Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automa...Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automation has greatly increased across industries worldwide.Real-time detection requires edge devices with limited computational time.This study proposes a novel hybrid deep learning system for human activity recognition(HAR),aiming to enhance the recognition accuracy and reduce the computational time.The proposed system combines a pretrained image classification model with a sequence analysis model.First,the dataset was divided into a training set(70%),validation set(10%),and test set(20%).Second,all the videos were converted into frames and deep-based features were extracted from each frame using convolutional neural networks(CNNs)with a vision transformer.Following that,bidirectional long short-term memory(BiLSTM)-and temporal convolutional network(TCN)-based models were trained using the training set,and their performances were evaluated using the validation set and test set.Four benchmark datasets(UCF11,UCF50,UCF101,and JHMDB)were used to evaluate the performance of the proposed HAR-based system.The experimental results showed that the combination of ConvNeXt and the TCN-based model achieved a recognition accuracy of 97.73%for UCF11,98.81%for UCF50,98.46%for UCF101,and 83.38%for JHMDB,respectively.This represents improvements in the recognition accuracy of 4%,2.67%,3.67%,and 7.08%for the UCF11,UCF50,UCF101,and JHMDB datasets,respectively,over existing models.Moreover,the proposed HAR-based system obtained superior recognition accuracy,shorter computational times,and minimal memory usage compared to the existing models.展开更多
Existing through-wall human activity recognition methods often rely on Doppler information or reflective signal characteristics of the human body.However,static individuals,lacking prominent motion features,do not gen...Existing through-wall human activity recognition methods often rely on Doppler information or reflective signal characteristics of the human body.However,static individuals,lacking prominent motion features,do not generate Doppler information.Moreover,radar signals experience significant attenuation due to absorption and scattering effects as they penetrate walls,limiting recognition performance.To address these challenges,this study proposes a novel through-wall human activity recognition method based on MIMO radar.Utilizing a MIMO radar operating at 1–2 GHz,we capture activity data of individuals through walls and process it into range-angle maps to represent activity features.To tackle the issue of minimal variation in reflection areas caused by static individuals,a multi-scale activity feature extraction module is designed,capable of extracting effective features from radar signals across multiple scales.Simultaneously,a temporal attention mechanism is employed to extract keyframe information from sequential signals,focusing on critical moments of activity.Furthermore,this study introduces an activity recognition network based on a Deformable Transformer,which efficiently extracts both global and local features from radar signals,delivering precise human posture and activity sequences.In experimental scenarios involving 24 cm-thick brick walls,the proposed method achieves an impressive 97.1%accuracy in activity recognition classification.展开更多
The Turpan-Hami(Tuha)Basin of China,a critical region on the Silk Road Economic Belt and a major national energy base,occupies a significant position in energy security and in the major industrial clusters in Xinjiang...The Turpan-Hami(Tuha)Basin of China,a critical region on the Silk Road Economic Belt and a major national energy base,occupies a significant position in energy security and in the major industrial clusters in Xinjiang Uygur Autonomous Region,China.Understanding spatial and temporal evolution of human activities in this area is essential for harmonizing ecological protection with energy development,safeguarding the ecological security of the Silk Road Economic Belt,and promoting the sustainable development of the area.However,despite rapid socioeconomic advances,the trajectories of human activity intensity and the principal driving mechanisms over the past three decades remain inadequately understood.To address these gaps,this study constructed a land use dataset for the Tuha Basin from 1990 to 2020,utilizing Google Earth Engine(GEE)and random forest classification algorithm.We assessed the intensity of human activities and their spatial autocorrelation patterns and further identified key drivers influencing spatial and temporal variations using the Geodetector model.Our findings indicated that the intensity of human activities in the Tuha Basin has exhibited a"first decline and then recovery"trend over the past 30 a,accompanied by significant spatial clustering.In recent years,the aggregation of hot spots has diminished,while clustering of cold spots has intensified,suggesting a dispersion of human activity centers.Nevertheless,urban areas in the Hami and Turpan cities,along with their surrounding areas,continued to serve as core areas of human activities.Topographic features(slope gradient and aspect)and their interactions with economic variables emerged as dominant determinants shaping the spatial patterns and temporal dynamics of human activity intensity.This result provides critical insights into fostering sustainable regional development and ecological conservation in the Tuha Basin and offers valuable methodological and empirical references for studies on land use dynamics and human activity intensity in similar arid areas.展开更多
Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computation...Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computational complexity,limited generalizability under varying conditions,and compromised real-time performance.To counter these,this paper introduces an Active Learning-aided Heuristic Deep Spatio-Textural Ensemble Learning(ALH-DSEL)framework.The model initially identifies keyframes from the surveillance videos with a Multi-Constraint Active Learning(MCAL)approach,with features extracted from DenseNet121.The frames are then segmented employing an optimized Fuzzy C-Means clustering algorithm with Firefly to identify areas of interest.A deep ensemble feature extractor,comprising DenseNet121,EfficientNet-B7,MobileNet,and GLCM,extracts varied spatial and textural features.Fused characteristics are enhanced through PCA and Min-Max normalization and discriminated by a maximum voting ensemble of RF,AdaBoost,and XGBoost.The experimental results show that ALH-DSEL provides higher accuracy,precision,recall,and F1-score,validating its superiority for real-time HAR in surveillance scenarios.展开更多
This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, ...This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, MotionSense, and PAMAP2—to develop a generalized MLP model for classifying six human activities. Performance analysis of the fused model for each dataset reveals accuracy rates of 95.83 for WISDM, 97 for DaLiAc, 94.65 for MotionSense, and 98.54 for PAMAP2. A comparative evaluation was conducted between the fused MLP model and the individual dataset models, with the latter tested on separate validation sets. The results indicate that the MLP model, trained on the fused dataset, exhibits superior performance relative to the models trained on individual datasets. This finding suggests that multisource data fusion significantly enhances the generalization and accuracy of HAR systems. The improved performance underscores the potential of integrating diverse data sources to create more robust and comprehensive models for activity recognition.展开更多
Due to long-term human activity interference,the Hainan Tropical Rainforest National Park(HTRNP)of China has experienced ecological problems such as habitat fragmentation and biodiversity loss,and with the expanding s...Due to long-term human activity interference,the Hainan Tropical Rainforest National Park(HTRNP)of China has experienced ecological problems such as habitat fragmentation and biodiversity loss,and with the expanding scope and intensity of human activity impact,the regional ecological security is facing serious challenges.A scientific assessment of the interrelationship between human activity intensity and habitat quality in the HTRNP is a prerequisite for achieving effective management of ecological disturbances caused by human activities and can also provide scientific strategies for the sustainable development of the region.Based on the land use change data in 2000,2010,and 2020,the spatial and temporal variations and the relationship between habitat quality(HQ)and human activity intensity(HAI)in the HTRNP were explored using the integrated valuation of ecosystem services and trade-offs(InVEST)model.System dynamics and land use simulation models were also combined to conduct multi-scenario simulations of their relationships.The results showed that during 2000–2020,the habitat quality of the HTRNP improved,the intensity of human activities decreased each year,and there was a negative correlation between the two.Second,the system dynamic model could be well coupled with the land use simulation model by combining socio-economic and natural factors.The simulation scenarios of the coupling model showed that the harmonious development(HD)scenario is effective in curbing the increasing trend of human activity intensity and decreasing trend of habitat quality,with a weaker trade-off between the two compared with the baseline development(BD)and investment priority oriented(IPO)scenarios.To maintain the authenticity and integrity of the HTRNP,effective measures such as ecological corridor construction,ecological restoration,and the implementation of ecological compensation policies need to be strengthened.展开更多
Human activities in a transborder watershed are complex under the influence of domestic policies,international relations,and global events.Understanding the forces driving human activity change is important for the de...Human activities in a transborder watershed are complex under the influence of domestic policies,international relations,and global events.Understanding the forces driving human activity change is important for the development of transborder watershed.In this study,we used global historical land cover data,the hemeroby index model,and synthesized major historical events to analyze how human activity intensity changed in the Heilongjiang River(Amur River in Russia)watershed(HLRW).The results showed that there was a strong spatial heterogeneity in the variation of human activity intensity in the HLRW over the past century(1900-2016).On the Chinese side,the human activity intensity change shifted from the plain areas for agricultural reclamation to the mountainous areas for timber extraction.On the Russian side,human activity intensity changes mostly concentrated along the Trans-Siberian Railway and the Baikal-Amur Mainline.Localized variation of human activity intensity tended to respond to regional events while regionalized variation tends to reflect national policy change or broad international events.The similarities and differences between China and Russia in policies and positions in international events resulted in synchronous and asynchronous changes in human activity intensity.Meanwhile,policy shifts were often confined by the natural features of the watershed.These results reveal the historical origins and fundamental connotations of watershed development and contribute to formulating regional management policies that coordinate population,eco-nomic,social,and environmental activities.展开更多
Artificial intelligence(AI)technology has become integral in the realm of medicine and healthcare,particularly in human activity recognition(HAR)applications such as fitness and rehabilitation tracking.This study intr...Artificial intelligence(AI)technology has become integral in the realm of medicine and healthcare,particularly in human activity recognition(HAR)applications such as fitness and rehabilitation tracking.This study introduces a robust coupling analysis framework that integrates four AI-enabled models,combining both machine learning(ML)and deep learning(DL)approaches to evaluate their effectiveness in HAR.The analytical dataset comprises 561 features sourced from the UCI-HAR database,forming the foundation for training the models.Additionally,the MHEALTH database is employed to replicate the modeling process for comparative purposes,while inclusion of the WISDM database,renowned for its challenging features,supports the framework’s resilience and adaptability.The ML-based models employ the methodologies including adaptive neuro-fuzzy inference system(ANFIS),support vector machine(SVM),and random forest(RF),for data training.In contrast,a DL-based model utilizes one-dimensional convolution neural network(1dCNN)to automate feature extraction.Furthermore,the recursive feature elimination(RFE)algorithm,which drives an ML-based estimator to eliminate low-participation features,helps identify the optimal features for enhancing model performance.The best accuracies of the ANFIS,SVM,RF,and 1dCNN models with meticulous featuring process achieve around 90%,96%,91%,and 93%,respectively.Comparative analysis using the MHEALTH dataset showcases the 1dCNN model’s remarkable perfect accuracy(100%),while the RF,SVM,and ANFIS models equipped with selected features achieve accuracies of 99.8%,99.7%,and 96.5%,respectively.Finally,when applied to the WISDM dataset,the DL-based and ML-based models attain accuracies of 91.4%and 87.3%,respectively,aligning with prior research findings.In conclusion,the proposed framework yields HAR models with commendable performance metrics,exhibiting its suitability for integration into the healthcare services system through AI-driven applications.展开更多
RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still...RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still,they have shortcomings:1)requiring complex hand-crafted data cleaning processes and 2)only addressing single-person activity recognition based on specific RF signals.To solve these problems,this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM.This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing.Concretely,we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes singlehuman activities and human-to-human interactions.Compared with existing CNN-and LSTM-based methods,the Transformer-based method has more data fitting power,generalization,and scalability.Furthermore,using RF signals,our method achieves an excellent classification effect on human behaviorbased classification tasks.Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy(99.1%).The dataset we collected for detecting RFID-based indoor human activities will be published.展开更多
The rapidly advancing Convolutional Neural Networks(CNNs)have brought about a paradigm shift in various computer vision tasks,while also garnering increasing interest and application in sensor-based Human Activity Rec...The rapidly advancing Convolutional Neural Networks(CNNs)have brought about a paradigm shift in various computer vision tasks,while also garnering increasing interest and application in sensor-based Human Activity Recognition(HAR)efforts.However,the significant computational demands and memory requirements hinder the practical deployment of deep networks in resource-constrained systems.This paper introduces a novel network pruning method based on the energy spectral density of data in the frequency domain,which reduces the model’s depth and accelerates activity inference.Unlike traditional pruning methods that focus on the spatial domain and the importance of filters,this method converts sensor data,such as HAR data,to the frequency domain for analysis.It emphasizes the low-frequency components by calculating their energy spectral density values.Subsequently,filters that meet the predefined thresholds are retained,and redundant filters are removed,leading to a significant reduction in model size without compromising performance or incurring additional computational costs.Notably,the proposed algorithm’s effectiveness is empirically validated on a standard five-layer CNNs backbone architecture.The computational feasibility and data sensitivity of the proposed scheme are thoroughly examined.Impressively,the classification accuracy on three benchmark HAR datasets UCI-HAR,WISDM,and PAMAP2 reaches 96.20%,98.40%,and 92.38%,respectively.Concurrently,our strategy achieves a reduction in Floating Point Operations(FLOPs)by 90.73%,93.70%,and 90.74%,respectively,along with a corresponding decrease in memory consumption by 90.53%,93.43%,and 90.05%.展开更多
Human Activity Recognition (HAR) is an important way for lower limb exoskeleton robots to implement human-computer collaboration with users. Most of the existing methods in this field focus on a simple scenario recogn...Human Activity Recognition (HAR) is an important way for lower limb exoskeleton robots to implement human-computer collaboration with users. Most of the existing methods in this field focus on a simple scenario recognizing activities for specific users, which does not consider the individual differences among users and cannot adapt to new users. In order to improve the generalization ability of HAR model, this paper proposes a novel method that combines the theories in transfer learning and active learning to mitigate the cross-subject issue, so that it can enable lower limb exoskeleton robots being used in more complex scenarios. First, a neural network based on convolutional neural networks (CNN) is designed, which can extract temporal and spatial features from sensor signals collected from different parts of human body. It can recognize human activities with high accuracy after trained by labeled data. Second, in order to improve the cross-subject adaptation ability of the pre-trained model, we design a cross-subject HAR algorithm based on sparse interrogation and label propagation. Through leave-one-subject-out validation on two widely-used public datasets with existing methods, our method achieves average accuracies of 91.77% on DSAD and 80.97% on PAMAP2, respectively. The experimental results demonstrate the potential of implementing cross-subject HAR for lower limb exoskeleton robots.展开更多
One exciting area within computer vision is classifying human activities, which has diverse applications like medical informatics, human-computer interaction, surveillance, and task monitoring systems. In the healthca...One exciting area within computer vision is classifying human activities, which has diverse applications like medical informatics, human-computer interaction, surveillance, and task monitoring systems. In the healthcare field, understanding and classifying patients’ activities is crucial for providing doctors with essential information for medication reactions and diagnosis. While some research methods already exist, utilizing machine learning and soft computational algorithms to recognize human activity from videos and images, there’s ongoing exploration of more advanced computer vision techniques. This paper introduces a straightforward and effective automated approach that involves five key steps: preprocessing, feature extraction technique, feature selection, feature fusion, and finally classification. To evaluate the proposed approach, two commonly used benchmark datasets KTH and Weizmann are employed for training, validation, and testing of ML classifiers. The study’s findings show that the first and second datasets had remarkable accuracy rates of 99.94% and 99.80%, respectively. When compared to existing methods, our approach stands out in terms of sensitivity, accuracy, precision, and specificity evaluation metrics. In essence, this paper demonstrates a practical method for automatically classifying human activities using an optimal feature fusion and deep learning approach, promising a great result that could benefit various fields, particularly in healthcare.展开更多
Human activity intensity is a synthesis index for describing the effects and influences of human activities on land surface. This paper presents the concept of human activity intensity of land surface and construction...Human activity intensity is a synthesis index for describing the effects and influences of human activities on land surface. This paper presents the concept of human activity intensity of land surface and construction land equivalent, builds an algorithm model for human activity intensity, and establishes a method for converting different land use/cover types into construction land equivalent as well. An application in China based on the land use data from 1984 to 2008 is also included. The results show that China＇s human activity intensity rose slowly before 2000, while rapidly after 2000. It experienced an increase from 7.63% in 1984 to 8.54% in 2008. It could be generally divided into five levels： Very High, High, Medium, Low, and Very Low, according to the human activity intensity at county level in 2008, which is rated by above 27%, 16%-27%, 10%-16%, 6%-10%, and below 6%. China＇s human activity intensity was spatially split into eastern and western parts by the line of Helan Mountains-Longmen Mountains-Jinghong. The eastern part was characterized by the levels of Very High, High, and Medium, and the levels of Low and Very Low were zonally distributed in the mountainous and hilly areas. In contrast, the western part was featured by the Low and Very Low levels, and the levels of Medium and High were scattered in Gansu Hexi Corridor, the east of Qinghai, and the northern and southern slopes of Tianshan Mountains in Xinjiang.展开更多
Human activity recognition is commonly used in several Internet of Things applications to recognize different contexts and respond to them.Deep learning has gained momentum for identifying activities through sensors,s...Human activity recognition is commonly used in several Internet of Things applications to recognize different contexts and respond to them.Deep learning has gained momentum for identifying activities through sensors,smartphones or even surveillance cameras.However,it is often difficult to train deep learning models on constrained IoT devices.The focus of this paper is to propose an alternative model by constructing a Deep Learning-based Human Activity Recognition framework for edge computing,which we call DL-HAR.The goal of this framework is to exploit the capabilities of cloud computing to train a deep learning model and deploy it on less-powerful edge devices for recognition.The idea is to conduct the training of the model in the Cloud and distribute it to the edge nodes.We demonstrate how the DL-HAR can perform human activity recognition at the edge while improving efficiency and accuracy.In order to evaluate the proposed framework,we conducted a comprehensive set of experiments to validate the applicability of DL-HAR.Experimental results on the benchmark dataset show a significant increase in performance compared with the state-of-the-art models.展开更多
Changes in the status of freshwater resources are a topic of major global, regional and local concern. This is especially so in the arid and semi-arid regions of China, where shortage of water resources plays a crucia...Changes in the status of freshwater resources are a topic of major global, regional and local concern. This is especially so in the arid and semi-arid regions of China, where shortage of water resources plays a crucial role in limiting sustainable socioeconomic development, as well as in sustaining natural ecosystems. Recent climate change, as well as the effects of localized human activity, such as the use of water for irrigation agriculture, may have significant effects on the status of the water resources in the region. Here, we report the results of a study of changes in the areas of lakes in Gonghe Basin, northeastern Tibetan Plateau of China, over the last 60 years. The data were acquired from optical satellite images and demonstrate that the total water area of lakes in Gonghe Basin decreased significantly from the 1950s to 1980s. The cause is ascribed mainly to human activity including exploitation of farmland, against a background of increasing population; in addition, climatic data for the region demonstrate a minor drying trend during this period as the temperature increased slightly. After the construction of several reservoirs, significant amounts of water were redistributed to promote irrigation agriculture and we conclude that this caused a significant shrinkage of the natural lakes. However, both the area of farmland and the population size remained approximately constant after 1990. We conclude that the variation of the total area of lakes during the second period was mainly controlled by climatic factors （precipitation and temperature）. As the regional temperature reached a new high, the area of some of the lakes decreased sharply before finally maintaining a relatively steady state. We emphasize that anthropogenic climate change and human activity have both significantly influenced the status of water resources in the arid and semi-arid regions of China.展开更多
Research on the spatio-temporal correlation between the intensity of human activities and the temperature of earth surfaces is of great significance in many aspects,including fully understanding the causes and mechani...Research on the spatio-temporal correlation between the intensity of human activities and the temperature of earth surfaces is of great significance in many aspects,including fully understanding the causes and mechanisms of climate change,actively adapting to climate change,pursuing rational development,and protecting the ecological environment.Taking the north slope of Tianshan Mountains,located in the arid area of northwestern China and extremely sensitive to climate change,as the research area,this study retrieves the surface temperature of the mountain based on MODIS data,while characterizing the intensity of human activities thereby data on the night light,population distribution and land use.The evolution characteristics of human activity intensity and surface temperature in the study area from 2000 to 2018 were analyzed,and the spatio-temporal correlation between them was further explored.It is found that:(1)The average human activity intensity(0.11)in the research area has kept relatively low since this century,and the overall trend has been slowly rising in a stepwise manner(0.0024·a-1);in addition,the increase in human activity intensity has lagged behind that in construction land and population by 1-2 years.(2)The annual average surface temperature in the area is 7.18℃with a pronounced growth.The rate of change(0.02℃·a-1)is about 2.33 times that of the world.The striking boost in spring(0.068℃·a-1)contributes the most to the overall warming trend.Spatially,the surface temperature is low in the south and high in the north,due to the prominent influence of the underlying surface characteristics,such as elevation and vegetation coverage.(3)The intensity of human activity and the surface temperature are remarkably positively correlated in the human activity areas there,showing a strong distribution in the east section and a weak one in the west section.The expression of its spatial differentiation and correlation is comprehensively affected by such factors as scopes of human activities,manifestations,and land-use changes.Vegetation-related human interventions,such as agriculture and forestry planting,urban greening,and afforestation,can effectively reduce the surface warming caused by human activities.This study not only puts forward new ideas to finely portray the intensity of human activities but also offers a scientific reference for regional human-land coordination and overall development.展开更多
This paper proposes a hybrid approach for recognizing human activities from trajectories. First, an improved hidden Markov model （HMM） parameter learning algorithm, HMM-PSO, is proposed, which achieves a better bala...This paper proposes a hybrid approach for recognizing human activities from trajectories. First, an improved hidden Markov model （HMM） parameter learning algorithm, HMM-PSO, is proposed, which achieves a better balance between the global and local exploitation by the nonlinear update strategy and repulsion operation. Then, the event probability sequence （EPS） which consists of a series of events is computed to describe the unique characteristic of human activities. The anatysis on EPS indicates that it is robust to the changes in viewing direction and contributes to improving the recognition rate. Finally, the effectiveness of the proposed approach is evaluated by data experiments on current popular datasets.展开更多
Human Activity Recognition(HAR)is an active research area due to its applications in pervasive computing,human-computer interaction,artificial intelligence,health care,and social sciences.Moreover,dynamic environments...Human Activity Recognition(HAR)is an active research area due to its applications in pervasive computing,human-computer interaction,artificial intelligence,health care,and social sciences.Moreover,dynamic environments and anthropometric differences between individuals make it harder to recognize actions.This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications.It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network.Moreover,the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information.Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction.For temporal sequence,this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short TermMemory(BiLSTM)to capture longtermdependencies.Two state-of-the-art datasets,UCF101 and HMDB51,are used for evaluation purposes.In addition,seven state-of-the-art optimizers are used to fine-tune the proposed network parameters.Furthermore,this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network(CNN),where two streams use RGB data.In contrast,the other uses optical flow images.Finally,the proposed ensemble approach using max hard voting outperforms state-ofthe-art methods with 96.30%and 90.07%accuracies on the UCF101 and HMDB51 datasets.展开更多
After defining landslide and debris flow, human activity, and precipitation indices, using with landslide and debris flow disaster data in low-latitude plateau of China, reflecting human activity and precipitation dat...After defining landslide and debris flow, human activity, and precipitation indices, using with landslide and debris flow disaster data in low-latitude plateau of China, reflecting human activity and precipitation data, the influence of human activity and precipitation on mid-long term evolution of landslide and debris flow was studied with the wavelet technique. Results indicate that mid-long evolution of landslide and debris flow disaster trends to increase 0.9 unit every year, and presents obvious stage feature. The abrupt point from rare to frequent periods took place in 1993. There is significant in-phase resonance oscillation between human activity and landslide and debris flow frequency on a scale of 11-16 years, in which the variation of human activity occurs about 0.2-2.8 years before landslide and debris flow variation. Thus, the increase of landslide and debris flow frequency in low latitude plateau of China may be mainly caused by geo-environmental degradation induced by human activity. After the impact of human activity is removed, there is sig- nificant in-phase resonance oscillation between landslide and debris flow frequency and summer rainfall in low-latitude plateau of China in quasi-three-year and quasi-six-year scales, in which the variation of summer precipitation occurs about 0.0-0.8 years before landslide and debris flow variation. Summer precipitation is one of important external causes which impacts landslide and debris flow frequency in low-latitude plateau of China. The mid-long term evolution predicting model of landslide and debris flow disasters frequency in low-latitude plateau region with better fitting and predicting ability was built by considering human activity and summer rainfall.展开更多
Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artificial Intelligence(AI)techni-ques.These techniques are applied in several areas like security,surveillance,...Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artificial Intelligence(AI)techni-ques.These techniques are applied in several areas like security,surveillance,healthcare,human-robot interaction,and entertainment.Since wearable sensor-based HAR system includes in-built sensors,human activities can be categorized based on sensor values.Further,it can also be employed in other applications such as gait diagnosis,observation of children/adult’s cognitive nature,stroke-patient hospital direction,Epilepsy and Parkinson’s disease examination,etc.Recently-developed Artificial Intelligence(AI)techniques,especially Deep Learning(DL)models can be deployed to accomplish effective outcomes on HAR process.With this motivation,the current research paper focuses on designing Intelligent Hyperparameter Tuned Deep Learning-based HAR(IHPTDL-HAR)technique in healthcare environment.The proposed IHPTDL-HAR technique aims at recogniz-ing the human actions in healthcare environment and helps the patients in mana-ging their healthcare service.In addition,the presented model makes use of Hierarchical Clustering(HC)-based outlier detection technique to remove the out-liers.IHPTDL-HAR technique incorporates DL-based Deep Belief Network(DBN)model to recognize the activities of users.Moreover,Harris Hawks Opti-mization(HHO)algorithm is used for hyperparameter tuning of DBN model.Finally,a comprehensive experimental analysis was conducted upon benchmark dataset and the results were examined under different aspects.The experimental results demonstrate that the proposed IHPTDL-HAR technique is a superior per-former compared to other recent techniques under different measures.展开更多
基金funded by the Ongoing Research Funding Program(ORF-2025-890),King Saud University,Riyadh,Saudi Arabia.
文摘Activity recognition is a challenging topic in the field of computer vision that has various applications,including surveillance systems,industrial automation,and human-computer interaction.Today,the demand for automation has greatly increased across industries worldwide.Real-time detection requires edge devices with limited computational time.This study proposes a novel hybrid deep learning system for human activity recognition(HAR),aiming to enhance the recognition accuracy and reduce the computational time.The proposed system combines a pretrained image classification model with a sequence analysis model.First,the dataset was divided into a training set(70%),validation set(10%),and test set(20%).Second,all the videos were converted into frames and deep-based features were extracted from each frame using convolutional neural networks(CNNs)with a vision transformer.Following that,bidirectional long short-term memory(BiLSTM)-and temporal convolutional network(TCN)-based models were trained using the training set,and their performances were evaluated using the validation set and test set.Four benchmark datasets(UCF11,UCF50,UCF101,and JHMDB)were used to evaluate the performance of the proposed HAR-based system.The experimental results showed that the combination of ConvNeXt and the TCN-based model achieved a recognition accuracy of 97.73%for UCF11,98.81%for UCF50,98.46%for UCF101,and 83.38%for JHMDB,respectively.This represents improvements in the recognition accuracy of 4%,2.67%,3.67%,and 7.08%for the UCF11,UCF50,UCF101,and JHMDB datasets,respectively,over existing models.Moreover,the proposed HAR-based system obtained superior recognition accuracy,shorter computational times,and minimal memory usage compared to the existing models.
基金supported by National Natural Science Foundation of China(No.62272242)Postgraduate Research&Practice Innovation Program of Jiangsu Province(Nos.KYCX21_0800,KYCX23_1082).
文摘Existing through-wall human activity recognition methods often rely on Doppler information or reflective signal characteristics of the human body.However,static individuals,lacking prominent motion features,do not generate Doppler information.Moreover,radar signals experience significant attenuation due to absorption and scattering effects as they penetrate walls,limiting recognition performance.To address these challenges,this study proposes a novel through-wall human activity recognition method based on MIMO radar.Utilizing a MIMO radar operating at 1–2 GHz,we capture activity data of individuals through walls and process it into range-angle maps to represent activity features.To tackle the issue of minimal variation in reflection areas caused by static individuals,a multi-scale activity feature extraction module is designed,capable of extracting effective features from radar signals across multiple scales.Simultaneously,a temporal attention mechanism is employed to extract keyframe information from sequential signals,focusing on critical moments of activity.Furthermore,this study introduces an activity recognition network based on a Deformable Transformer,which efficiently extracts both global and local features from radar signals,delivering precise human posture and activity sequences.In experimental scenarios involving 24 cm-thick brick walls,the proposed method achieves an impressive 97.1%accuracy in activity recognition classification.
基金supported by the Third Xinjiang Scientific Expedition Program(2022xjkk1205)the National Natural Science Foundation of China(42377461).
文摘The Turpan-Hami(Tuha)Basin of China,a critical region on the Silk Road Economic Belt and a major national energy base,occupies a significant position in energy security and in the major industrial clusters in Xinjiang Uygur Autonomous Region,China.Understanding spatial and temporal evolution of human activities in this area is essential for harmonizing ecological protection with energy development,safeguarding the ecological security of the Silk Road Economic Belt,and promoting the sustainable development of the area.However,despite rapid socioeconomic advances,the trajectories of human activity intensity and the principal driving mechanisms over the past three decades remain inadequately understood.To address these gaps,this study constructed a land use dataset for the Tuha Basin from 1990 to 2020,utilizing Google Earth Engine(GEE)and random forest classification algorithm.We assessed the intensity of human activities and their spatial autocorrelation patterns and further identified key drivers influencing spatial and temporal variations using the Geodetector model.Our findings indicated that the intensity of human activities in the Tuha Basin has exhibited a"first decline and then recovery"trend over the past 30 a,accompanied by significant spatial clustering.In recent years,the aggregation of hot spots has diminished,while clustering of cold spots has intensified,suggesting a dispersion of human activity centers.Nevertheless,urban areas in the Hami and Turpan cities,along with their surrounding areas,continued to serve as core areas of human activities.Topographic features(slope gradient and aspect)and their interactions with economic variables emerged as dominant determinants shaping the spatial patterns and temporal dynamics of human activity intensity.This result provides critical insights into fostering sustainable regional development and ecological conservation in the Tuha Basin and offers valuable methodological and empirical references for studies on land use dynamics and human activity intensity in similar arid areas.
文摘Human Activity Recognition(HAR)has become increasingly critical in civic surveillance,medical care monitoring,and institutional protection.Current deep learning-based approaches often suffer from excessive computational complexity,limited generalizability under varying conditions,and compromised real-time performance.To counter these,this paper introduces an Active Learning-aided Heuristic Deep Spatio-Textural Ensemble Learning(ALH-DSEL)framework.The model initially identifies keyframes from the surveillance videos with a Multi-Constraint Active Learning(MCAL)approach,with features extracted from DenseNet121.The frames are then segmented employing an optimized Fuzzy C-Means clustering algorithm with Firefly to identify areas of interest.A deep ensemble feature extractor,comprising DenseNet121,EfficientNet-B7,MobileNet,and GLCM,extracts varied spatial and textural features.Fused characteristics are enhanced through PCA and Min-Max normalization and discriminated by a maximum voting ensemble of RF,AdaBoost,and XGBoost.The experimental results show that ALH-DSEL provides higher accuracy,precision,recall,and F1-score,validating its superiority for real-time HAR in surveillance scenarios.
基金supported by the Royal Golden Jubilee(RGJ)Ph.D.Programme(Grant No.PHD/0079/2561)through the National Research Council of Thailand(NRCT)and Thailand Research Fund(TRF).
文摘This research investigates the application of multisource data fusion using a Multi-Layer Perceptron (MLP) for Human Activity Recognition (HAR). The study integrates four distinct open-source datasets—WISDM, DaLiAc, MotionSense, and PAMAP2—to develop a generalized MLP model for classifying six human activities. Performance analysis of the fused model for each dataset reveals accuracy rates of 95.83 for WISDM, 97 for DaLiAc, 94.65 for MotionSense, and 98.54 for PAMAP2. A comparative evaluation was conducted between the fused MLP model and the individual dataset models, with the latter tested on separate validation sets. The results indicate that the MLP model, trained on the fused dataset, exhibits superior performance relative to the models trained on individual datasets. This finding suggests that multisource data fusion significantly enhances the generalization and accuracy of HAR systems. The improved performance underscores the potential of integrating diverse data sources to create more robust and comprehensive models for activity recognition.
基金Under the auspices of the National Social Science Found of China(No.21XGL019)Hainan Provincial Natural Science Foundation of China(No.421RC1034)Professor/Doctor Research Foundation of Huizhou University(No.2022JB080)。
文摘Due to long-term human activity interference,the Hainan Tropical Rainforest National Park(HTRNP)of China has experienced ecological problems such as habitat fragmentation and biodiversity loss,and with the expanding scope and intensity of human activity impact,the regional ecological security is facing serious challenges.A scientific assessment of the interrelationship between human activity intensity and habitat quality in the HTRNP is a prerequisite for achieving effective management of ecological disturbances caused by human activities and can also provide scientific strategies for the sustainable development of the region.Based on the land use change data in 2000,2010,and 2020,the spatial and temporal variations and the relationship between habitat quality(HQ)and human activity intensity(HAI)in the HTRNP were explored using the integrated valuation of ecosystem services and trade-offs(InVEST)model.System dynamics and land use simulation models were also combined to conduct multi-scenario simulations of their relationships.The results showed that during 2000–2020,the habitat quality of the HTRNP improved,the intensity of human activities decreased each year,and there was a negative correlation between the two.Second,the system dynamic model could be well coupled with the land use simulation model by combining socio-economic and natural factors.The simulation scenarios of the coupling model showed that the harmonious development(HD)scenario is effective in curbing the increasing trend of human activity intensity and decreasing trend of habitat quality,with a weaker trade-off between the two compared with the baseline development(BD)and investment priority oriented(IPO)scenarios.To maintain the authenticity and integrity of the HTRNP,effective measures such as ecological corridor construction,ecological restoration,and the implementation of ecological compensation policies need to be strengthened.
基金Under the auspices of National Key Research and Development Program of China(No.2017YFA0604403)National Natural Science Foundation of China(No.41801108)。
文摘Human activities in a transborder watershed are complex under the influence of domestic policies,international relations,and global events.Understanding the forces driving human activity change is important for the development of transborder watershed.In this study,we used global historical land cover data,the hemeroby index model,and synthesized major historical events to analyze how human activity intensity changed in the Heilongjiang River(Amur River in Russia)watershed(HLRW).The results showed that there was a strong spatial heterogeneity in the variation of human activity intensity in the HLRW over the past century(1900-2016).On the Chinese side,the human activity intensity change shifted from the plain areas for agricultural reclamation to the mountainous areas for timber extraction.On the Russian side,human activity intensity changes mostly concentrated along the Trans-Siberian Railway and the Baikal-Amur Mainline.Localized variation of human activity intensity tended to respond to regional events while regionalized variation tends to reflect national policy change or broad international events.The similarities and differences between China and Russia in policies and positions in international events resulted in synchronous and asynchronous changes in human activity intensity.Meanwhile,policy shifts were often confined by the natural features of the watershed.These results reveal the historical origins and fundamental connotations of watershed development and contribute to formulating regional management policies that coordinate population,eco-nomic,social,and environmental activities.
基金funded by the National Science and Technology Council,Taiwan(Grant No.NSTC 112-2121-M-039-001)by China Medical University(Grant No.CMU112-MF-79).
文摘Artificial intelligence(AI)technology has become integral in the realm of medicine and healthcare,particularly in human activity recognition(HAR)applications such as fitness and rehabilitation tracking.This study introduces a robust coupling analysis framework that integrates four AI-enabled models,combining both machine learning(ML)and deep learning(DL)approaches to evaluate their effectiveness in HAR.The analytical dataset comprises 561 features sourced from the UCI-HAR database,forming the foundation for training the models.Additionally,the MHEALTH database is employed to replicate the modeling process for comparative purposes,while inclusion of the WISDM database,renowned for its challenging features,supports the framework’s resilience and adaptability.The ML-based models employ the methodologies including adaptive neuro-fuzzy inference system(ANFIS),support vector machine(SVM),and random forest(RF),for data training.In contrast,a DL-based model utilizes one-dimensional convolution neural network(1dCNN)to automate feature extraction.Furthermore,the recursive feature elimination(RFE)algorithm,which drives an ML-based estimator to eliminate low-participation features,helps identify the optimal features for enhancing model performance.The best accuracies of the ANFIS,SVM,RF,and 1dCNN models with meticulous featuring process achieve around 90%,96%,91%,and 93%,respectively.Comparative analysis using the MHEALTH dataset showcases the 1dCNN model’s remarkable perfect accuracy(100%),while the RF,SVM,and ANFIS models equipped with selected features achieve accuracies of 99.8%,99.7%,and 96.5%,respectively.Finally,when applied to the WISDM dataset,the DL-based and ML-based models attain accuracies of 91.4%and 87.3%,respectively,aligning with prior research findings.In conclusion,the proposed framework yields HAR models with commendable performance metrics,exhibiting its suitability for integration into the healthcare services system through AI-driven applications.
基金the Strategic Priority Research Program of Chinese Academy of Sciences(Grant No.XDC02040300)for this study.
文摘RFID-based human activity recognition(HAR)attracts attention due to its convenience,noninvasiveness,and privacy protection.Existing RFID-based HAR methods use modeling,CNN,or LSTM to extract features effectively.Still,they have shortcomings:1)requiring complex hand-crafted data cleaning processes and 2)only addressing single-person activity recognition based on specific RF signals.To solve these problems,this paper proposes a novel device-free method based on Time-streaming Multiscale Transformer called TransTM.This model leverages the Transformer's powerful data fitting capabilities to take raw RFID RSSI data as input without pre-processing.Concretely,we propose a multiscale convolutional hybrid Transformer to capture behavioral features that recognizes singlehuman activities and human-to-human interactions.Compared with existing CNN-and LSTM-based methods,the Transformer-based method has more data fitting power,generalization,and scalability.Furthermore,using RF signals,our method achieves an excellent classification effect on human behaviorbased classification tasks.Experimental results on the actual RFID datasets show that this model achieves a high average recognition accuracy(99.1%).The dataset we collected for detecting RFID-based indoor human activities will be published.
基金supported by National Natural Science Foundation of China(Nos.61902158 and 62202210).
文摘The rapidly advancing Convolutional Neural Networks(CNNs)have brought about a paradigm shift in various computer vision tasks,while also garnering increasing interest and application in sensor-based Human Activity Recognition(HAR)efforts.However,the significant computational demands and memory requirements hinder the practical deployment of deep networks in resource-constrained systems.This paper introduces a novel network pruning method based on the energy spectral density of data in the frequency domain,which reduces the model’s depth and accelerates activity inference.Unlike traditional pruning methods that focus on the spatial domain and the importance of filters,this method converts sensor data,such as HAR data,to the frequency domain for analysis.It emphasizes the low-frequency components by calculating their energy spectral density values.Subsequently,filters that meet the predefined thresholds are retained,and redundant filters are removed,leading to a significant reduction in model size without compromising performance or incurring additional computational costs.Notably,the proposed algorithm’s effectiveness is empirically validated on a standard five-layer CNNs backbone architecture.The computational feasibility and data sensitivity of the proposed scheme are thoroughly examined.Impressively,the classification accuracy on three benchmark HAR datasets UCI-HAR,WISDM,and PAMAP2 reaches 96.20%,98.40%,and 92.38%,respectively.Concurrently,our strategy achieves a reduction in Floating Point Operations(FLOPs)by 90.73%,93.70%,and 90.74%,respectively,along with a corresponding decrease in memory consumption by 90.53%,93.43%,and 90.05%.
文摘Human Activity Recognition (HAR) is an important way for lower limb exoskeleton robots to implement human-computer collaboration with users. Most of the existing methods in this field focus on a simple scenario recognizing activities for specific users, which does not consider the individual differences among users and cannot adapt to new users. In order to improve the generalization ability of HAR model, this paper proposes a novel method that combines the theories in transfer learning and active learning to mitigate the cross-subject issue, so that it can enable lower limb exoskeleton robots being used in more complex scenarios. First, a neural network based on convolutional neural networks (CNN) is designed, which can extract temporal and spatial features from sensor signals collected from different parts of human body. It can recognize human activities with high accuracy after trained by labeled data. Second, in order to improve the cross-subject adaptation ability of the pre-trained model, we design a cross-subject HAR algorithm based on sparse interrogation and label propagation. Through leave-one-subject-out validation on two widely-used public datasets with existing methods, our method achieves average accuracies of 91.77% on DSAD and 80.97% on PAMAP2, respectively. The experimental results demonstrate the potential of implementing cross-subject HAR for lower limb exoskeleton robots.
文摘One exciting area within computer vision is classifying human activities, which has diverse applications like medical informatics, human-computer interaction, surveillance, and task monitoring systems. In the healthcare field, understanding and classifying patients’ activities is crucial for providing doctors with essential information for medication reactions and diagnosis. While some research methods already exist, utilizing machine learning and soft computational algorithms to recognize human activity from videos and images, there’s ongoing exploration of more advanced computer vision techniques. This paper introduces a straightforward and effective automated approach that involves five key steps: preprocessing, feature extraction technique, feature selection, feature fusion, and finally classification. To evaluate the proposed approach, two commonly used benchmark datasets KTH and Weizmann are employed for training, validation, and testing of ML classifiers. The study’s findings show that the first and second datasets had remarkable accuracy rates of 99.94% and 99.80%, respectively. When compared to existing methods, our approach stands out in terms of sensitivity, accuracy, precision, and specificity evaluation metrics. In essence, this paper demonstrates a practical method for automatically classifying human activities using an optimal feature fusion and deep learning approach, promising a great result that could benefit various fields, particularly in healthcare.
基金National Natural Science Foundation of China,No.41171449,No.41301121,No.41430636The Key Research Program of the Chinese Academy of Sciences,No.KZZD-EW-06-01
文摘Human activity intensity is a synthesis index for describing the effects and influences of human activities on land surface. This paper presents the concept of human activity intensity of land surface and construction land equivalent, builds an algorithm model for human activity intensity, and establishes a method for converting different land use/cover types into construction land equivalent as well. An application in China based on the land use data from 1984 to 2008 is also included. The results show that China＇s human activity intensity rose slowly before 2000, while rapidly after 2000. It experienced an increase from 7.63% in 1984 to 8.54% in 2008. It could be generally divided into five levels： Very High, High, Medium, Low, and Very Low, according to the human activity intensity at county level in 2008, which is rated by above 27%, 16%-27%, 10%-16%, 6%-10%, and below 6%. China＇s human activity intensity was spatially split into eastern and western parts by the line of Helan Mountains-Longmen Mountains-Jinghong. The eastern part was characterized by the levels of Very High, High, and Medium, and the levels of Low and Very Low were zonally distributed in the mountainous and hilly areas. In contrast, the western part was featured by the Low and Very Low levels, and the levels of Medium and High were scattered in Gansu Hexi Corridor, the east of Qinghai, and the northern and southern slopes of Tianshan Mountains in Xinjiang.
文摘Human activity recognition is commonly used in several Internet of Things applications to recognize different contexts and respond to them.Deep learning has gained momentum for identifying activities through sensors,smartphones or even surveillance cameras.However,it is often difficult to train deep learning models on constrained IoT devices.The focus of this paper is to propose an alternative model by constructing a Deep Learning-based Human Activity Recognition framework for edge computing,which we call DL-HAR.The goal of this framework is to exploit the capabilities of cloud computing to train a deep learning model and deploy it on less-powerful edge devices for recognition.The idea is to conduct the training of the model in the Cloud and distribute it to the edge nodes.We demonstrate how the DL-HAR can perform human activity recognition at the edge while improving efficiency and accuracy.In order to evaluate the proposed framework,we conducted a comprehensive set of experiments to validate the applicability of DL-HAR.Experimental results on the benchmark dataset show a significant increase in performance compared with the state-of-the-art models.
基金supported by the National Natural Science Foundation of China (41372180)the Open Foundation of MOE Key Laboratory of Western China’s Environmental System,Lanzhou University and the Fundamental Research Funds for the Central Universities (lzujbky-2015-bt01)
文摘Changes in the status of freshwater resources are a topic of major global, regional and local concern. This is especially so in the arid and semi-arid regions of China, where shortage of water resources plays a crucial role in limiting sustainable socioeconomic development, as well as in sustaining natural ecosystems. Recent climate change, as well as the effects of localized human activity, such as the use of water for irrigation agriculture, may have significant effects on the status of the water resources in the region. Here, we report the results of a study of changes in the areas of lakes in Gonghe Basin, northeastern Tibetan Plateau of China, over the last 60 years. The data were acquired from optical satellite images and demonstrate that the total water area of lakes in Gonghe Basin decreased significantly from the 1950s to 1980s. The cause is ascribed mainly to human activity including exploitation of farmland, against a background of increasing population; in addition, climatic data for the region demonstrate a minor drying trend during this period as the temperature increased slightly. After the construction of several reservoirs, significant amounts of water were redistributed to promote irrigation agriculture and we conclude that this caused a significant shrinkage of the natural lakes. However, both the area of farmland and the population size remained approximately constant after 1990. We conclude that the variation of the total area of lakes during the second period was mainly controlled by climatic factors （precipitation and temperature）. As the regional temperature reached a new high, the area of some of the lakes decreased sharply before finally maintaining a relatively steady state. We emphasize that anthropogenic climate change and human activity have both significantly influenced the status of water resources in the arid and semi-arid regions of China.
基金National Natural Science Foundation of China(41461086)National Natural Science Foundation of China(41761108)。
文摘Research on the spatio-temporal correlation between the intensity of human activities and the temperature of earth surfaces is of great significance in many aspects,including fully understanding the causes and mechanisms of climate change,actively adapting to climate change,pursuing rational development,and protecting the ecological environment.Taking the north slope of Tianshan Mountains,located in the arid area of northwestern China and extremely sensitive to climate change,as the research area,this study retrieves the surface temperature of the mountain based on MODIS data,while characterizing the intensity of human activities thereby data on the night light,population distribution and land use.The evolution characteristics of human activity intensity and surface temperature in the study area from 2000 to 2018 were analyzed,and the spatio-temporal correlation between them was further explored.It is found that:(1)The average human activity intensity(0.11)in the research area has kept relatively low since this century,and the overall trend has been slowly rising in a stepwise manner(0.0024·a-1);in addition,the increase in human activity intensity has lagged behind that in construction land and population by 1-2 years.(2)The annual average surface temperature in the area is 7.18℃with a pronounced growth.The rate of change(0.02℃·a-1)is about 2.33 times that of the world.The striking boost in spring(0.068℃·a-1)contributes the most to the overall warming trend.Spatially,the surface temperature is low in the south and high in the north,due to the prominent influence of the underlying surface characteristics,such as elevation and vegetation coverage.(3)The intensity of human activity and the surface temperature are remarkably positively correlated in the human activity areas there,showing a strong distribution in the east section and a weak one in the west section.The expression of its spatial differentiation and correlation is comprehensively affected by such factors as scopes of human activities,manifestations,and land-use changes.Vegetation-related human interventions,such as agriculture and forestry planting,urban greening,and afforestation,can effectively reduce the surface warming caused by human activities.This study not only puts forward new ideas to finely portray the intensity of human activities but also offers a scientific reference for regional human-land coordination and overall development.
基金supported by the National Natural Science Foundation of China(60573159)the Guangdong High Technique Project(201100000514)
文摘This paper proposes a hybrid approach for recognizing human activities from trajectories. First, an improved hidden Markov model （HMM） parameter learning algorithm, HMM-PSO, is proposed, which achieves a better balance between the global and local exploitation by the nonlinear update strategy and repulsion operation. Then, the event probability sequence （EPS） which consists of a series of events is computed to describe the unique characteristic of human activities. The anatysis on EPS indicates that it is robust to the changes in viewing direction and contributes to improving the recognition rate. Finally, the effectiveness of the proposed approach is evaluated by data experiments on current popular datasets.
基金This work was supported by financial support from Universiti Sains Malaysia(USM)under FRGS grant number FRGS/1/2020/TK03/USM/02/1the School of Computer Sciences USM for their support.
文摘Human Activity Recognition(HAR)is an active research area due to its applications in pervasive computing,human-computer interaction,artificial intelligence,health care,and social sciences.Moreover,dynamic environments and anthropometric differences between individuals make it harder to recognize actions.This study focused on human activity in video sequences acquired with an RGB camera because of its vast range of real-world applications.It uses two-stream ConvNet to extract spatial and temporal information and proposes a fine-tuned deep neural network.Moreover,the transfer learning paradigm is adopted to extract varied and fixed frames while reusing object identification information.Six state-of-the-art pre-trained models are exploited to find the best model for spatial feature extraction.For temporal sequence,this study uses dense optical flow following the two-stream ConvNet and Bidirectional Long Short TermMemory(BiLSTM)to capture longtermdependencies.Two state-of-the-art datasets,UCF101 and HMDB51,are used for evaluation purposes.In addition,seven state-of-the-art optimizers are used to fine-tune the proposed network parameters.Furthermore,this study utilizes an ensemble mechanism to aggregate spatial-temporal features using a four-stream Convolutional Neural Network(CNN),where two streams use RGB data.In contrast,the other uses optical flow images.Finally,the proposed ensemble approach using max hard voting outperforms state-ofthe-art methods with 96.30%and 90.07%accuracies on the UCF101 and HMDB51 datasets.
基金supported by National Natural Science Foundation of China(Grant No.U0933603)National Science and Technology Sup-port Program(Grant No.2011BAC09B07)
文摘After defining landslide and debris flow, human activity, and precipitation indices, using with landslide and debris flow disaster data in low-latitude plateau of China, reflecting human activity and precipitation data, the influence of human activity and precipitation on mid-long term evolution of landslide and debris flow was studied with the wavelet technique. Results indicate that mid-long evolution of landslide and debris flow disaster trends to increase 0.9 unit every year, and presents obvious stage feature. The abrupt point from rare to frequent periods took place in 1993. There is significant in-phase resonance oscillation between human activity and landslide and debris flow frequency on a scale of 11-16 years, in which the variation of human activity occurs about 0.2-2.8 years before landslide and debris flow variation. Thus, the increase of landslide and debris flow frequency in low latitude plateau of China may be mainly caused by geo-environmental degradation induced by human activity. After the impact of human activity is removed, there is sig- nificant in-phase resonance oscillation between landslide and debris flow frequency and summer rainfall in low-latitude plateau of China in quasi-three-year and quasi-six-year scales, in which the variation of summer precipitation occurs about 0.0-0.8 years before landslide and debris flow variation. Summer precipitation is one of important external causes which impacts landslide and debris flow frequency in low-latitude plateau of China. The mid-long term evolution predicting model of landslide and debris flow disasters frequency in low-latitude plateau region with better fitting and predicting ability was built by considering human activity and summer rainfall.
基金supported by Korea Institute for Advancement of Technology(KIAT)grant fundedthe Korea Government(MOTIE)(P0012724,The Competency Development Program for Industry Specialist)the Soonchunhyang University Research Fund.
文摘Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artificial Intelligence(AI)techni-ques.These techniques are applied in several areas like security,surveillance,healthcare,human-robot interaction,and entertainment.Since wearable sensor-based HAR system includes in-built sensors,human activities can be categorized based on sensor values.Further,it can also be employed in other applications such as gait diagnosis,observation of children/adult’s cognitive nature,stroke-patient hospital direction,Epilepsy and Parkinson’s disease examination,etc.Recently-developed Artificial Intelligence(AI)techniques,especially Deep Learning(DL)models can be deployed to accomplish effective outcomes on HAR process.With this motivation,the current research paper focuses on designing Intelligent Hyperparameter Tuned Deep Learning-based HAR(IHPTDL-HAR)technique in healthcare environment.The proposed IHPTDL-HAR technique aims at recogniz-ing the human actions in healthcare environment and helps the patients in mana-ging their healthcare service.In addition,the presented model makes use of Hierarchical Clustering(HC)-based outlier detection technique to remove the out-liers.IHPTDL-HAR technique incorporates DL-based Deep Belief Network(DBN)model to recognize the activities of users.Moreover,Harris Hawks Opti-mization(HHO)algorithm is used for hyperparameter tuning of DBN model.Finally,a comprehensive experimental analysis was conducted upon benchmark dataset and the results were examined under different aspects.The experimental results demonstrate that the proposed IHPTDL-HAR technique is a superior per-former compared to other recent techniques under different measures.
