained disease classification task and therefore achieves better classification. To demonstrate the effectiveness of KFWC, we conduct a series of experiments on a clinical dataset collected in cooperation with a Grade III Level A ophthalmology hospital in China. The AUC score of KFWC reaches 99.71%, with 6.69% over the best baseline and 4.14% over ophthalmologists. KFWC can also provide good interpretability and effectively alleviate the pressure of data collection and annotation in the field of fine-grained disease classification for wet-AMD. The model proposed in this paper effectively solves the difficulties of small data volume and high image similarity in the wet-AMD fine-grained classification task through a knowledge-driven approach. Besides, this method effectively relieves the pressure of data collection and annotation in the field of fine-grained classification. In the diagnosis of wet-AMD, KFWC is superior to previous work and human ophthalmologists. [ABSTRACT FROM AUTHOR]

This paper develops a hierarchical deep learning machine (HDLM) to efficiently achieve both quantitative and qualitative online transient stability prediction (TSP). For the sake of improving its online efficiency, multiple generators’ fault-on trajectories as well as the two closest data-points in pre-/post-fault stages are acquired by PMUs to form its raw inputs. An anti-noise graphical transient characterization technique is tactfully designed to transform multiplex trajectories into 2-D images, within which system-wide transients are concisely described. Then, following the divide-and-conquer philosophy, the HDLM trains a two-level convolutional neural network (CNN) based regression model. With stability margin regressions hierarchically refined, it manages to perform reliable and adaptive online TSP almost immediately after fault clearance. Test results on the IEEE 39-bus test system and the real-world Guangdong Power Grid in South China demonstrate the HDLM's superior performances on both stability status and stability margin predictions. [ABSTRACT FROM AUTHOR]

Cotton (Gossypium hirsutum) is one of the most important cash crops in China, The timely and effective removal of weeds in cotton seedling stage is an important measure to ensure high and stable yield of cotton. Nowadays, weed recognition based on machine vision is widely used. The fast and effective recognition of crop and weed in the field under natural illumination is one of the key technologies for the development of intelligent mechanization weeding pattern. In the one hand, cotton and weeds have similar color feature in the field. Feature presentation of the natural property of target is difficult to be obtained by the hand-engineered feature extractor. The spatial consistency of the obtained features is not good, and the real-time performance of recognition system is reduced for the complex feature extraction algorithm. On the other hand, the effect of image preprocessing has important influence on recognition results. In order to solve the main problems in the current research, we explored the way to improve the recognition accuracy, stability and real-time performance, and a recognition method of crop and weed based on Faster R-CNN．In this paper, cotton seedling at 2-5 leaves stages and weeding during the same stage were used as research objects under natural illumination. Weed identification from digital images taken under natural illumination at field level is still challenging in agricultural image processing applications, though a lot of research has been conducted related to this topic. To address this problem, images including cottons and weeds were taken vertically from top to bottom. A method based on Faster R-CNN convolutional neural network was proposed to identify weeds from cotton plants more accurately and quickly. The residual network was used to extract image features, with ReLU as the activation function and Max-pooling as the down-sampling method. In the region of proposal network, feature pyramid network was introduced to generate target candidate frame, and Softmax regression classifier was utilized to optimize the CNN network. The proposed methodology was implemented on 200 digital images taken under natural illumination. The experimental results demonstrated that, the average accuracy of weed identification reached 95.5%, and the average time for individual weed plant identification was 1.51 s, which was reduced to 0.09 s by using GPU. To test the efficiency of the proposed methodology, YOLOv3 method was also carried out on the same training and test datasets. The weed identification results were assessed by mean average precision and average precision. The experimental results showed that better performance was achieved by using our proposed methodology, and better identification accuracy was reached as well. This indicated that the proposed method had a good effect on weed detection under natural illumination, and it will greatly promote the development of precise weed control. [ABSTRACT FROM AUTHOR]

[Display omitted] • In the case of a small number of known deposits and mineral occurrences, the prospecting prediction model based on the Siamese network can extract effective features without relying on prior knowledge to realize supervised prospecting prediction. • On the basis of sliding window data enhancement, this paper generates paired sample data sets by randomly combining deposit samples and non-deposit samples, which can further enhance the data and improve the generalization ability and prediction accuracy of the model to a certain extent. • Compared with the prediction results of the weight of evidence method and the convolutional neural network, the prediction results of the Siamese network are better, and 7 new gold prospecting areas have been discovered in Zhangjiakou, Hebei Province, China. Supervised neural networks constitute an important research area for the intelligent prediction of locations for prospecting mineral deposits. Accurate training of a supervised neural network requires many training samples, something that is often difficult to obtain. This paper reports the use of geological, geochemical, and geophysical data by a Siamese network to overcome the problem of insufficient training samples, and implements a supervised deep learning prospecting prediction method based on the Siamese network. Intelligent prediction for gold deposits prospecting is carried out for the Chongli mineral concentrate area in Zhangjiakou, Hebei Province, China, and compared with the weight of evidence method and convolutional neural network model. The results show that:(a) the performance of the Siamese network model is no less than that of the convolutional neural network (CNN) model and better than that of the weight of evidence (WOE) method; (b) the gold prospective areas differentiated by the established models are strongly consistent with geological and metallogenic characteristics in the study area. This study suggests Siamese network model as an effective mineral prospectivity modeling tool. This method is also suitable for prospecting prediction using geoscience data in other areas. [ABSTRACT FROM AUTHOR]

Cloud recognition is a basic task in ground meteorological observation. It is of great significance to accurately identify cloud types from long-time-series satellite cloud images for improving the reliability and accuracy of weather forecasting. However, different from ground-based cloud images with a small observation range and easy operation, satellite cloud images have a wider cloud coverage area and contain more surface features. Hence, it is difficult to effectively extract the structural shape, area size, contour shape, hue, shadow and texture of clouds through traditional deep learning methods. In order to analyze the regional cloud type characteristics effectively, we construct a China region meteorological satellite cloud image dataset named CRMSCD, which consists of nine cloud types and the clear sky (cloudless). In this paper, we propose a novel neural network model, UATNet, which can realize the pixel-level classification of meteorological satellite cloud images. Our model efficiently integrates the spatial and multi-channel information of clouds. Specifically, several transformer blocks with modified self-attention computation (swin transformer blocks) and patch merging operations are used to build a hierarchical transformer, and spatial displacement is introduced to construct long-distance cross-window connections. In addition, we introduce a Channel Cross fusion with Transformer (CCT) to guide the multi-scale channel fusion, and design an Attention-based Squeeze and Excitation (ASE) to effectively connect the fused multi-scale channel information to the decoder features. The experimental results demonstrate that the proposed model achieved 82.33% PA, 67.79% MPA, 54.51% MIoU and 70.96% FWIoU on CRMSCD. Compared with the existing models, our method produces more precise segmentation performance, which demonstrates its superiority on meteorological satellite cloud recognition tasks. [ABSTRACT FROM AUTHOR]

• Multi-feature & stacking were used for automatic grading cataract into six levels based on the fundus images. • Two support vector machine (SVM) classifiers are used as base-learners and fully connected neural network (FCNN) are used as meta-learner to output the final classification result. • Multi-feature & Stacking achieve higher grading performance and lower volatility than grading using high-level features and texture features. • Automatic cataract grading can be used to help government assist people more accurately in poor areas. The shortage of ophthalmologists in rural areas in China causes a lot of cataract patients not getting timely diagnosis and effective treatment. We develop an algorithm and platform to automatically diagnose and grade cataract based on fundus images of patients. This method can help government assisting poor population more accurately. The novel six-level cataract grading method proposed in this paper focuses on the multi-feature fusion based on stacking. We extract two kinds of features which can effectively distinguish different levels of cataract. One is high-level features extracted from residual network (ResNet18). The other is texture features extarcted by gray level co-occurrence matrix (GLCM). Then a frame is proposed to automatically grade cataract by the extracted features. In the frame, two support vector machine (SVM) classifiers are used as base-learners to obtain the probability outputs of each fundus image, and fully connected neural network (FCNN) are used as meta-learner to output the final classification result, which consists of two fully-connected layers. The accuracy of six-level grading achieved by the proposed method is up to 92.66% on average, the highest of which reaches 93.33%. The proposed method achieves 94.75% accuracy on four-level grading for cataract, which is at least 1.75% higher than those of the exiting methods. Six-category cataract classification algorithm show that Multi-feature & Stacking proposed in this paper helps achieve higher grading performance and lower volatility than grading using high-level features and texture features respectively. We also apply our algorithm into four-level cataract grading system and it shows higher accuracy compared with previous reports. [ABSTRACT FROM AUTHOR]

The identification of overpass structures in road networks has great significance for multi-scale modeling of roads, congestion analysis, and vehicle navigation. The traditional vector-based methods identify overpasses by the methodologies coming from computational geometry and graph theory, and they overly rely on the artificially designed features and have poor adaptability to complex scenes. This paper presents a novel method of identifying overpasses based on a target detection model (Faster-RCNN). This method utilizes raster representation of vector data and convolutional neural networks (CNNs) to learn task adaptive features from raster data, then identifies the location of an overpass by a Region Proposal network (RPN). The contribution of this paper is: (1) An overpass labelling geodatabase (OLGDB) for the OpenStreetMap (OSM) road network data of six typical cities in China is established; (2) Three different CNNs (ZF-net, VGG-16, Inception-ResNet V2) are integrated into Faster-RCNN and evaluated by accuracy performance; (3) The optimal combination of learning rate and batchsize is determined by fine-tuning; and (4) Five geometric metrics (perimeter, area, squareness, circularity, and W/L) are synthetized into image bands to enhance the training data, and their contribution to the overpass identification task is determined. The experimental results have shown that the proposed method has good accuracy performance (around 90%), and could be improved with the expansion of OLGDB and switching to more sophisticated target detection models. The deep learning target detection model has great application potential in large-scale road network pattern recognition, it can task-adaptively learn road structure features and easily extend to other road network patterns. [ABSTRACT FROM AUTHOR]

In this paper, a deep learning long-short-term memory (LSTM) method is applied to the forecasting of the critical frequency of the ionosphere F2 layer (foF2). Hourly values of foF2 from 10 ionospheric stations in China and Australia (based on availability) from 2006 to 2019 are used for training and verifying. While 2015 and 2019 are exclusive for verifying the forecasting accuracy. The inputs of the LSTM model are sequential data for the previous values, which include local time (LT), day number, solar zenith angle, the sunspot number (SSN), the daily F10.7 solar flux, geomagnetic the Ap and Kp indices, geographic coordinates, neutral winds, and the observed value of foF2 at the previous moment. To evaluate the forecasting ability of the deep learning LSTM model, two different neural network forecasting models: a back-propagation neural network (BPNN) and a genetic algorithm optimized backpropagation neural network (GABP) were established for comparative analysis. The foF2 parameters were forecasted under geomagnetic quiet and geomagnetic disturbed conditions during solar activity maximum (2015) and minimum (2019), respectively. The forecasting results of these models are compared with those of the international reference ionosphere model (IRI2016) and the measurements. The diurnal and seasonal variations of foF2 for the 4 models were compared and analyzed from 8 selected verification stations. The forecasting results reveal that the deep learning LSTM model presents the optimal performance of all models in forecasting the time series of foF2, while the IRI2016 model has the poorest forecasting performance, and the BPNN model and GABP model are between two of them. [ABSTRACT FROM AUTHOR]

Land use change (LUC) exhibits obvious spatio-temporal dependency. Previous cellular automata (CA)-based methods usually treated the LUC dynamics as Markov processes and proposed a series of CA-Markov models, which however, were intrinsically unable to capture the long-term temporal dependency. Meanwhile, such models used only numerical proportion of neighboring land use (LU) types to represent neighborhood effects of LUC, which inevitably neglected the complicated spatial heterogeneity and thus caused inaccurate simulation results. To address these problems, this paper presents a novel CA model integrated with deep learning (DL) techniques to model spatio-temporal LUC dynamics. Our DL-CA model firstly uses a convolutional neural network to capture latent spatial features for complete representation of neighborhood effects. A recurrent neural network then extracts historical information of LUC from time-series land use maps. A random forest is appended as binary change predictor to avoid the imbalanced sample problem during model training. Land use data collected from 2000 to 2014 of the Dongguan City, China were used to verify our proposed DL-CA model. The input data from 2000 to 2009 were used for model training, the 2010 data for model validation, and the data collected from 2011 to 2014 were used for model evaluation. In addition, four traditional CA models of multilayer perceptron (MLP)-CA, support vector machine (SVM)-CA, logistic regression (LR)-CA and random forest (RF)-CA were also developed for accuracy comparisons. The simulation results demonstrate that the proposed DL-CA model accurately captures long-term spatio-temporal dependency for more accurate LUC prediction results. The DL-CA model raised prediction accuracy by 9.3%–11.67% in 2011–2014 in contrast to traditional CA models. • A novel CA model is integrated with deep learning to model the dynamics of LUC. • LUC is treated as a long-term process and the temporal dependency is modeled by RNN. • For the spatial dependency in LUC, CNN is built to learn the latent spatial features. • RF is used as bootstrap to avoid imbalanced sampling during CNN-RNN model training. [ABSTRACT FROM AUTHOR]

How to acquire landslide disaster information quickly and accurately has become the focus and difficulty of disaster prevention and relief by remote sensing. Landslide disasters are generally featured by sudden occurrence, proposing high demand for emergency data acquisition. The low-altitude Unmanned Aerial Vehicle (UAV) remote sensing technology is widely applied to acquire landslide disaster data, due to its convenience, high efficiency, and ability to fly at low altitude under cloud. However, the spectrum information of UAV images is generally deficient and manual interpretation is difficult for meeting the need of quick acquisition of emergency data. Based on this, UAV images of high-occurrence areas of landslide disaster in Wenchuan County and Baoxing County in Sichuan Province, China were selected for research in the paper. Firstly, the acquired UAV images were pre-processed to generate orthoimages. Subsequently, multi-resolution segmentation was carried out to obtain image objects, and the barycenter of each object was calculated to generate a landslide sample database (including positive and negative samples) for deep learning. Next, four landslide feature models of deep learning and transfer learning, namely Histograms of Oriented Gradients (HOG), Bag of Visual Word (BOVW), Convolutional Neural Network (CNN), and Transfer Learning (TL) were compared, and it was found that the TL model possesses the best feature extraction effect, so a landslide extraction method based on the TL model and object-oriented image analysis (TLOEL) was proposed; finally, the TLOEL method was compared with the object-oriented nearest neighbor classification (NNC) method. The research results show that the accuracy of the TLOEL method is higher than the NNC method, which can not only achieve the edge extraction of large landslides, but also detect and extract middle and small landslides accurately that are scatteredly distributed. [ABSTRACT FROM AUTHOR]

This paper aims at studying the data-driven short-term provincial load forecasting (STLF) problem via an in-depth exploration of benefits brought by the feature engineering and model selection. Three core issues regarding model selections, feature selections, and feature encoding mechanism selections are deeply investigated. The candidate models are grouped into three types: the time series model, classical regression models, and the deep learning models. Three categories of features, historical loads, calendar effects, and weather factors, are considered and utilized in various encoding mechanisms. In experimental studies, an hourly provincial load dataset from Jiangsu Province in China and the corresponding weather records are utilized. The experiments are extensively performed in three parts according to model types. A time series model is conducted individually and the greedy forward wrapper-based feature selections (GFW-FS) are separately performed in six classical regression models to determine suitable encoded features. Deep learning approaches for developing STLF models are also considered. A deep neural network (DNN) model considering selected features of shallow neural networks (SNN) is developed. Meanwhile, a novel convolutional neural network (CNN) based model using GFW-FS is constructed. Through a comparative error analysis of the test set, the intrinsic linear nature among extracted features and the target in the 24-h-ahead provincial STLF problem is discovered. Feature effects are also evaluated. Data-driven models and their considered features, which are more effective to the STLF problem, are reported. • A short-term provincial load forecasting problem is studied via an in-depth analytics of data-driven models and developed features. • Classical machine learning and recent deep learning methods are applied to develop short-term provincial load forecasting models. • A greedy forward wrapper-based feature selection framework is developed to construct effective features for forecasting models. • Extensive computational experiments are conducted to explore best performed modeling methods and valuable features in the forecasting. [ABSTRACT FROM AUTHOR]

Horizontal visibility (HVIS) is a primary index used for assessing air quality. Although satellite images provide information regarding atmospheric aerosols, atmospheric visibility is not directly measured. In this paper, a deep learning approach is proposed to retrieve HVIS using moderate-resolution imaging spectroradiometer (MODIS) aerosol optical depth (AOD) data, the European Centre for Medium-Range Weather Forecasts reanalysis dataset, and ground-based visibility observations. The deep neural network model comprises a multi-layer unsupervised restricted Boltzmann machine (RBM) and a layer for supervised learning. The dropout mechanism was used in the training process to overcome the errors caused by over-fitting. The results demonstrate that the correlation coefficient values between HVIS observations and retrievals during training, pre-validating, and evaluation were 0.74, 0.723, and 0.697, respectively. The retrieved HVIS in Eastern China exhibited a north-to-south increasing trend, increasing and decreasing in summer and winter, respectively. In conclusion, the proposed model presents an effective and more reliable method for HVIS retrieval. However, the small samples, low AOD, low albedo, high total column water, high longitude, and the low vertical wind component at 10 m likely cause HVIS bias. [ABSTRACT FROM AUTHOR]

Accurate, fast, and reliable estimation of field rock strength is significant for the assessment of rock mass strength. This paper presents a novel method to continuously estimate the field strength parameters of rock by incorporating a deep convolutional neural network (DCNN) technique into the drilling process. A DCNN model is established based on the stochastic pooling method and softmax loss. The proposed CNN model framework includes : (1) collecting drilling performance parameters of intact rock, (2) establishing a strength parameter database via standard laboratory tests , (3) training the proposed CNN model using images created from the dataset of drilling performance parameters and rock strength parameters (unconfined compressive strength (UCS), cohesion, and internal friction angle), and (4) performing a new estimation to evaluate the reliability of the trained CNN model. The experimental results show that the predicted rock strength parameters are within the accepted error range of 10% compared to the results from the conventional standard test. The proposed CNN shows superior performance for UCS estimation of various rock types and achieves higher accuracy than the Mohr–Coulomb criterion. In particular, the proposed method can overcome major limitations of analytical models based on the force limit and energy equilibrium method. This method enables continuous and reliable field measurements of rock strength parameters at a speed several orders of magnitude faster than the standard test. This continuous and practical technique has been applied in an engineering problem in the Hanjiang-to-Weihe River Project of China and shows potential for field applications in rock engineering. [ABSTRACT FROM AUTHOR]

Quality management in shield tunneling projects is a challenging problem. To improve the quality of segment erection, the shield driver must constantly adjust the shield's attitude and position to reduce a snakelike motion for fitting the design tunnel axis based on the driver's experience, which is untimely and less reliable during excavation. Considering the disadvantage of this control method, this paper presents a predictive framework for the attitude and position in shield tunneling by applying a hybrid deep learning model. This framework contains a wavelet transform noise filter, convolutional neural network feature extractor, and long short-term memory predictor for determining the attitude and position of the shield machine in the future. The prediction framework is tested with the collected data of Mixshield operated in the river-crossing tunnel project of Yangtze Sanyang Road, Wuhan, China. Six variables characterizing the shield attitude and position are selected to validate the feasibility and performance of our method. Results reveal that the proposed model outperforms the other three similar models in predictive accuracy and provides decision support for adjusting the attitude and position in shield tunneling. • A new method for predicting the attitude and position of shield machine was proposed. • A hybrid deep learning model consisting of wavelet transform, CNN, and LSTM was constructed for the prediction task. • The on-site data collected from the largest river-crossing tunnel in China has been tested. • The prediction results comparing with other three models and the potential practice value were discussed. [ABSTRACT FROM AUTHOR]

Predicting the passenger flow of metro networks is of great importance for traffic management and public safety. However, such predictions are very challenging, as passenger flow is affected by complex spatial dependencies (nearby and distant) and temporal dependencies (recent and periodic). In this paper, we propose a novel deep-learning-based approach, named STGCNNmetro (spatiotemporal graph convolutional neural networks for metro), to collectively predict two types of passenger flow volumes—inflow and outflow—in each metro station of a city. Specifically, instead of representing metro stations by grids and employing conventional convolutional neural networks (CNNs) to capture spatiotemporal dependencies, STGCNNmetro transforms the city metro network to a graph and makes predictions using graph convolutional neural networks (GCNNs). First, we apply stereogram graph convolution operations to seamlessly capture the irregular spatiotemporal dependencies along the metro network. Second, a deep structure composed of GCNNs is constructed to capture the distant spatiotemporal dependencies at the citywide level. Finally, we integrate three temporal patterns (recent, daily, and weekly) and fuse the spatiotemporal dependencies captured from these patterns to form the final prediction values. The STGCNNmetro model is an end-to-end framework which can accept raw passenger flow-volume data, automatically capture the effective features of the citywide metro network, and output predictions. We test this model by predicting the short-term passenger flow volume in the citywide metro network of Shanghai, China. Experiments show that the STGCNNmetro model outperforms seven well-known baseline models (LSVR, PCA-kNN, NMF-kNN, Bayesian, MLR, M-CNN, and LSTM). We additionally explore the sensitivity of the model to its parameters and discuss the distribution of prediction errors. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach for semantic segmentation of building roofs in dense urban environments with a Deep Convolution Neural Network (DCNN) using Chinese Very High Resolution (VHR) satellite (i.e., GF2) imagery. To provide an operational end-to-end approach for accurately mapping build roofs with feature extraction and image segmentation, a fully convolutional DCNN with both convolutional and deconvolutional layers is designed to perform building roof segmentation. We selected typical cities with dense and diverse urban environments in different metropolitan regions of China as study areas, and sample images were collected over cities. High performance GPU-mounted workstations are employed to perform the model training and optimization. With the building roof samples collected over different cities, the predictive model with convolution layers is developed for building roof segmentation. The validation shows that the overall accuracy (OA) and the mean Intersection Over Union (mIOU) of DCNN-based semantic segmentation results are 94.67% and 0.85, respectively, and the CRF-refined segmentation results achieved OA of 94.69% and mIOU of 0.83. The results suggest that the proposed approach is a promising solution for building roof mapping with VHR images over large areas in dense urban environments with different building patterns. With the operational acquisition of GF2 VHR imagery, it is expected to develop an automated pipeline of operational built-up area monitoring, and the timely update of building roof map could be applied in urban management and assessment of human settlement-related sustainable development goals over large areas. [ABSTRACT FROM AUTHOR]