RESUMO
GPS-based maneuvering target localization and tracking is a crucial aspect of autonomous driving and is widely used in navigation, transportation, autonomous vehicles, and other fields.The classical tracking approach employs a Kalman filter with precise system parameters to estimate the state. However, it is difficult to model their uncertainty because of the complex motion of maneuvering targets and the unknown sensor characteristics. Furthermore, GPS data often involve unknown color noise, making it challenging to obtain accurate system parameters, which can degrade the performance of the classical methods. To address these issues, we present a state estimation method based on the Kalman filter that does not require predefined parameters but instead uses attention learning. We use a transformer encoder with a long short-term memory (LSTM) network to extract dynamic characteristics, and estimate the system model parameters online using the expectation maximization (EM) algorithm, based on the output of the attention learning module. Finally, the Kalman filter computes the dynamic state estimates using the parameters of the learned system, dynamics, and measurement characteristics. Based on GPS simulation data and the Geolife Beijing vehicle GPS trajectory dataset, the experimental results demonstrated that our method outperformed classical and pure model-free network estimation approaches in estimation accuracy, providing an effective solution for practical maneuvering-target tracking applications.
RESUMO
The environment and development are major issues of general concern. After much suffering from the harm of environmental pollution, human beings began to pay attention to environmental protection and started to carry out pollutant prediction research. A large number of air pollutant predictions have tried to predict pollutants by revealing their evolution patterns, emphasizing the fitting analysis of time series but ignoring the spatial transmission effect of adjacent areas, leading to low prediction accuracy. To solve this problem, we propose a time series prediction network with the self-optimization ability of a spatio-temporal graph neural network (BGGRU) to mine the changing pattern of the time series and the spatial propagation effect. The proposed network includes spatial and temporal modules. The spatial module uses a graph sampling and aggregation network (GraphSAGE) in order to extract the spatial information of the data. The temporal module uses a Bayesian graph gated recurrent unit (BGraphGRU), which applies a graph network to the gated recurrent unit (GRU) so as to fit the data's temporal information. In addition, this study used Bayesian optimization to solve the problem of the model's inaccuracy caused by inappropriate hyperparameters of the model. The high accuracy of the proposed method was verified by the actual PM2.5 data of Beijing, China, which provided an effective method for predicting the PM2.5 concentration.
RESUMO
The prediction of time series is of great significance for rational planning and risk prevention. However, time series data in various natural and artificial systems are nonstationary and complex, which makes them difficult to predict. An improved deep prediction method is proposed herein based on the dual variational mode decomposition of a nonstationary time series. First, criteria were determined based on information entropy and frequency statistics to determine the quantity of components in the variational mode decomposition, including the number of subsequences and the conditions for dual decomposition. Second, a deep prediction model was built for the subsequences obtained after the dual decomposition. Third, a general framework was proposed to integrate the data decomposition and deep prediction models. The method was verified on practical time series data with some contrast methods. The results show that it performed better than single deep network and traditional decomposition methods. The proposed method can effectively extract the characteristics of a nonstationary time series and obtain reliable prediction results.
RESUMO
Compared with mechanism-based modeling methods, data-driven modeling based on big data has become a popular research field in recent years because of its applicability. However, it is not always better to have more data when building a forecasting model in practical areas. Due to the noise and conflict, redundancy, and inconsistency of big time-series data, the forecasting accuracy may reduce on the contrary. This paper proposes a deep network by selecting and understanding data to improve performance. Firstly, a data self-screening layer (DSSL) with a maximal information distance coefficient (MIDC) is designed to filter input data with high correlation and low redundancy; then, a variational Bayesian gated recurrent unit (VBGRU) is used to improve the anti-noise ability and robustness of the model. Beijing's air quality and meteorological data are conducted in a verification experiment of 24 h PM2.5 concentration forecasting, proving that the proposed model is superior to other models in accuracy.
RESUMO
Time-series data generally exists in many application fields, and the classification of time-series data is one of the important research directions in time-series data mining. In this paper, univariate time-series data are taken as the research object, deep learning and broad learning systems (BLSs) are the basic methods used to explore the classification of multi-modal time-series data features. Long short-term memory (LSTM), gated recurrent unit, and bidirectional LSTM networks are used to learn and test the original time-series data, and a Gramian angular field and recurrence plot are used to encode time-series data to images, and a BLS is employed for image learning and testing. Finally, to obtain the final classification results, Dempster-Shafer evidence theory (D-S evidence theory) is considered to fuse the probability outputs of the two categories. Through the testing of public datasets, the method proposed in this paper obtains competitive results, compensating for the deficiencies of using only time-series data or images for different types of datasets.
RESUMO
State estimation is widely used in various automated systems, including IoT systems, unmanned systems, robots, etc. In traditional state estimation, measurement data are instantaneous and processed in real time. With modern systems' development, sensors can obtain more and more signals and store them. Therefore, how to use these measurement big data to improve the performance of state estimation has become a hot research issue in this field. This paper reviews the development of state estimation and future development trends. First, we review the model-based state estimation methods, including the Kalman filter, such as the extended Kalman filter (EKF), unscented Kalman filter (UKF), cubature Kalman filter (CKF), etc. Particle filters and Gaussian mixture filters that can handle mixed Gaussian noise are discussed, too. These methods have high requirements for models, while it is not easy to obtain accurate system models in practice. The emergence of robust filters, the interacting multiple model (IMM), and adaptive filters are also mentioned here. Secondly, the current research status of data-driven state estimation methods is introduced based on network learning. Finally, the main research results for hybrid filters obtained in recent years are summarized and discussed, which combine model-based methods and data-driven methods. This paper is based on state estimation research results and provides a more detailed overview of model-driven, data-driven, and hybrid-driven approaches. The main algorithm of each method is provided so that beginners can have a clearer understanding. Additionally, it discusses the future development trends for researchers in state estimation.
RESUMO
Trend prediction based on sensor data in a multi-sensor system is an important topic. As the number of sensors increases, we can measure and store more and more data. However, the increase in data has not effectively improved prediction performance. This paper focuses on this problem and presents a distributed predictor that can overcome unrelated data and sensor noise: First, we define the causality entropy to calculate the measurement's causality. Then, the series causality coefficient (SCC) is proposed to select the high causal measurement as the input data. To overcome the traditional deep learning network's over-fitting to the sensor noise, the Bayesian method is used to obtain the weight distribution characteristics of the sub-predictor network. A multi-layer perceptron (MLP) is constructed as the fusion layer to fuse the results from different sub-predictors. The experiments were implemented to verify the effectiveness of the proposed method by meteorological data from Beijing. The results show that the proposed predictor can effectively model the multi-sensor system's big measurement data to improve prediction performance.
RESUMO
Smart agricultural sensing has enabled great advantages in practical applications recently, making it one of the most important and valuable systems. For outdoor plantation farms, the prediction of climate data, such as temperature, wind speed, and humidity, enables the planning and control of agricultural production to improve the yield and quality of crops. However, it is not easy to accurately predict climate trends because the sensing data are complex, nonlinear, and contain multiple components. This study proposes a hybrid deep learning predictor, in which an empirical mode decomposition (EMD) method is used to decompose the climate data into fixed component groups with different frequency characteristics, then a gated recurrent unit (GRU) network is trained for each group as the sub-predictor, and finally the results from the GRU are added to obtain the prediction result. Experiments based on climate data from an agricultural Internet of Things (IoT) system verify the development of the proposed model. The prediction results show that the proposed predictor can obtain more accurate predictions of temperature, wind speed, and humidity data to meet the needs of precision agricultural production.
Assuntos
Agricultura , Aprendizado Profundo , Produtos Agrícolas , TemperaturaRESUMO
Intelligence has been considered as the major challenge in promoting economic potential and production efficiency of precision agriculture. In order to apply advanced deep-learning technology to complete various agricultural tasks in online and offline ways, a large number of crop vision datasets with domain-specific annotation are urgently needed. To encourage further progress in challenging realistic agricultural conditions, we present the CropDeep species classification and detection dataset, consisting of 31,147 images with over 49,000 annotated instances from 31 different classes. In contrast to existing vision datasets, images were collected with different cameras and equipment in greenhouses, captured in a wide variety of situations. It features visually similar species and periodic changes with more representative annotations, which have supported a stronger benchmark for deep-learning-based classification and detection. To further verify the application prospect, we provide extensive baseline experiments using state-of-the-art deep-learning classification and detection models. Results show that current deep-learning-based methods achieve well performance in classification accuracy over 99%. While current deep-learning methods achieve only 92% detection accuracy, illustrating the difficulty of the dataset and improvement room of state-of-the-art deep-learning models when applied to crops production and management. Specifically, we suggest that the YOLOv3 network has good potential application in agricultural detection tasks.
Assuntos
Agricultura/métodos , Aprendizado Profundo , Algoritmos , Aprendizado de Máquina , Redes Neurais de ComputaçãoRESUMO
In this paper, a novel semi-supervised segmentation framework based on a spot-divergence supervoxelization of multi-sensor fusion data is proposed for autonomous forest machine (AFMs) applications in complex environments. Given the multi-sensor measuring system, our framework addresses three successive steps: firstly, the relationship of multi-sensor coordinates is jointly calibrated to form higher-dimensional fusion data. Then, spot-divergence supervoxels representing the size-change property are given to produce feature vectors covering comprehensive information of multi-sensors at a time. Finally, the Gaussian density peak clustering is proposed to segment supervoxels into sematic objects in the semi-supervised way, which non-requires parameters preset in manual. It is demonstrated that the proposed framework achieves a balancing act both for supervoxel generation and sematic segmentation. Comparative experiments show that the well performance of segmenting various objects in terms of segmentation accuracy (F-score up to 95.6%) and operation time, which would improve intelligent capability of AFMs.
RESUMO
Online denoising is motivated by real-time applications in the industrial process, where the data must be utilizable soon after it is collected. Since the noise in practical process is usually colored, it is quite a challenge for denoising techniques. In this paper, a novel online denoising method was proposed to achieve the processing of the practical measurement data with colored noise, and the characteristics of the colored noise were considered in the dynamic model via an adaptive parameter. The proposed method consists of two parts within a closed loop: the first one is to estimate the system state based on the second-order adaptive statistics model and the other is to update the adaptive parameter in the model using the Yule-Walker algorithm. Specifically, the state estimation process was implemented via the Kalman filter in a recursive way, and the online purpose was therefore attained. Experimental data in a reinforced concrete structure test was used to verify the effectiveness of the proposed method. Results show the proposed method not only dealt with the signals with colored noise, but also achieved a tradeoff between efficiency and accuracy.
RESUMO
In this paper, a new algorithm to improve the accuracy of estimating diameter at breast height (DBH) for tree trunks in forest areas is proposed. First, the information is collected by a two-dimensional terrestrial laser scanner (2DTLS), which emits laser pulses to generate a point cloud. After extraction and filtration, the laser point clusters of the trunks are obtained, which are optimized by an arithmetic means method. Then, an algebraic circle fitting algorithm in polar form is non-linearly optimized by the Levenberg-Marquardt method to form a new hybrid algorithm, which is used to acquire the diameters and positions of the trees. Compared with previous works, this proposed method improves the accuracy of diameter estimation of trees significantly and effectively reduces the calculation time. Moreover, the experimental results indicate that this method is stable and suitable for the most challenging conditions, which has practical significance in improving the operating efficiency of forest harvester and reducing the risk of causing accidents.
RESUMO
The harvest year of maize seeds has a significant impact on seed vitality and maize yield. Therefore, it is vital to identify new seeds. In this study, an on-line near-infrared (NIR) spectra collection device (899-1715 nm) was designed and employed for distinguishing maize seeds harvested in different years. Compared with least squares support vector machine (LS-SVM), k-nearest neighbor (KNN), and extreme learning machine (ELM), the partial least squares discriminant analysis (PLS-DA) model has the optimal recognition performance for maize seed harvest years. Six different preprocessing methods, including Savitzky-Golay smoothing (SGS), standard normal variate transformation (SNV), multiplicative scatter correction (MSC), Savitzky-Golay 1 derivative (SG-D1), Savitzky-Golay 2 derivative (SG-D2), and normalization (Norm), were used to improve the quality of the spectra. The Monte Carlo cross-validation uninformative variable elimination (MC-UVE), competitive adaptive reweighted sampling (CARS), bootstrapping soft shrinkage (BOSS), successive projections algorithm (SPA), and their combinations were used to obtain effective wavelengths and decrease spectral dimensionality. The MC-UVE-BOSS-PLS-DA model achieved the classification with an accuracy of 88.75% using 93 features based on Norm preprocessed spectral data. This study showed that the self-designed NIR collection system could be used to identify the harvested years of maize seed.
RESUMO
The conservation of avian diversity plays a critical role in maintaining ecological balance and ecosystem function, as well as having a profound impact on human survival and livelihood. With species' continuous and rapid decline, information and intelligent technology have provided innovative knowledge about how functional biological diversity interacts with environmental changes. Especially in complex natural scenes, identifying bird species with a real-time and accurate pattern is vital to protect the ecological environment and maintain biodiversity changes. Aiming at the fine-grained problem in bird image recognition, this paper proposes a fine-grained detection neural network based on optimizing the YOLOV5 structure via a graph pyramid attention convolution operation. Firstly, the Cross Stage Partial (CSP) structure is introduced to a brand-new backbone classification network (GPA-Net) for significantly reducing the whole model's parameters. Then, the graph pyramid structure is applied to learn the bird image features of different scales, which enhances the fine-grained learning ability and embeds high-order features to reduce parameters. Thirdly, YOLOV5 with the soft non-maximum suppression (NMS) strategy is adopted to design the detector composition, improving the detection capability for small targets. Detailed experiments demonstrated that the proposed model achieves better or equivalent accuracy results, over-performing current advanced models in bird species identification, and is more stable and suitable for practical applications in biodiversity conservation.
Assuntos
Biodiversidade , Aves , Redes Neurais de Computação , Animais , Conservação dos Recursos NaturaisRESUMO
Introduction: Global navigation satellite system (GNSS) signals can be lost in viaducts, urban canyons, and tunnel environments. It has been a significant challenge to achieve the accurate location of pedestrians during Global Positioning System (GPS) signal outages. This paper proposes a location estimation only with inertial measurements. Methods: A method is designed based on deep network models with feature mode matching. First, a framework is designed to extract the features of inertial measurements and match them with deep networks. Second, feature extraction and classification methods are investigated to achieve mode partitioning and to lay the foundation for checking different deep networks. Third, typical deep network models are analyzed to match various features. The selected models can be trained for different modes of inertial measurements to obtain localization information. The experiments are performed with the inertial mileage dataset from Oxford University. Results and discussion: The results demonstrate that the appropriate networks based on different feature modes have more accurate position estimation, which can improve the localization accuracy of pedestrians in GPS signal outages.
RESUMO
The primary objective of document-level event extraction is to extract relevant event information from lengthy texts. However, many existing methods for document-level event extraction fail to fully incorporate the contextual information that spans across sentences. To overcome this limitation, the present study proposes a document-level event extraction model called Integration Across Texts and Reciprocal Feedback (IATRF). The proposed model constructs a heterogeneous graph and employs a graph convolutional network to enhance the connection between document and entity information. This approach facilitates the acquisition of semantic information enriched with document-level context. Additionally, a Transformer classifier is introduced to transform multiple event types into a multi-label classification task. To tackle the challenge of event argument recognition, this paper introduces the Reciprocal Feedback Argument Extraction strategy. Experimental results conducted on both our COSM dataset and the publicly available ChFinAnn dataset demonstrate that the proposed model outperforms previous methods in terms of F1 value, thus confirming its effectiveness. The IATRF model effectively solves the problems of long-distance document context-aware representation and cross-sentence argument dispersion.
RESUMO
Heavy metal contamination in wheat not only endangers human health, but also causes crop quality degradation, leads to economic losses and affects social stability. Therefore, this paper proposes a Pyraformer-based model to predict the safety risk level of Chinese wheat contaminated with heavy metals. First, based on the heavy metal sampling data of wheat and the dietary consumption data of residents, a wheat risk level dataset was constructed using the risk evaluation method; a data-driven approach was used to classify the dataset into risk levels using the K-Means++ clustering algorithm; and, finally, on the constructed dataset, Pyraformer was used to predict the risk assessment indicator and, thus, the risk level. In this paper, the proposed model was compared to the constructed dataset, and for the dataset with the lowest risk level, the precision and recall of this model still reached more than 90%, which was 25.38-4.15% and 18.42-5.26% higher, respectively. The model proposed in this paper provides a technical means for hierarchical management and early warning of heavy metal contamination of wheat in China, and also provides a scientific basis for dynamic monitoring and integrated prevention of heavy metal contamination of wheat in farmland.
RESUMO
Diseases and pests are essential threat factors that affect agricultural production, food security supply, and ecological plant diversity. However, the accurate recognition of various diseases and pests is still challenging for existing advanced information and intelligence technologies. Disease and pest recognition is typically a fine-grained visual classification problem, which is easy to confuse the traditional coarse-grained methods due to the external similarity between different categories and the significant differences among each subsample of the same category. Toward this end, this paper proposes an effective graph-related high-order network with feature aggregation enhancement (GHA-Net) to handle the fine-grained image recognition of plant pests and diseases. In our approach, an improved CSP-stage backbone network is first formed to offer massive channel-shuffled features in multiple granularities. Secondly, relying on the multilevel attention mechanism, the feature aggregation enhancement module is designed to exploit distinguishable fine-grained features representing different discriminating parts. Meanwhile, the graphic convolution module is constructed to analyse the graph-correlated representation of part-specific interrelationships by regularizing semantic features into the high-order tensor space. With the collaborative learning of three modules, our approach can grasp the robust contextual details of diseases and pests for better fine-grained identification. Extensive experiments on several public fine-grained disease and pest datasets demonstrate that the proposed GHA-Net achieves better performances in accuracy and efficiency surpassing several other existing models and is more suitable for fine-grained identification applications in complex scenes.
Assuntos
Redes Neurais de Computação , Semântica , Coleta de DadosRESUMO
Rice is a major food crop around the world, and its various quality and safety problems are closely related to human health. As an important area of food safety research, the rice supply chain has attracted increasing attention. Based on blockchain technology, this study investigated problems of data privacy and circulation efficiency caused by complex rice supply networks, long circulation cycles, and various risk factors in each link. First, we deconstructed the quality and safety of each link of the rice supply chain at the information level and established a key information classification table for each link. On that basis, we built a rice supply chain information supervision model based on blockchain. Various encryption algorithms are used to secure the sensitive data of enterprises in the supply chain to meet regulators' needs for efficient supervision. Moreover, we propose a practical Byzantine fault-tolerant consensus algorithm that scores the credit of enterprise nodes, optimizes the selection strategy of master nodes, and ensures high efficiency and low cost. Then, we built a prototype system based on the open-source framework of hyperledger fabric, analyzed the model's viability, and implemented the system using cases. The results indicated that the proposed system can optimize the information supervision process of rice supply chain regulators and provide a feasible solution for the quality and safety supervision of grain and oil.
Assuntos
Blockchain , Oryza , Algoritmos , Humanos , TecnologiaRESUMO
The nonstationary time series is generated in various natural and man-made systems, of which the prediction is vital for advanced control and management. The neural networks have been explored in the time series prediction, but the problem remains in modeling the data's nonstationary and nonlinear features. Referring to the time series feature and network property, a novel network is designed with dynamic optimization of the model structure. Firstly, the echo state network (ESN) is introduced into the broad learning system (BLS). The broad echo state network (BESN) can increase the training efficiency with the incremental learning algorithm by removing the error backpropagation. Secondly, an optimization algorithm is proposed to reduce the redundant information in the training process of BESN units. The number of neurons in BESN with a fixed step size is pruned according to the contribution degree. Finally, the improved network is applied in the different datasets. The tests in the time series of natural and man-made systems prove that the proposed network performs better on the nonstationary time series prediction than the typical methods, including the ESN, BLS, and recurrent neural network.