Human body posture recognition has attracted considerable attention in recent years in wireless body area networks(WBAN). In order to precisely recognize human body posture,many recognition algorithms have been propos...Human body posture recognition has attracted considerable attention in recent years in wireless body area networks(WBAN). In order to precisely recognize human body posture,many recognition algorithms have been proposed.However, the recognition rate is relatively low. In this paper, we apply back propagation(BP) neural network as a classifier to recognizing human body posture, where signals are collected from VG350 acceleration sensor and a posture signal collection system based on WBAN is designed. Human body signal vector magnitude(SVM) and tri-axial acceleration sensor data are used to describe the human body postures. We are able to recognize 4postures: Walk, Run, Squat and Sit. Our posture recognition rate is up to 91.67%. Furthermore, we find an implied relationship between hidden layer neurons and the posture recognition rate. The proposed human body posture recognition algorithm lays the foundation for the subsequent applications.展开更多
Over the years,the continuous development of new technology has promoted research in the field of posture recognition and also made the application field of posture recognition have been greatly expanded.The purpose o...Over the years,the continuous development of new technology has promoted research in the field of posture recognition and also made the application field of posture recognition have been greatly expanded.The purpose of this paper is to introduce the latest methods of posture recognition and review the various techniques and algorithms of posture recognition in recent years,such as scale-invariant feature transform,histogram of oriented gradients,support vectormachine(SVM),Gaussian mixturemodel,dynamic time warping,hiddenMarkovmodel(HMM),lightweight network,convolutional neural network(CNN).We also investigate improved methods of CNN,such as stacked hourglass networks,multi-stage pose estimation networks,convolutional posemachines,and high-resolution nets.The general process and datasets of posture recognition are analyzed and summarized,and several improved CNNmethods and threemain recognition techniques are compared.In addition,the applications of advanced neural networks in posture recognition,such as transfer learning,ensemble learning,graph neural networks,and explainable deep neural networks,are introduced.It was found that CNN has achieved great success in posture recognition and is favored by researchers.Still,a more in-depth research is needed in feature extraction,information fusion,and other aspects.Among classification methods,HMM and SVM are the most widely used,and lightweight network gradually attracts the attention of researchers.In addition,due to the lack of 3Dbenchmark data sets,data generation is a critical research direction.展开更多
This study proposes a pose estimation-convolutional neural network-bidirectional gated recurrent unit(PSECNN-BiGRU)fusion model for human posture recognition to address low accuracy issues in abnormal posture recognit...This study proposes a pose estimation-convolutional neural network-bidirectional gated recurrent unit(PSECNN-BiGRU)fusion model for human posture recognition to address low accuracy issues in abnormal posture recognition due to the loss of some feature information and the deterioration of comprehensive performance in model detection in complex home environments.Firstly,the deep convolutional network is integrated with the Mediapipe framework to extract high-precision,multi-dimensional information from the key points of the human skeleton,thereby obtaining a human posture feature set.Thereafter,a double-layer BiGRU algorithm is utilized to extract multi-layer,bidirectional temporal features from the human posture feature set,and a CNN network with an exponential linear unit(ELU)activation function is adopted to perform deep convolution of the feature map to extract the spatial feature of the human posture.Furthermore,a squeeze and excitation networks(SENet)module is introduced to adaptively learn the importance weights of each channel,enhancing the network’s focus on important features.Finally,comparative experiments are performed on available datasets,including the public human activity recognition using smartphone dataset(UCIHAR),the public human activity recognition 70 plus dataset(HAR70PLUS),and the independently developed home abnormal behavior recognition dataset(HABRD)created by the authors’team.The results show that the average accuracy of the proposed PSE-CNN-BiGRU fusion model for human posture recognition is 99.56%,89.42%,and 98.90%,respectively,which are 5.24%,5.83%,and 3.19%higher than the average accuracy of the five models proposed in the comparative literature,including CNN,GRU,and others.The F1-score for abnormal posture recognition reaches 98.84%(heartache),97.18%(fall),99.6%(bellyache),and 98.27%(climbing)on the self-builtHABRDdataset,thus verifying the effectiveness,generalization,and robustness of the proposed model in enhancing human posture recognition.展开更多
Posture recognition plays an important role in many applications,such as security system and monitoring system.Joint quaternion combined with support vector machine(SVM) can solve the problem of moving human posture r...Posture recognition plays an important role in many applications,such as security system and monitoring system.Joint quaternion combined with support vector machine(SVM) can solve the problem of moving human posture recognition.It is a simple and effective algorithm that only three joints are used as the feature points in the whole human skeleton.Using the quaternion of the three joints,a feature vector with five parameters in gait cycle is extracted.The efficiency of the proposed method is demonstrated through an experimental study,and walking and running postures can be distinguished accurately.展开更多
In this paper, we propose a low-cost posture recognition scheme using a single webcam for the signaling hand with nature sways and possible oc-clusions. It goes for developing the untouchable low-complexity utility ba...In this paper, we propose a low-cost posture recognition scheme using a single webcam for the signaling hand with nature sways and possible oc-clusions. It goes for developing the untouchable low-complexity utility based on friendly hand-posture signaling. The scheme integrates the dominant temporal-difference detection, skin color detection and morphological filtering for efficient cooperation in constructing the hand profile molds. Those molds provide representative hand profiles for more stable posture recognition than accurate hand shapes with in effect trivial details. The resultant bounding box of tracking the signaling molds can be treated as a regular-type object-matched ROI to facilitate the stable extraction of robust HOG features. With such commonly applied features on hand, the prototype SVM is adequately capable of obtaining fast and stable hand postures recognition under natural hand movement and non-hand object occlusion. Experimental results demonstrate that our scheme can achieve hand-posture recognition with enough accuracy under background clutters that the targeted hand can be allowed with medium movement and palm-grasped object. Hence, the proposed method can be easily embedded in the mobile phone as application software.展开更多
This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophagea...This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.展开更多
With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors we...With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors were widely applied due to their low cost. This paper explored the implementation of a human hand posture recognition system using ToF sensors and residual neural networks. Firstly, this paper reviewed the typical applications of human hand recognition. Secondly, this paper designed a hand gesture recognition system using a ToF sensor VL53L5. Subsequently, data preprocessing was conducted, followed by training the constructed residual neural network. Then, the recognition results were analyzed, indicating that gesture recognition based on the residual neural network achieved an accuracy of 98.5% in a 5-class classification scenario. Finally, the paper discussed existing issues and future research directions.展开更多
Smart healthcare applications depend on data from wearable sensors(WSs)mounted on a patient’s body for frequent monitoring information.Healthcare systems depend on multi-level data for detecting illnesses and consequ...Smart healthcare applications depend on data from wearable sensors(WSs)mounted on a patient’s body for frequent monitoring information.Healthcare systems depend on multi-level data for detecting illnesses and consequently delivering correct diagnostic measures.The collection of WS data and integration of that data for diagnostic purposes is a difficult task.This paper proposes an Errorless Data Fusion(EDF)approach to increase posture recognition accuracy.The research is based on a case study in a health organization.With the rise in smart healthcare systems,WS data fusion necessitates careful attention to provide sensitive analysis of the recognized illness.As a result,it is dependent on WS inputs and performs group analysis at a similar rate to improve diagnostic efficiency.Sensor breakdowns,the constant time factor,aggregation,and analysis results all cause errors,resulting in rejected or incorrect suggestions.This paper resolves this problem by using EDF,which is related to patient situational discovery through healthcare surveillance systems.Features of WS data are examined extensively using active and iterative learning to identify errors in specific postures.This technology improves position detection accuracy,analysis duration,and error rate,regardless of user movements.Wearable devices play a critical role in the management and treatment of patients.They can ensure that patients are provided with a unique treatment for their medical needs.This paper discusses the EDF technique for optimizing posture identification accuracy through multi-feature analysis.At first,the patients’walking patterns are tracked at various time intervals.The characteristics are then evaluated in relation to the stored data using a random forest classifier.展开更多
Achieving human skin-like sensitivity and wide-range pressure detection remains a significant challenge in the developmentof wearable pressure sensors.In this study,we engineered and fabricated a fibrous polyimide fib...Achieving human skin-like sensitivity and wide-range pressure detection remains a significant challenge in the developmentof wearable pressure sensors.In this study,we engineered and fabricated a fibrous polyimide fiber(PIF)/carbon nanotube(CNT)composite aerogel with a gradient structure using a layer-by-layer freeze casting technique,aiming to overcome thelimitations of traditional pressure sensors.Finite element analysis(FEA)reveals that this innovative gradient structure mimicsthe unique microstructure of human skin,enabling the sensor to detect a broad spectrum of pressure stimuli,ranging fromsubtle pressures as low as 10 Pa to intense pressures up to 1.58 MPa with exceptional sensitivity.Moreover,the sensor exhibitsextraordinary pressure resolution across the entire pressure range,particularly at 1 MPa(0.001%).Additionally,the sensordemonstrates remarkable thermal stability,operating reliably across a wide temperature range from−150 to 200°C,makingit suitable for extreme environments such as deep space exploration.When integrated with machine learning algorithms,thesensor shows great potential for real-time physiological monitoring,fitness tracking,and motion recognition.The proposedgradient fibrous pressure sensor,with its high sensitivity and resolution over a wide pressure range,paves the way for newopportunities in human–machine interaction.展开更多
绵羊的姿态与其健康及福利密切相关。随着智能化畜牧业需求的增长,自动、准确地检测绵羊姿态尤为尤为重要。本研究提出基于Mask R-CNN基准网络的新型RDS-Mask R-CNN绵羊姿态检测算法,以Res2Net101作为特征提取网络,同时引入可变形卷积(D...绵羊的姿态与其健康及福利密切相关。随着智能化畜牧业需求的增长,自动、准确地检测绵羊姿态尤为尤为重要。本研究提出基于Mask R-CNN基准网络的新型RDS-Mask R-CNN绵羊姿态检测算法,以Res2Net101作为特征提取网络,同时引入可变形卷积(Deformable convolution network,DCN),以更精准捕捉绵羊在不同位置的姿态特征,并运用软非极大值抑制(Soft non-maximum suppression,Soft NMS)算法实现重叠实例目标的准确分割。结果表明:1)目标检测框架算法对比:与该领域最经典的YOLOv3和Faster R-CNN相比,改进的算法在平均精度均值(Mean average precision,mAP)上分别提升了16.68%和8.64%;2)不同改进策略的算法对比:改进算法相较于基准网络,边界框平均精度均值(Bounding box mean average precision,Bbox mAP)提高6.21%,分割平均精度均值(Segmentation mean average precision,Segm mAP)提高6.61%,分别达到87.34%和81.50%;3)相较于Mask R-CNN,改进模型在识别绵羊站立与躺卧姿态时边界框平均精度(Bounding box average precision,Bbox AP)分别提高了6.84%和5.58%,分割平均精度(Segmentation average precision,Segm AP)分别提高了7.25%和5.17%;4)模型可解释性可视化结果表明RDS-Mask R-CNN能精准捕获绵羊站立和躺卧姿态关键部位深度特征,表明模型自动检测可行且具有可解释性。综上,本研究提出的RDS-Mask R-CNN算法,有效提升了绵羊姿态检测的精准度,为智慧养殖提供了技术支撑。展开更多
Human action recognition and posture prediction aim to recognize and predict respectively the action and postures of persons in videos.They are both active research topics in computer vision community,which have attra...Human action recognition and posture prediction aim to recognize and predict respectively the action and postures of persons in videos.They are both active research topics in computer vision community,which have attracted considerable attention from academia and industry.They are also the precondition for intelligent interaction and human-computer cooperation,and they help the machine perceive the external environment.In the past decade,tremendous progress has been made in the field,especially after the emergence of deep learning technologies.Hence,it is necessary to make a comprehensive review of recent developments.In this paper,firstly,we attempt to present the background,and then discuss research progresses.Secondly,we introduce datasets,various typical feature representation methods,and explore advanced human action recognition and posture prediction algorithms.Finally,facing the challenges in the field,this paper puts forward the research focus,and introduces the importance of action recognition and posture prediction by taking interactive cognition in self-driving vehicle as an example.展开更多
基金supported by the National Natural Science Foundation of China(No.61074165 and No.61273064)Jilin Provincial Science&Technology Department Key Scientific and Technological Project(No.20140204034GX)Jilin Province Development and Reform Commission Project(No.2015Y043)
文摘Human body posture recognition has attracted considerable attention in recent years in wireless body area networks(WBAN). In order to precisely recognize human body posture,many recognition algorithms have been proposed.However, the recognition rate is relatively low. In this paper, we apply back propagation(BP) neural network as a classifier to recognizing human body posture, where signals are collected from VG350 acceleration sensor and a posture signal collection system based on WBAN is designed. Human body signal vector magnitude(SVM) and tri-axial acceleration sensor data are used to describe the human body postures. We are able to recognize 4postures: Walk, Run, Squat and Sit. Our posture recognition rate is up to 91.67%. Furthermore, we find an implied relationship between hidden layer neurons and the posture recognition rate. The proposed human body posture recognition algorithm lays the foundation for the subsequent applications.
基金supported by British Heart Foundation Accelerator Award,UK(AA/18/3/34220)Royal Society International Exchanges Cost Share Award,UK(RP202G0230)+7 种基金Hope Foundation for Cancer Research,UK(RM60G0680)Medical Research Council Confidence in Concept Award,UK(MC_PC_17171)Sino-UK Industrial Fund,UK(RP202G0289)Global Challenges Research Fund(GCRF),UK(P202PF11)LIAS Pioneering Partnerships award,UK(P202ED10)Data Science Enhancement Fund,UK(P202RE237)Fight for Sight,UK(24NN201)Sino-UK Education Fund,UK(OP202006).
文摘Over the years,the continuous development of new technology has promoted research in the field of posture recognition and also made the application field of posture recognition have been greatly expanded.The purpose of this paper is to introduce the latest methods of posture recognition and review the various techniques and algorithms of posture recognition in recent years,such as scale-invariant feature transform,histogram of oriented gradients,support vectormachine(SVM),Gaussian mixturemodel,dynamic time warping,hiddenMarkovmodel(HMM),lightweight network,convolutional neural network(CNN).We also investigate improved methods of CNN,such as stacked hourglass networks,multi-stage pose estimation networks,convolutional posemachines,and high-resolution nets.The general process and datasets of posture recognition are analyzed and summarized,and several improved CNNmethods and threemain recognition techniques are compared.In addition,the applications of advanced neural networks in posture recognition,such as transfer learning,ensemble learning,graph neural networks,and explainable deep neural networks,are introduced.It was found that CNN has achieved great success in posture recognition and is favored by researchers.Still,a more in-depth research is needed in feature extraction,information fusion,and other aspects.Among classification methods,HMM and SVM are the most widely used,and lightweight network gradually attracts the attention of researchers.In addition,due to the lack of 3Dbenchmark data sets,data generation is a critical research direction.
基金funded by the Henan Provincial Science and Technology Research Project(222102210086)the Starry Sky Creative Space Innovation Space Innovation Incubation Project of Zhengzhou University of Light Industry(2023ZCKJ211).
文摘This study proposes a pose estimation-convolutional neural network-bidirectional gated recurrent unit(PSECNN-BiGRU)fusion model for human posture recognition to address low accuracy issues in abnormal posture recognition due to the loss of some feature information and the deterioration of comprehensive performance in model detection in complex home environments.Firstly,the deep convolutional network is integrated with the Mediapipe framework to extract high-precision,multi-dimensional information from the key points of the human skeleton,thereby obtaining a human posture feature set.Thereafter,a double-layer BiGRU algorithm is utilized to extract multi-layer,bidirectional temporal features from the human posture feature set,and a CNN network with an exponential linear unit(ELU)activation function is adopted to perform deep convolution of the feature map to extract the spatial feature of the human posture.Furthermore,a squeeze and excitation networks(SENet)module is introduced to adaptively learn the importance weights of each channel,enhancing the network’s focus on important features.Finally,comparative experiments are performed on available datasets,including the public human activity recognition using smartphone dataset(UCIHAR),the public human activity recognition 70 plus dataset(HAR70PLUS),and the independently developed home abnormal behavior recognition dataset(HABRD)created by the authors’team.The results show that the average accuracy of the proposed PSE-CNN-BiGRU fusion model for human posture recognition is 99.56%,89.42%,and 98.90%,respectively,which are 5.24%,5.83%,and 3.19%higher than the average accuracy of the five models proposed in the comparative literature,including CNN,GRU,and others.The F1-score for abnormal posture recognition reaches 98.84%(heartache),97.18%(fall),99.6%(bellyache),and 98.27%(climbing)on the self-builtHABRDdataset,thus verifying the effectiveness,generalization,and robustness of the proposed model in enhancing human posture recognition.
基金the Key Project of the National Natural Science Foundation of China(No.61134009)National Natural Science Foundations of China(Nos.61473077,61473078,61503075)+6 种基金Cooperative Research Funds of the National Natural Science Funds Overseas and Hong Kong and Macao Scholars,China(No.61428302)Program for Changjiang Scholars from the Ministry of Education,ChinaSpecialized Research Fund for Shanghai Leading Talents,ChinaProject of the Shanghai Committee of Science and Technology,China(No.13JC1407500)Innovation Program of Shanghai Municipal Education Commission,China(No.14ZZ067)Shanghai Pujiang Program,China(No.15PJ1400100)the Fundamental Research Funds for the Central Universities,China(Nos.15D110423,2232015D3-32)
文摘Posture recognition plays an important role in many applications,such as security system and monitoring system.Joint quaternion combined with support vector machine(SVM) can solve the problem of moving human posture recognition.It is a simple and effective algorithm that only three joints are used as the feature points in the whole human skeleton.Using the quaternion of the three joints,a feature vector with five parameters in gait cycle is extracted.The efficiency of the proposed method is demonstrated through an experimental study,and walking and running postures can be distinguished accurately.
文摘In this paper, we propose a low-cost posture recognition scheme using a single webcam for the signaling hand with nature sways and possible oc-clusions. It goes for developing the untouchable low-complexity utility based on friendly hand-posture signaling. The scheme integrates the dominant temporal-difference detection, skin color detection and morphological filtering for efficient cooperation in constructing the hand profile molds. Those molds provide representative hand profiles for more stable posture recognition than accurate hand shapes with in effect trivial details. The resultant bounding box of tracking the signaling molds can be treated as a regular-type object-matched ROI to facilitate the stable extraction of robust HOG features. With such commonly applied features on hand, the prototype SVM is adequately capable of obtaining fast and stable hand postures recognition under natural hand movement and non-hand object occlusion. Experimental results demonstrate that our scheme can achieve hand-posture recognition with enough accuracy under background clutters that the targeted hand can be allowed with medium movement and palm-grasped object. Hence, the proposed method can be easily embedded in the mobile phone as application software.
基金funded by Vietnam National Foundation for Science and Technology Development(NAFOSTED)under grant number:NCUD.02-2024.11.
文摘This study presents CGB-Net,a novel deep learning architecture specifically developed for classifying twelve distinct sleep positions using a single abdominal accelerometer,with direct applicability to gastroesophageal reflux disease(GERD)monitoring.Unlike conventional approaches limited to four basic postures,CGB-Net enables fine-grained classification of twelve clinically relevant sleep positions,providing enhanced resolution for personalized health assessment.The architecture introduces a unique integration of three complementary components:1D Convolutional Neural Networks(1D-CNN)for efficient local spatial feature extraction,Gated Recurrent Units(GRU)to capture short-termtemporal dependencieswith reduced computational complexity,and Bidirectional Long Short-Term Memory(Bi-LSTM)networks for modeling long-term temporal context in both forward and backward directions.This complementary integration allows the model to better represent dynamic and contextual information inherent in the sensor data,surpassing the performance of simpler or previously published hybrid models.Experiments were conducted on a benchmark dataset consisting of 18 volunteers(age range:19–24 years,mean 20.56±1.1 years;height 164.78±8.18 cm;weight 55.39±8.30 kg;BMI 20.24±2.04),monitored via a single abdominal accelerometer.A subjectindependent evaluation protocol with multiple random splits was employed to ensure robustness and generalizability.The proposed model achieves an average Accuracy of 87.60% and F1-score of 83.38%,both reported with standard deviations over multiple runs,outperforming several baseline and state-of-the-art methods.By releasing the dataset publicly and detailing themodel design,this work aims to facilitate reproducibility and advance research in sleep posture classification for clinical applications.
文摘With the advancement of technology and the increase in user demands, gesture recognition played a pivotal role in the field of human-computer interaction. Among various sensing devices, Time-of-Flight (ToF) sensors were widely applied due to their low cost. This paper explored the implementation of a human hand posture recognition system using ToF sensors and residual neural networks. Firstly, this paper reviewed the typical applications of human hand recognition. Secondly, this paper designed a hand gesture recognition system using a ToF sensor VL53L5. Subsequently, data preprocessing was conducted, followed by training the constructed residual neural network. Then, the recognition results were analyzed, indicating that gesture recognition based on the residual neural network achieved an accuracy of 98.5% in a 5-class classification scenario. Finally, the paper discussed existing issues and future research directions.
文摘Smart healthcare applications depend on data from wearable sensors(WSs)mounted on a patient’s body for frequent monitoring information.Healthcare systems depend on multi-level data for detecting illnesses and consequently delivering correct diagnostic measures.The collection of WS data and integration of that data for diagnostic purposes is a difficult task.This paper proposes an Errorless Data Fusion(EDF)approach to increase posture recognition accuracy.The research is based on a case study in a health organization.With the rise in smart healthcare systems,WS data fusion necessitates careful attention to provide sensitive analysis of the recognized illness.As a result,it is dependent on WS inputs and performs group analysis at a similar rate to improve diagnostic efficiency.Sensor breakdowns,the constant time factor,aggregation,and analysis results all cause errors,resulting in rejected or incorrect suggestions.This paper resolves this problem by using EDF,which is related to patient situational discovery through healthcare surveillance systems.Features of WS data are examined extensively using active and iterative learning to identify errors in specific postures.This technology improves position detection accuracy,analysis duration,and error rate,regardless of user movements.Wearable devices play a critical role in the management and treatment of patients.They can ensure that patients are provided with a unique treatment for their medical needs.This paper discusses the EDF technique for optimizing posture identification accuracy through multi-feature analysis.At first,the patients’walking patterns are tracked at various time intervals.The characteristics are then evaluated in relation to the stored data using a random forest classifier.
基金National Natural Science Foundation of China(52373093)Excellent Youth Found of Natural Science Foundation of Henan Province(242300421062)Central Plains Youth Top notch Talent Program of Henan Province,and the 111 project(D18023).
文摘Achieving human skin-like sensitivity and wide-range pressure detection remains a significant challenge in the developmentof wearable pressure sensors.In this study,we engineered and fabricated a fibrous polyimide fiber(PIF)/carbon nanotube(CNT)composite aerogel with a gradient structure using a layer-by-layer freeze casting technique,aiming to overcome thelimitations of traditional pressure sensors.Finite element analysis(FEA)reveals that this innovative gradient structure mimicsthe unique microstructure of human skin,enabling the sensor to detect a broad spectrum of pressure stimuli,ranging fromsubtle pressures as low as 10 Pa to intense pressures up to 1.58 MPa with exceptional sensitivity.Moreover,the sensor exhibitsextraordinary pressure resolution across the entire pressure range,particularly at 1 MPa(0.001%).Additionally,the sensordemonstrates remarkable thermal stability,operating reliably across a wide temperature range from−150 to 200°C,makingit suitable for extreme environments such as deep space exploration.When integrated with machine learning algorithms,thesensor shows great potential for real-time physiological monitoring,fitness tracking,and motion recognition.The proposedgradient fibrous pressure sensor,with its high sensitivity and resolution over a wide pressure range,paves the way for newopportunities in human–machine interaction.
文摘绵羊的姿态与其健康及福利密切相关。随着智能化畜牧业需求的增长,自动、准确地检测绵羊姿态尤为尤为重要。本研究提出基于Mask R-CNN基准网络的新型RDS-Mask R-CNN绵羊姿态检测算法,以Res2Net101作为特征提取网络,同时引入可变形卷积(Deformable convolution network,DCN),以更精准捕捉绵羊在不同位置的姿态特征,并运用软非极大值抑制(Soft non-maximum suppression,Soft NMS)算法实现重叠实例目标的准确分割。结果表明:1)目标检测框架算法对比:与该领域最经典的YOLOv3和Faster R-CNN相比,改进的算法在平均精度均值(Mean average precision,mAP)上分别提升了16.68%和8.64%;2)不同改进策略的算法对比:改进算法相较于基准网络,边界框平均精度均值(Bounding box mean average precision,Bbox mAP)提高6.21%,分割平均精度均值(Segmentation mean average precision,Segm mAP)提高6.61%,分别达到87.34%和81.50%;3)相较于Mask R-CNN,改进模型在识别绵羊站立与躺卧姿态时边界框平均精度(Bounding box average precision,Bbox AP)分别提高了6.84%和5.58%,分割平均精度(Segmentation average precision,Segm AP)分别提高了7.25%和5.17%;4)模型可解释性可视化结果表明RDS-Mask R-CNN能精准捕获绵羊站立和躺卧姿态关键部位深度特征,表明模型自动检测可行且具有可解释性。综上,本研究提出的RDS-Mask R-CNN算法,有效提升了绵羊姿态检测的精准度,为智慧养殖提供了技术支撑。
基金supported by the National Natural Science Foundation of China(Nos.61871038 and 61931012)the Premium Funding Project for Academic Human Resources Development of Beijing Union University(No.BPHR2020AZ02)the Generic Pre-research Program of the Equipment Development Department in Military Commission(No.41412040302).
文摘Human action recognition and posture prediction aim to recognize and predict respectively the action and postures of persons in videos.They are both active research topics in computer vision community,which have attracted considerable attention from academia and industry.They are also the precondition for intelligent interaction and human-computer cooperation,and they help the machine perceive the external environment.In the past decade,tremendous progress has been made in the field,especially after the emergence of deep learning technologies.Hence,it is necessary to make a comprehensive review of recent developments.In this paper,firstly,we attempt to present the background,and then discuss research progresses.Secondly,we introduce datasets,various typical feature representation methods,and explore advanced human action recognition and posture prediction algorithms.Finally,facing the challenges in the field,this paper puts forward the research focus,and introduces the importance of action recognition and posture prediction by taking interactive cognition in self-driving vehicle as an example.
