Aiming at the problem that the traditional SRP-PHAT sound source localization method performs intensive search in a 360-degree space,resulting in high computational complexity and difficulty in meeting real-time requi...Aiming at the problem that the traditional SRP-PHAT sound source localization method performs intensive search in a 360-degree space,resulting in high computational complexity and difficulty in meeting real-time requirements,an innovative high-precision sound source localization method is proposed.This method combines the selective SRP-PHAT algorithm with real-time visual analysis.Its core innovations include using face detection to dynamically determine the scanning angle range to achieve visually guided selective scanning,distinguishing face sound sources from background noise through a sound source classification mechanism,and implementing intelligent background orientation selection to ensure comprehensive monitoring of environmental noise.Experimental results show that the method achieves a positioning accuracy of±5 degrees and a processing speed of more than 10FPS in complex real environments,and its performance is significantly better than the traditional full-angle scanning method.展开更多
This paper investigates the problem of decentralized multi-robot cooperative localization.This problem involves collaboratively estimating the poses of a group of robots with respect to a common reference coordinate s...This paper investigates the problem of decentralized multi-robot cooperative localization.This problem involves collaboratively estimating the poses of a group of robots with respect to a common reference coordinate system using robot-to-robot relative measurements and intermittent absolute measurements in a distributed manner.To address this problem,we present a decentralized fusion method that enables batch updating to handle relative measurements from multiple robots simultaneously.This method can improve both the accuracy and computational efficiency of cooperative localization.To reduce communication costs and reliance on connectivity,we do not maintain the inter-robot state correlations.Instead,we adopt a covariance intersection(CI)technique to design an upper bound that replaces unknown joint correlations.We propose an optimization method to determine a tight upper bound for the correlations in the joint update.The consistency and convergence of our proposed algorithm is theoretically analyzed.Furthermore,we conduct Monte Carlo numerical simulations and real-world experiments to demonstrate that the proposed method outperforms existing approaches in terms of both accuracy and consistency.展开更多
The task of food image recognition,a nuanced subset of fine-grained image recognition,grapples with substantial intra-class variation and minimal inter-class differences.These challenges are compounded by the irregula...The task of food image recognition,a nuanced subset of fine-grained image recognition,grapples with substantial intra-class variation and minimal inter-class differences.These challenges are compounded by the irregular and multi-scale nature of food images.Addressing these complexities,our study introduces an advanced model that leverages multiple attention mechanisms and multi-stage local fusion,grounded in the ConvNeXt architecture.Our model employs hybrid attention(HA)mechanisms to pinpoint critical discriminative regions within images,substantially mitigating the influence of background noise.Furthermore,it introduces a multi-stage local fusion(MSLF)module,fostering long-distance dependencies between feature maps at varying stages.This approach facilitates the assimilation of complementary features across scales,significantly bolstering the model’s capacity for feature extraction.Furthermore,we constructed a dataset named Roushi60,which consists of 60 different categories of common meat dishes.Empirical evaluation of the ETH Food-101,ChineseFoodNet,and Roushi60 datasets reveals that our model achieves recognition accuracies of 91.12%,82.86%,and 92.50%,respectively.These figures not only mark an improvement of 1.04%,3.42%,and 1.36%over the foundational ConvNeXt network but also surpass the performance of most contemporary food image recognition methods.Such advancements underscore the efficacy of our proposed model in navigating the intricate landscape of food image recognition,setting a new benchmark for the field.展开更多
The fast growth of mobile autonomous machines from traditional equipment to unmanned autonomous vehicles has fueled the demand for accurate and reliable localization solutions in diverse application domains.Ultra Wide...The fast growth of mobile autonomous machines from traditional equipment to unmanned autonomous vehicles has fueled the demand for accurate and reliable localization solutions in diverse application domains.Ultra Wide Band(UWB)technology has emerged as a promising candidate for addressing this need,offering high precision,immunity to multipath interference,and robust performance in challenging environments.In this comprehensive survey,we systematically explore UWB-based localization for mobile autonomous machines,spanning from fundamental principles to future trends.To the best of our knowledge,this review paper stands as the pioneer in systematically dissecting the algorithms of UWB-based localization for mobile autonomous machines,covering a spectrum from bottom-ranging schemes to advanced sensor fusion,error mitigation,and optimization techniques.By synthesizing existing knowledge,evaluating current methodologies,and highlighting future trends,this review aims to catalyze progress and innovation in the field,unlocking new opportunities for mobile autonomous machine applications across diverse industries and domains.Thus,it serves as a valuable resource for researchers,practitioners,and stakeholders interested in advancing the state-of-the-art UWB-based localization for mobile autonomous machines.展开更多
Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure p...Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure prompt diagnosis and effective treatment.Deep learning-based automated diagnosis for diabetic retinopathy can facilitate early detection and treatment.However,traditional deep learning models that focus on local views often learn feature representations that are less discriminative at the semantic level.On the other hand,models that focus on global semantic-level information might overlook critical,subtle local pathological features.To address this issue,we propose an adaptive multi-scale feature fusion network called(AMSFuse),which can adaptively combine multi-scale global and local features without compromising their individual representation.Specifically,our model incorporates global features for extracting high-level contextual information from retinal images.Concurrently,local features capture fine-grained details,such as microaneurysms,hemorrhages,and exudates,which are critical for DR diagnosis.These global and local features are adaptively fused using a fusion block,followed by an Integrated Attention Mechanism(IAM)that refines the fused features by emphasizing relevant regions,thereby enhancing classification accuracy for DR classification.Our model achieves 86.3%accuracy on the APTOS dataset and 96.6%RFMiD,both of which are comparable to state-of-the-art methods.展开更多
To overcome the shortcomings of 1 D and 2 D Otsu’s thresholding techniques, the 3 D Otsu method has been developed.Among all Otsu’s methods, 3 D Otsu technique provides the best threshold values for the multi-level ...To overcome the shortcomings of 1 D and 2 D Otsu’s thresholding techniques, the 3 D Otsu method has been developed.Among all Otsu’s methods, 3 D Otsu technique provides the best threshold values for the multi-level thresholding processes. In this paper, to improve the quality of segmented images, a simple and effective multilevel thresholding method is introduced. The proposed approach focuses on preserving edge detail by computing the 3 D Otsu along the fusion phenomena. The advantages of the presented scheme include higher quality outcomes, better preservation of tiny details and boundaries and reduced execution time with rising threshold levels. The fusion approach depends upon the differences between pixel intensity values within a small local space of an image;it aims to improve localized information after the thresholding process. The fusion of images based on local contrast can improve image segmentation performance by minimizing the loss of local contrast, loss of details and gray-level distributions. Results show that the proposed method yields more promising segmentation results when compared to conventional1 D Otsu, 2 D Otsu and 3 D Otsu methods, as evident from the objective and subjective evaluations.展开更多
In order to effectively reduce the uncertainty error of mobile robot localization with a single sensor and improve the accuracy and robustness of robot localization and mapping,a mobile robot localization algorithm ba...In order to effectively reduce the uncertainty error of mobile robot localization with a single sensor and improve the accuracy and robustness of robot localization and mapping,a mobile robot localization algorithm based on multi-sensor information fusion(MSIF)was proposed.In this paper,simultaneous localization and mapping(SLAM)was realized on the basis of laser Rao-Blackwellized particle filter(RBPF)-SLAM algorithm and graph-based optimization theory was used to constrain and optimize the pose estimation results of Monte Carlo localization.The feature point extraction and quadrilateral closed loop matching algorithm based on oriented FAST and rotated BRIEF(ORB)were improved aiming at the problems of generous calculation and low tracking accuracy in visual information processing by means of the three-dimensional(3D)point feature in binocular visual reconstruction environment.Factor graph model was used for the information fusion under the maximum posterior probability criterion for laser RBPF-SLAM localization and binocular visual localization.The results of simulation and experiment indicate that localization accuracy of the above-mentioned method is higher than that of traditional RBPF-SLAM algorithm and general improved algorithms,and the effectiveness and usefulness of the proposed method are verified.展开更多
The localization of the blanket jamming is studied and a new method of solving the localization ambiguity is proposed. Radars only can acquire angle information without range information when encountering the blanket ...The localization of the blanket jamming is studied and a new method of solving the localization ambiguity is proposed. Radars only can acquire angle information without range information when encountering the blanket jamming. Netted radars could get position information of the blanket jamming by make use of radars' relative position and the angle information, when there is one blanket jamming. In the presence of error, the localization method and the accuracy analysis of one blanket jamming are given. However, if there are more than one blanket jamming, and the two blanket jamming and two radars are coplanar, the localization of jamming could be error due to localization ambiguity. To solve this confusion, the Kalman filter model is established for all intersections, and through the initiation and association algorithm of multi-target, the false intersection can be eliminated. Simulations show that the presented method is valid.展开更多
For existing indoor localization algorithm has low accuracy, high cost in deployment and maintenance, lack of robustness, and low sensor utilization, this paper proposes a particle filter algorithm based on multi-sens...For existing indoor localization algorithm has low accuracy, high cost in deployment and maintenance, lack of robustness, and low sensor utilization, this paper proposes a particle filter algorithm based on multi-sensor fusion. The pedestrian’s localization in indoor environment is described as dynamic system state estimation problem. The algorithm combines the smart mobile terminal with indoor localization, and filters the result of localization with the particle filter. In this paper, a dynamic interval particle filter algorithm based on pedestrian dead reckoning (PDR) information and RSSI localization information have been used to improve the filtering precision and the stability. Moreover, the localization results will be uploaded to the server in time, and the location fingerprint database will be built incrementally, which can adapt the dynamic changes of the indoor environment. Experimental results show that the algorithm based on multi-sensor improves the localization accuracy and robustness compared with the location algorithm based on Wi-Fi.展开更多
Improved local tangent space alignment (ILTSA) is a recent nonlinear dimensionality reduction method which can efficiently recover the geometrical structure of sparse or non-uniformly distributed data manifold. In thi...Improved local tangent space alignment (ILTSA) is a recent nonlinear dimensionality reduction method which can efficiently recover the geometrical structure of sparse or non-uniformly distributed data manifold. In this paper, based on combination of modified maximum margin criterion and ILTSA, a novel feature extraction method named orthogonal discriminant improved local tangent space alignment (ODILTSA) is proposed. ODILTSA can preserve local geometry structure and maximize the margin between different classes simultaneously. Based on ODILTSA, a novel face recognition method which combines augmented complex wavelet features and original image features is developed. Experimental results on Yale, AR and PIE face databases demonstrate the effectiveness of ODILTSA and the feature fusion method.展开更多
In this paper, an innovative collaborative data fusion approach to ego-vehicle localization is presented. This approach called Optimized Kalman Swarm (OKS) is a data fusion and filtering method, fusing data from a low...In this paper, an innovative collaborative data fusion approach to ego-vehicle localization is presented. This approach called Optimized Kalman Swarm (OKS) is a data fusion and filtering method, fusing data from a low cost GPS, an INS, an Odometer and a Steering wheel angle encoder. The OKS is developed addressing the challenge of managing reactivity and robustness during a real time ego-localization process. For ego-vehicle localization, especially for highly dynamic on-road maneuvers, a filter needs to be robust and reactive at the same time. In these situations, the balance between reactivity and robustness concepts is crucial. The OKS filter represents an intelligent cooperative-reactive localization algorithm inspired by dynamic Particle Swarm Optimization (PSO). It combines advantages coming from two filters: Particle Filter (PF) and Extended Kalman filter (EKF). The OKS is tested using real embedded sensors data collected in the Satory’s test tracks. The OKS is also compared with both the well-known EKF and the Particle Filters (PF). The results show the efficiency of the OKS for a high dynamic driving scenario with damaged and low quality GPS data.展开更多
s-Lap is a new gene sequence from pig retinal pigment epithelial(RPE) cells, which was found and cloned in the early period of apoptosis of RPE cells damaged with visible light. We cloned the coding area sequence of t...s-Lap is a new gene sequence from pig retinal pigment epithelial(RPE) cells, which was found and cloned in the early period of apoptosis of RPE cells damaged with visible light. We cloned the coding area sequence of the novel gene of s-Lap and constructed its recombinant eukaryotic plasmid pcDNA3.1-GFP/s-lap with the recombinant DNA technique. The expression and localization of s-lap/GFP fusion protein in CHO and B_~16 cell lines were studied with the instantaneously transfected pcDNA3.1-GFP/s-lap recombinant plasmid. ~s-Lap/GFP fusion protein can be expressed in CHO and B_~16 cells with a high rate expression in the nuclei.展开更多
In response to the evolving challenges posed by small unmanned aerial vehicles(UAVs),which have the potential to transport harmful payloads or cause significant damage,we present AV-FDTI,an innovative Audio-Visual Fus...In response to the evolving challenges posed by small unmanned aerial vehicles(UAVs),which have the potential to transport harmful payloads or cause significant damage,we present AV-FDTI,an innovative Audio-Visual Fusion system designed for Drone Threat Identification.AV-FDTI leverages the fusion of audio and omnidirectional camera feature inputs,providing a comprehensive solution to enhance the precision and resilience of drone classification and 3D localization.Specifically,AV-FDTI employs a CRNN network to capture vital temporal dynamics within the audio domain and utilizes a pretrained ResNet50 model for image feature extraction.Furthermore,we adopt a visual information entropy and cross-attention-based mechanism to enhance the fusion of visual and audio data.Notably,our system is trained based on automated Leica tracking annotations,offering accurate ground truth data with millimeter-level accuracy.Comprehensive comparative evaluations demonstrate the superiority of our solution over the existing systems.In our commitment to advancing this field,we will release this work as open-source code and wearable AV-FDTI design,contributing valuable resources to the research community.展开更多
In this paper,an effective target locating approach based on the fingerprint fusion posi-tioning(FFP)method is proposed which integrates the time-difference of arrival(TDOA)and the received signal strength according t...In this paper,an effective target locating approach based on the fingerprint fusion posi-tioning(FFP)method is proposed which integrates the time-difference of arrival(TDOA)and the received signal strength according to the statistical variance of target position in the stationary 3D scenarios.The FFP method fuses the pedestrian dead reckoning(PDR)estimation to solve the moving target localization problem.We also introduce auxiliary parameters to estimate the target motion state.Subsequently,we can locate the static pedestrians and track the the moving target.For the case study,eight access stationary points are placed on a bookshelf and hypermarket;one target node is moving inside hypermarkets in 2D and 3D scenarios or stationary on the bookshelf.We compare the performance of our proposed method with existing localization algorithms such as k-nearest neighbor,weighted k-nearest neighbor,pure TDOA and fingerprinting combining Bayesian frameworks including the extended Kalman filter,unscented Kalman filter and particle fil-ter(PF).The proposed approach outperforms obviously the counterpart methodologies in terms of the root mean square error and the cumulative distribution function of localization errors,espe-cially in the 3D scenarios.Simulation results corroborate the effectiveness of our proposed approach.展开更多
This paper presents the Optimized Kalman Particle Swarm (OKPS) filter. This filter results from two years of research and improves the Swarm Particle Filter (SPF). The OKPS has been designed to be both cooperative and...This paper presents the Optimized Kalman Particle Swarm (OKPS) filter. This filter results from two years of research and improves the Swarm Particle Filter (SPF). The OKPS has been designed to be both cooperative and reactive. It combines the advantages of the Particle Filter (PF) and the metaheuristic Particle Swarm Optimization (PSO) for ego-vehicles localization applications. In addition to a simple fusion between the swarm optimization and the particular filtering (which leads to the Swarm Particle Filter), the OKPS uses some attributes of the Extended Kalman filter (EKF). The OKPS filter innovates by fitting its particles with a capacity of self-diagnose by means of the EKF covariance uncertainty matrix. The particles can therefore evolve by exchanging information to assess the optimized position of the ego-vehicle. The OKPS fuses data coming from embedded sensors (low cost INS, GPS and Odometer) to perform a robust ego-vehicle positioning. The OKPS is compared to the EKF filter and to filters using particles (PF and SPF) on real data from our equipped vehicle.展开更多
A gradient metamaterial with varying-stiffness local resonators is proposed to open the multiple bandgaps and further form a broad fusion bandgap.First,three local resonators with linearly increasing stiffness are per...A gradient metamaterial with varying-stiffness local resonators is proposed to open the multiple bandgaps and further form a broad fusion bandgap.First,three local resonators with linearly increasing stiffness are periodically attached to the spring-mass chain to construct the gradient metamaterial.The dispersion relation is then derived based on Bloch's theorem to reveal the fusion bandgap theoretically.The dynamic characteristic of the finite spring-mass chain is investigated to validate the fusion of multiple bandgaps.Finally,the effects of the design parameters on multiple bandgaps are discussed.The results show that the metamaterial with a non-uniform stiffness gradient pattern is capable of opening a broad fusion bandgap and effectively attenuating the longitudinal waves within a broad frequency region.展开更多
Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inc...Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inconsistent from one view to another.This study develops a deep global multiple-scale and local patches attention(GMS-LPA)dual-branch network for pose-invariant FER to weaken the influence of pose variation and selfocclusion on recognition accuracy.In this research,the designed GMS-LPA network contains four main parts,i.e.,the feature extraction module,the global multiple-scale(GMS)module,the local patches attention(LPA)module,and the model-level fusion model.The feature extraction module is designed to extract and normalize texture information to the same size.The GMS model can extract deep global features with different receptive fields,releasing the sensitivity of deeper convolution layers to pose-variant and self-occlusion.The LPA module is built to force the network to focus on local salient features,which can lower the effect of pose variation and self-occlusion on recognition results.Subsequently,the extracted features are fused with a model-level strategy to improve recognition accuracy.Extensive experimentswere conducted on four public databases,and the recognition results demonstrated the feasibility and validity of the proposed methods.展开更多
基金the research result of the 2024 Guangxi Higher Education Undergraduate Teaching Reform Project“OBE-Guided,Digitally Empowered‘Hadoop Big Data Development Technology’Course Ideological and Political Construction Innovation Exploration and Practice”(Project No.:2024JGA396).
文摘Aiming at the problem that the traditional SRP-PHAT sound source localization method performs intensive search in a 360-degree space,resulting in high computational complexity and difficulty in meeting real-time requirements,an innovative high-precision sound source localization method is proposed.This method combines the selective SRP-PHAT algorithm with real-time visual analysis.Its core innovations include using face detection to dynamically determine the scanning angle range to achieve visually guided selective scanning,distinguishing face sound sources from background noise through a sound source classification mechanism,and implementing intelligent background orientation selection to ensure comprehensive monitoring of environmental noise.Experimental results show that the method achieves a positioning accuracy of±5 degrees and a processing speed of more than 10FPS in complex real environments,and its performance is significantly better than the traditional full-angle scanning method.
文摘This paper investigates the problem of decentralized multi-robot cooperative localization.This problem involves collaboratively estimating the poses of a group of robots with respect to a common reference coordinate system using robot-to-robot relative measurements and intermittent absolute measurements in a distributed manner.To address this problem,we present a decentralized fusion method that enables batch updating to handle relative measurements from multiple robots simultaneously.This method can improve both the accuracy and computational efficiency of cooperative localization.To reduce communication costs and reliance on connectivity,we do not maintain the inter-robot state correlations.Instead,we adopt a covariance intersection(CI)technique to design an upper bound that replaces unknown joint correlations.We propose an optimization method to determine a tight upper bound for the correlations in the joint update.The consistency and convergence of our proposed algorithm is theoretically analyzed.Furthermore,we conduct Monte Carlo numerical simulations and real-world experiments to demonstrate that the proposed method outperforms existing approaches in terms of both accuracy and consistency.
基金The support of this research was by Hubei Provincial Natural Science Foundation(2022CFB449)Science Research Foundation of Education Department of Hubei Province(B2020061),are gratefully acknowledged.
文摘The task of food image recognition,a nuanced subset of fine-grained image recognition,grapples with substantial intra-class variation and minimal inter-class differences.These challenges are compounded by the irregular and multi-scale nature of food images.Addressing these complexities,our study introduces an advanced model that leverages multiple attention mechanisms and multi-stage local fusion,grounded in the ConvNeXt architecture.Our model employs hybrid attention(HA)mechanisms to pinpoint critical discriminative regions within images,substantially mitigating the influence of background noise.Furthermore,it introduces a multi-stage local fusion(MSLF)module,fostering long-distance dependencies between feature maps at varying stages.This approach facilitates the assimilation of complementary features across scales,significantly bolstering the model’s capacity for feature extraction.Furthermore,we constructed a dataset named Roushi60,which consists of 60 different categories of common meat dishes.Empirical evaluation of the ETH Food-101,ChineseFoodNet,and Roushi60 datasets reveals that our model achieves recognition accuracies of 91.12%,82.86%,and 92.50%,respectively.These figures not only mark an improvement of 1.04%,3.42%,and 1.36%over the foundational ConvNeXt network but also surpass the performance of most contemporary food image recognition methods.Such advancements underscore the efficacy of our proposed model in navigating the intricate landscape of food image recognition,setting a new benchmark for the field.
文摘The fast growth of mobile autonomous machines from traditional equipment to unmanned autonomous vehicles has fueled the demand for accurate and reliable localization solutions in diverse application domains.Ultra Wide Band(UWB)technology has emerged as a promising candidate for addressing this need,offering high precision,immunity to multipath interference,and robust performance in challenging environments.In this comprehensive survey,we systematically explore UWB-based localization for mobile autonomous machines,spanning from fundamental principles to future trends.To the best of our knowledge,this review paper stands as the pioneer in systematically dissecting the algorithms of UWB-based localization for mobile autonomous machines,covering a spectrum from bottom-ranging schemes to advanced sensor fusion,error mitigation,and optimization techniques.By synthesizing existing knowledge,evaluating current methodologies,and highlighting future trends,this review aims to catalyze progress and innovation in the field,unlocking new opportunities for mobile autonomous machine applications across diverse industries and domains.Thus,it serves as a valuable resource for researchers,practitioners,and stakeholders interested in advancing the state-of-the-art UWB-based localization for mobile autonomous machines.
基金supported by the National Natural Science Foundation of China(No.62376287)the International Science and Technology Innovation Joint Base of Machine Vision and Medical Image Processing in Hunan Province(2021CB1013)the Natural Science Foundation of Hunan Province(Nos.2022JJ30762,2023JJ70016).
文摘Globally,diabetic retinopathy(DR)is the primary cause of blindness,affecting millions of people worldwide.This widespread impact underscores the critical need for reliable and precise diagnostic techniques to ensure prompt diagnosis and effective treatment.Deep learning-based automated diagnosis for diabetic retinopathy can facilitate early detection and treatment.However,traditional deep learning models that focus on local views often learn feature representations that are less discriminative at the semantic level.On the other hand,models that focus on global semantic-level information might overlook critical,subtle local pathological features.To address this issue,we propose an adaptive multi-scale feature fusion network called(AMSFuse),which can adaptively combine multi-scale global and local features without compromising their individual representation.Specifically,our model incorporates global features for extracting high-level contextual information from retinal images.Concurrently,local features capture fine-grained details,such as microaneurysms,hemorrhages,and exudates,which are critical for DR diagnosis.These global and local features are adaptively fused using a fusion block,followed by an Integrated Attention Mechanism(IAM)that refines the fused features by emphasizing relevant regions,thereby enhancing classification accuracy for DR classification.Our model achieves 86.3%accuracy on the APTOS dataset and 96.6%RFMiD,both of which are comparable to state-of-the-art methods.
文摘To overcome the shortcomings of 1 D and 2 D Otsu’s thresholding techniques, the 3 D Otsu method has been developed.Among all Otsu’s methods, 3 D Otsu technique provides the best threshold values for the multi-level thresholding processes. In this paper, to improve the quality of segmented images, a simple and effective multilevel thresholding method is introduced. The proposed approach focuses on preserving edge detail by computing the 3 D Otsu along the fusion phenomena. The advantages of the presented scheme include higher quality outcomes, better preservation of tiny details and boundaries and reduced execution time with rising threshold levels. The fusion approach depends upon the differences between pixel intensity values within a small local space of an image;it aims to improve localized information after the thresholding process. The fusion of images based on local contrast can improve image segmentation performance by minimizing the loss of local contrast, loss of details and gray-level distributions. Results show that the proposed method yields more promising segmentation results when compared to conventional1 D Otsu, 2 D Otsu and 3 D Otsu methods, as evident from the objective and subjective evaluations.
基金Natural Science Foundation of Shaanxi Province(No.2019JQ-004)Scientific Research Plan Projects of Shaanxi Education Department(No.18JK0438)Youth Talent Promotion Project of Shaanxi Province(No.20180112)。
文摘In order to effectively reduce the uncertainty error of mobile robot localization with a single sensor and improve the accuracy and robustness of robot localization and mapping,a mobile robot localization algorithm based on multi-sensor information fusion(MSIF)was proposed.In this paper,simultaneous localization and mapping(SLAM)was realized on the basis of laser Rao-Blackwellized particle filter(RBPF)-SLAM algorithm and graph-based optimization theory was used to constrain and optimize the pose estimation results of Monte Carlo localization.The feature point extraction and quadrilateral closed loop matching algorithm based on oriented FAST and rotated BRIEF(ORB)were improved aiming at the problems of generous calculation and low tracking accuracy in visual information processing by means of the three-dimensional(3D)point feature in binocular visual reconstruction environment.Factor graph model was used for the information fusion under the maximum posterior probability criterion for laser RBPF-SLAM localization and binocular visual localization.The results of simulation and experiment indicate that localization accuracy of the above-mentioned method is higher than that of traditional RBPF-SLAM algorithm and general improved algorithms,and the effectiveness and usefulness of the proposed method are verified.
文摘The localization of the blanket jamming is studied and a new method of solving the localization ambiguity is proposed. Radars only can acquire angle information without range information when encountering the blanket jamming. Netted radars could get position information of the blanket jamming by make use of radars' relative position and the angle information, when there is one blanket jamming. In the presence of error, the localization method and the accuracy analysis of one blanket jamming are given. However, if there are more than one blanket jamming, and the two blanket jamming and two radars are coplanar, the localization of jamming could be error due to localization ambiguity. To solve this confusion, the Kalman filter model is established for all intersections, and through the initiation and association algorithm of multi-target, the false intersection can be eliminated. Simulations show that the presented method is valid.
文摘For existing indoor localization algorithm has low accuracy, high cost in deployment and maintenance, lack of robustness, and low sensor utilization, this paper proposes a particle filter algorithm based on multi-sensor fusion. The pedestrian’s localization in indoor environment is described as dynamic system state estimation problem. The algorithm combines the smart mobile terminal with indoor localization, and filters the result of localization with the particle filter. In this paper, a dynamic interval particle filter algorithm based on pedestrian dead reckoning (PDR) information and RSSI localization information have been used to improve the filtering precision and the stability. Moreover, the localization results will be uploaded to the server in time, and the location fingerprint database will be built incrementally, which can adapt the dynamic changes of the indoor environment. Experimental results show that the algorithm based on multi-sensor improves the localization accuracy and robustness compared with the location algorithm based on Wi-Fi.
基金the National Natural Science Foundation of China(No.61004088)the Key Basic Research Foundation of Shanghai Municipal Science and Technology Commission(No.09JC1408000)
文摘Improved local tangent space alignment (ILTSA) is a recent nonlinear dimensionality reduction method which can efficiently recover the geometrical structure of sparse or non-uniformly distributed data manifold. In this paper, based on combination of modified maximum margin criterion and ILTSA, a novel feature extraction method named orthogonal discriminant improved local tangent space alignment (ODILTSA) is proposed. ODILTSA can preserve local geometry structure and maximize the margin between different classes simultaneously. Based on ODILTSA, a novel face recognition method which combines augmented complex wavelet features and original image features is developed. Experimental results on Yale, AR and PIE face databases demonstrate the effectiveness of ODILTSA and the feature fusion method.
文摘In this paper, an innovative collaborative data fusion approach to ego-vehicle localization is presented. This approach called Optimized Kalman Swarm (OKS) is a data fusion and filtering method, fusing data from a low cost GPS, an INS, an Odometer and a Steering wheel angle encoder. The OKS is developed addressing the challenge of managing reactivity and robustness during a real time ego-localization process. For ego-vehicle localization, especially for highly dynamic on-road maneuvers, a filter needs to be robust and reactive at the same time. In these situations, the balance between reactivity and robustness concepts is crucial. The OKS filter represents an intelligent cooperative-reactive localization algorithm inspired by dynamic Particle Swarm Optimization (PSO). It combines advantages coming from two filters: Particle Filter (PF) and Extended Kalman filter (EKF). The OKS is tested using real embedded sensors data collected in the Satory’s test tracks. The OKS is also compared with both the well-known EKF and the Particle Filters (PF). The results show the efficiency of the OKS for a high dynamic driving scenario with damaged and low quality GPS data.
文摘s-Lap is a new gene sequence from pig retinal pigment epithelial(RPE) cells, which was found and cloned in the early period of apoptosis of RPE cells damaged with visible light. We cloned the coding area sequence of the novel gene of s-Lap and constructed its recombinant eukaryotic plasmid pcDNA3.1-GFP/s-lap with the recombinant DNA technique. The expression and localization of s-lap/GFP fusion protein in CHO and B_~16 cell lines were studied with the instantaneously transfected pcDNA3.1-GFP/s-lap recombinant plasmid. ~s-Lap/GFP fusion protein can be expressed in CHO and B_~16 cells with a high rate expression in the nuclei.
基金National Research Foundation,Singapore,under its Medium-Sized Center for Advanced Robotics Technology Innovation(CARTIN)under project WP5 within the Delta-NTU Corporate Lab with funding support from A*STAR under its IAF-ICP program(Grant no:I2201E0013)and Delta Electronics Inc.
文摘In response to the evolving challenges posed by small unmanned aerial vehicles(UAVs),which have the potential to transport harmful payloads or cause significant damage,we present AV-FDTI,an innovative Audio-Visual Fusion system designed for Drone Threat Identification.AV-FDTI leverages the fusion of audio and omnidirectional camera feature inputs,providing a comprehensive solution to enhance the precision and resilience of drone classification and 3D localization.Specifically,AV-FDTI employs a CRNN network to capture vital temporal dynamics within the audio domain and utilizes a pretrained ResNet50 model for image feature extraction.Furthermore,we adopt a visual information entropy and cross-attention-based mechanism to enhance the fusion of visual and audio data.Notably,our system is trained based on automated Leica tracking annotations,offering accurate ground truth data with millimeter-level accuracy.Comprehensive comparative evaluations demonstrate the superiority of our solution over the existing systems.In our commitment to advancing this field,we will release this work as open-source code and wearable AV-FDTI design,contributing valuable resources to the research community.
基金partially supported by the National Natural Science Foun-dation of China(No.62071389).
文摘In this paper,an effective target locating approach based on the fingerprint fusion posi-tioning(FFP)method is proposed which integrates the time-difference of arrival(TDOA)and the received signal strength according to the statistical variance of target position in the stationary 3D scenarios.The FFP method fuses the pedestrian dead reckoning(PDR)estimation to solve the moving target localization problem.We also introduce auxiliary parameters to estimate the target motion state.Subsequently,we can locate the static pedestrians and track the the moving target.For the case study,eight access stationary points are placed on a bookshelf and hypermarket;one target node is moving inside hypermarkets in 2D and 3D scenarios or stationary on the bookshelf.We compare the performance of our proposed method with existing localization algorithms such as k-nearest neighbor,weighted k-nearest neighbor,pure TDOA and fingerprinting combining Bayesian frameworks including the extended Kalman filter,unscented Kalman filter and particle fil-ter(PF).The proposed approach outperforms obviously the counterpart methodologies in terms of the root mean square error and the cumulative distribution function of localization errors,espe-cially in the 3D scenarios.Simulation results corroborate the effectiveness of our proposed approach.
文摘This paper presents the Optimized Kalman Particle Swarm (OKPS) filter. This filter results from two years of research and improves the Swarm Particle Filter (SPF). The OKPS has been designed to be both cooperative and reactive. It combines the advantages of the Particle Filter (PF) and the metaheuristic Particle Swarm Optimization (PSO) for ego-vehicles localization applications. In addition to a simple fusion between the swarm optimization and the particular filtering (which leads to the Swarm Particle Filter), the OKPS uses some attributes of the Extended Kalman filter (EKF). The OKPS filter innovates by fitting its particles with a capacity of self-diagnose by means of the EKF covariance uncertainty matrix. The particles can therefore evolve by exchanging information to assess the optimized position of the ego-vehicle. The OKPS fuses data coming from embedded sensors (low cost INS, GPS and Odometer) to perform a robust ego-vehicle positioning. The OKPS is compared to the EKF filter and to filters using particles (PF and SPF) on real data from our equipped vehicle.
基金supported by the National Natural Science Foundation of China(Nos.12122206,52175125,12272129,12304309,and 12302039)the Zhejiang Provincial Natural Science Foundation of China(No.LQ24A020006)+1 种基金the Hong Kong Scholars Program of China(No.XJ2022012)the Natural Science Foundation of Hunan Province of China(No.2024JJ4004)。
文摘A gradient metamaterial with varying-stiffness local resonators is proposed to open the multiple bandgaps and further form a broad fusion bandgap.First,three local resonators with linearly increasing stiffness are periodically attached to the spring-mass chain to construct the gradient metamaterial.The dispersion relation is then derived based on Bloch's theorem to reveal the fusion bandgap theoretically.The dynamic characteristic of the finite spring-mass chain is investigated to validate the fusion of multiple bandgaps.Finally,the effects of the design parameters on multiple bandgaps are discussed.The results show that the metamaterial with a non-uniform stiffness gradient pattern is capable of opening a broad fusion bandgap and effectively attenuating the longitudinal waves within a broad frequency region.
基金supported by the National Natural Science Foundation of China (No.31872399)Advantage Discipline Construction Project (PAPD,No.6-2018)of Jiangsu University。
文摘Pose-invariant facial expression recognition(FER)is an active but challenging research topic in computer vision.Especially with the involvement of diverse observation angles,FER makes the training parameter models inconsistent from one view to another.This study develops a deep global multiple-scale and local patches attention(GMS-LPA)dual-branch network for pose-invariant FER to weaken the influence of pose variation and selfocclusion on recognition accuracy.In this research,the designed GMS-LPA network contains four main parts,i.e.,the feature extraction module,the global multiple-scale(GMS)module,the local patches attention(LPA)module,and the model-level fusion model.The feature extraction module is designed to extract and normalize texture information to the same size.The GMS model can extract deep global features with different receptive fields,releasing the sensitivity of deeper convolution layers to pose-variant and self-occlusion.The LPA module is built to force the network to focus on local salient features,which can lower the effect of pose variation and self-occlusion on recognition results.Subsequently,the extracted features are fused with a model-level strategy to improve recognition accuracy.Extensive experimentswere conducted on four public databases,and the recognition results demonstrated the feasibility and validity of the proposed methods.
