Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–...Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–PM_(2.5)and the limitations of existing algorithms pose a significant challenge in realizing the accurate joint retrieval of these two parameters at the same location.On this point,a multi-task learning(MTL)model,which enables the joint retrieval of PM_(2.5)concentration and AOD,is proposed and applied on the top-of-the-atmosphere reflectance data gathered by the Fengyun-4A Advanced Geosynchronous Radiation Imager(FY-4A AGRI),and compared to that of two single-task learning models—namely,Random Forest(RF)and Deep Neural Network(DNN).Specifically,MTL achieves a coefficient of determination(R^(2))of 0.88 and a root-mean-square error(RMSE)of 0.10 in AOD retrieval.In comparison to RF,the R^(2)increases by 0.04,the RMSE decreases by 0.02,and the percentage of retrieval results falling within the expected error range(Within-EE)rises by 5.55%.The R^(2)and RMSE of PM_(2.5)retrieval by MTL are 0.84 and 13.76μg m~(-3)respectively.Compared with RF,the R^(2)increases by 0.06,the RMSE decreases by 4.55μg m~(-3),and the Within-EE increases by 7.28%.Additionally,compared to DNN,MTL shows an increase of 0.01 in R^(2)and a decrease of 0.02 in RMSE in AOD retrieval,with a corresponding increase of 2.89%in Within-EE.For PM_(2.5)retrieval,MTL exhibits an increase of 0.05 in R^(2),a decrease of 1.76μg m~(-3)in RMSE,and an increase of 6.83%in Within-EE.The evaluation suggests that MTL is able to provide simultaneously improved AOD and PM_(2.5)retrievals,demonstrating a significant advantage in efficiently capturing the spatial distribution of PM_(2.5)concentration and AOD.展开更多
Medical institutions frequently utilize cloud servers for storing digital medical imaging data, aiming to lower both storage expenses and computational expenses. Nevertheless, the reliability of cloud servers as third...Medical institutions frequently utilize cloud servers for storing digital medical imaging data, aiming to lower both storage expenses and computational expenses. Nevertheless, the reliability of cloud servers as third-party providers is not always guaranteed. To safeguard against the exposure and misuse of personal privacy information, and achieve secure and efficient retrieval, a secure medical image retrieval based on a multi-attention mechanism and triplet deep hashing is proposed in this paper (abbreviated as MATDH). Specifically, this method first utilizes the contrast-limited adaptive histogram equalization method applicable to color images to enhance chest X-ray images. Next, a designed multi-attention mechanism focuses on important local features during the feature extraction stage. Moreover, a triplet loss function is utilized to learn discriminative hash codes to construct a compact and efficient triplet deep hashing. Finally, upsampling is used to restore the original resolution of the images during retrieval, thereby enabling more accurate matching. To ensure the security of medical image data, a lightweight image encryption method based on frequency domain encryption is designed to encrypt the chest X-ray images. The findings of the experiment indicate that, in comparison to various advanced image retrieval techniques, the suggested approach improves the precision of feature extraction and retrieval using the COVIDx dataset. Additionally, it offers enhanced protection for the confidentiality of medical images stored in cloud settings and demonstrates strong practicality.展开更多
Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasi...Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasing volume of visual-language pre-training model parameters,direct transfer learning consumes a substantial amount of computational and storage resources.Moreover,recently proposed parameter-efficient transfer learning methods mainly focus on the reconstruction of channel features,ignoring the spatial features which are vital for modeling key entity relationships.To address these issues,we design an efficient transfer learning framework for RSCIR,which is based on spatial feature efficient reconstruction(SPER).A concise and efficient spatial adapter is introduced to enhance the extraction of spatial relationships.The spatial adapter is able to spatially reconstruct the features in the backbone with few parameters while incorporating the prior information from the channel dimension.We conduct quantitative and qualitative experiments on two different commonly used RSCIR datasets.Compared with traditional methods,our approach achieves an improvement of 3%-11% in sumR metric.Compared with methods finetuning all parameters,our proposed method only trains less than 1% of the parameters,while maintaining an overall performance of about 96%.展开更多
In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilizati...In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.展开更多
This paper investigates the adaptive fuzzy finite-time output-feedback fault-tolerant control (FTC) problemfor a class of nonlinear underactuated wheeled mobile robots (UWMRs) system with intermittent actuatorfaults. ...This paper investigates the adaptive fuzzy finite-time output-feedback fault-tolerant control (FTC) problemfor a class of nonlinear underactuated wheeled mobile robots (UWMRs) system with intermittent actuatorfaults. The UWMR system includes unknown nonlinear dynamics and immeasurable states. Fuzzy logic systems(FLSs) are utilized to work out immeasurable functions. Furthermore, with the support of the backsteppingcontrol technique and adaptive fuzzy state observer, a fuzzy adaptive finite-time output-feedback FTC scheme isdeveloped under the intermittent actuator faults. It is testifying the scheme can ensure the controlled nonlinearUWMRs is stable and the estimation errors are convergent. Finally, the comparison results and simulationvalidate the effectiveness of the proposed fuzzy adaptive finite-time FTC approach.展开更多
Blockchain has been widely used in finance,the Internet of Things(IoT),supply chains,and other scenarios as a revolutionary technology.Consensus protocol plays a vital role in blockchain,which helps all participants t...Blockchain has been widely used in finance,the Internet of Things(IoT),supply chains,and other scenarios as a revolutionary technology.Consensus protocol plays a vital role in blockchain,which helps all participants to maintain the storage state consistently.However,with the improvement of network environment complexity and system scale,blockchain development is limited by the performance,security,and scalability of the consensus protocol.To address this problem,this paper introduces the collaborative filtering mechanism commonly used in the recommendation system into the Practical Byzantine Fault Tolerance(PBFT)and proposes a Byzantine fault-tolerant(BFT)consensus protocol based on collaborative filtering recommendation(CRBFT).Specifically,an improved collaborative filtering recommendation method is designed to use the similarity between a node’s recommendation opinions and those of the recommender as a basis for determining whether to adopt the recommendation opinions.This can amplify the recommendation voice of good nodes,weaken the impact of cunningmalicious nodes on the trust value calculation,andmake the calculated resultsmore accurate.In addition,the nodes are given voting power according to their trust value,and a weight randomelection algorithm is designed and implemented to reduce the risk of attack.The experimental results show that CRBFT can effectively eliminate various malicious nodes and improve the performance of blockchain systems in complex network environments,and the feasibility of CRBFT is also proven by theoretical analysis.展开更多
In recent years,motor drive systems have garnered increasing attention due to their high efficiency and superior control performance.This is especially apparent in aerospace,marine propulsion,and electric vehicles,whe...In recent years,motor drive systems have garnered increasing attention due to their high efficiency and superior control performance.This is especially apparent in aerospace,marine propulsion,and electric vehicles,where high performance,efficiency,and reliability are crucial.The ability of the drive system to maintain long-term fault-tolerant control(FTC)operation after a failure is essential.The likelihood of inverter failures surpasses that of other components in the drive system,highlighting its critical importance.Long-term FTC operation ensures the system retains its fundamental functions until safe repairs or replacements can be made.The focus of developing a FTC strategy has shifted from basic FTC operations to enhancing the post-fault quality to accommodate the realities of prolonged operation post-failure.This paper primarily investigates FTC strategies for inverter failures in various motor drive systems over the past decade.These strategies are categorized into three types based on post-fault operational quality:rescue,remedy,and reestablishment.The paper discusses each typical control strategy and its research focus,the strengths and weaknesses of various algorithms,and recent advancements in FTC.Finally,this review summarizes effective FTC techniques for inverter failures in motor drive systems and suggests directions for future research.展开更多
The Advanced Geosynchronous Radiation Imager(AGRI)is a mission-critical instrument for the Fengyun series of satellites.AGRI acquires full-disk images every 15 min and views East Asia every 5 min through 14 spectral b...The Advanced Geosynchronous Radiation Imager(AGRI)is a mission-critical instrument for the Fengyun series of satellites.AGRI acquires full-disk images every 15 min and views East Asia every 5 min through 14 spectral bands,enabling the detection of highly variable aerosol optical depth(AOD).Quantitative retrieval of AOD has hitherto been challenging,especially over land.In this study,an AOD retrieval algorithm is proposed that combines deep learning and transfer learning.The algorithm uses core concepts from both the Dark Target(DT)and Deep Blue(DB)algorithms to select features for the machinelearning(ML)algorithm,allowing for AOD retrieval at 550 nm over both dark and bright surfaces.The algorithm consists of two steps:①A baseline deep neural network(DNN)with skip connections is developed using 10 min Advanced Himawari Imager(AHI)AODs as the target variable,and②sunphotometer AODs from 89 ground-based stations are used to fine-tune the DNN parameters.Out-of-station validation shows that the retrieved AOD attains high accuracy,characterized by a coefficient of determination(R2)of 0.70,a mean bias error(MBE)of 0.03,and a percentage of data within the expected error(EE)of 70.7%.A sensitivity study reveals that the top-of-atmosphere reflectance at 650 and 470 nm,as well as the surface reflectance at 650 nm,are the two largest sources of uncertainty impacting the retrieval.In a case study of monitoring an extreme aerosol event,the AGRI AOD is found to be able to capture the detailed temporal evolution of the event.This work demonstrates the superiority of the transfer-learning technique in satellite AOD retrievals and the applicability of the retrieved AGRI AOD in monitoring extreme pollution events.展开更多
This study examines the database search behaviors of individuals, focusing on gender differences and the impact of planning habits on information retrieval. Data were collected from a survey of 198 respondents, catego...This study examines the database search behaviors of individuals, focusing on gender differences and the impact of planning habits on information retrieval. Data were collected from a survey of 198 respondents, categorized by their discipline, schooling background, internet usage, and information retrieval preferences. Key findings indicate that females are more likely to plan their searches in advance and prefer structured methods of information retrieval, such as using library portals and leading university websites. Males, however, tend to use web search engines and self-archiving methods more frequently. This analysis provides valuable insights for educational institutions and libraries to optimize their resources and services based on user behavior patterns.展开更多
This study presents a tracking and fault-tolerant controller architecture for uncertain steer-by-wire(SbW)systems using model predictive control in the presence of actuator malfunction and the nonlinear properties of ...This study presents a tracking and fault-tolerant controller architecture for uncertain steer-by-wire(SbW)systems using model predictive control in the presence of actuator malfunction and the nonlinear properties of tire lateral stiffness coefficients.By changing the internal model,the model predictive control(MPC)technique was used to achieve optimal tracking performance under the actuator output limitation variation problem and uncertain system parameters.System parameters and state estimates were simultaneously provided by the fault detection and isolation modules to detect actuator failure using the coupling estimation approach.The estimation accuracy was further improved by considering the replacement errors as virtual noise,which was also estimated during the estimation process.Simulation and experimental results demonstrate that the proposed fault-tolerant control technique can identify motor faults and conduct fault-tolerant control based on fault identification,showing good front-wheel steering angle tracking performance under both normal and fault conditions.展开更多
The photovoltaic system is experiencing great growth in the production of electrical energy these days.It plays a vital role in the production of electrical energy in isolated towns.It is generally either stand-alone ...The photovoltaic system is experiencing great growth in the production of electrical energy these days.It plays a vital role in the production of electrical energy in isolated towns.It is generally either stand-alone or connected to a network.The energy produced by the photovoltaic generator is in continuous form;the conversion from its continuous form to the alternating form requires a converter:the inverter.In order to improve the quality of the waveform,we moved from the classic solar inverter to multilevel inverters.These multilevel inverters are equipped with power switches which are required to withstand strong fluctuations in the voltage produced by the GPV(photovoltaic generator).It is obvious that the degradation of the inverter leads to a distortion of the wave quality.This article presents the simulation of the GPV-Chopper Boost-Inverter chain in fault-tolerant cascaded H-bridges in order to overcome the difficulties of voltage constraints experienced by power switches(IGBT:insulated gate bipolar transistor).The results of simulations carried out in Matlab/Simulink show good performance of the designed inverter model.展开更多
For the multicopter with more than four rotors,the rotor fault information is unobservable,which limits the applica-tion of active fault-tolerant on multicopters.This paper applies an existing fault-tolerant control m...For the multicopter with more than four rotors,the rotor fault information is unobservable,which limits the applica-tion of active fault-tolerant on multicopters.This paper applies an existing fault-tolerant control method for quadcopter to multi-copter with more than four rotors.Without relying on rotor fault information,this method is able to stabilize the multicopter with multiple rotor failures,which is validated on the hexacopter and octocopter using the hardware-in-the-loop simulations.Addi-tionally,the hardware-in-the-loop simulations demonstrate that a more significant tilt angle in flight will inhibit the maximum tolera-ble number of rotor failures of a multicopter.The more signifi-cant aerodynamic drag moment will make it difficult for the mul-ticopter to regain altitude control after rotor failure.展开更多
Background A medical content-based image retrieval(CBIR)system is designed to retrieve images from large imaging repositories that are visually similar to a user′s query image.CBIR is widely used in evidence-based di...Background A medical content-based image retrieval(CBIR)system is designed to retrieve images from large imaging repositories that are visually similar to a user′s query image.CBIR is widely used in evidence-based diagnosis,teaching,and research.Although the retrieval accuracy has largely improved,there has been limited development toward visualizing important image features that indicate the similarity of retrieved images.Despite the prevalence of 3D volumetric data in medical imaging such as computed tomography(CT),current CBIR systems still rely on 2D cross-sectional views for the visualization of retrieved images.Such 2D visualization requires users to browse through the image stacks to confirm the similarity of the retrieved images and often involves mental reconstruction of 3D information,including the size,shape,and spatial relations of multiple structures.This process is time-consuming and reliant on users'experience.Methods In this study,we proposed an importance-aware 3D volume visualization method.The rendering parameters were automatically optimized to maximize the visibility of important structures that were detected and prioritized in the retrieval process.We then integrated the proposed visualization into a CBIR system,thereby complementing the 2D cross-sectional views for relevance feedback and further analyses.Results Our preliminary results demonstrate that 3D visualization can provide additional information using multimodal positron emission tomography and computed tomography(PETCT)images of a non-small cell lung cancer dataset.展开更多
This study introduces the Orbit Weighting Scheme(OWS),a novel approach aimed at enhancing the precision and efficiency of Vector Space information retrieval(IR)models,which have traditionally relied on weighting schem...This study introduces the Orbit Weighting Scheme(OWS),a novel approach aimed at enhancing the precision and efficiency of Vector Space information retrieval(IR)models,which have traditionally relied on weighting schemes like tf-idf and BM25.These conventional methods often struggle with accurately capturing document relevance,leading to inefficiencies in both retrieval performance and index size management.OWS proposes a dynamic weighting mechanism that evaluates the significance of terms based on their orbital position within the vector space,emphasizing term relationships and distribution patterns overlooked by existing models.Our research focuses on evaluating OWS’s impact on model accuracy using Information Retrieval metrics like Recall,Precision,InterpolatedAverage Precision(IAP),andMeanAverage Precision(MAP).Additionally,we assessOWS’s effectiveness in reducing the inverted index size,crucial for model efficiency.We compare OWS-based retrieval models against others using different schemes,including tf-idf variations and BM25Delta.Results reveal OWS’s superiority,achieving a 54%Recall and 81%MAP,and a notable 38%reduction in the inverted index size.This highlights OWS’s potential in optimizing retrieval processes and underscores the need for further research in this underrepresented area to fully leverage OWS’s capabilities in information retrieval methodologies.展开更多
Crime scene investigation(CSI)image is key evidence carrier during criminal investiga-tion,in which CSI image retrieval can assist the public police to obtain criminal clues.Moreover,with the rapid development of deep...Crime scene investigation(CSI)image is key evidence carrier during criminal investiga-tion,in which CSI image retrieval can assist the public police to obtain criminal clues.Moreover,with the rapid development of deep learning,data-driven paradigm has become the mainstreammethod of CSI image feature extraction and representation,and in this process,datasets provideeffective support for CSI retrieval performance.However,there is a lack of systematic research onCSI image retrieval methods and datasets.Therefore,we present an overview of the existing worksabout one-class and multi-class CSI image retrieval based on deep learning.According to theresearch,based on their technical functionalities and implementation methods,CSI image retrievalis roughly classified into five categories:feature representation,metric learning,generative adversar-ial networks,autoencoder networks and attention networks.Furthermore,We analyzed the remain-ing challenges and discussed future work directions in this field.展开更多
Clothing attribute recognition has become an essential technology,which enables users to automatically identify the characteristics of clothes and search for clothing images with similar attributes.However,existing me...Clothing attribute recognition has become an essential technology,which enables users to automatically identify the characteristics of clothes and search for clothing images with similar attributes.However,existing methods cannot recognize newly added attributes and may fail to capture region-level visual features.To address the aforementioned issues,a region-aware fashion contrastive language-image pre-training(RaF-CLIP)model was proposed.This model aligned cropped and segmented images with category and multiple fine-grained attribute texts,achieving the matching of fashion region and corresponding texts through contrastive learning.Clothing retrieval found suitable clothing based on the user-specified clothing categories and attributes,and to further improve the accuracy of retrieval,an attribute-guided composed network(AGCN)as an additional component on RaF-CLIP was introduced,specifically designed for composed image retrieval.This task aimed to modify the reference image based on textual expressions to retrieve the expected target.By adopting a transformer-based bidirectional attention and gating mechanism,it realized the fusion and selection of image features and attribute text features.Experimental results show that the proposed model achieves a mean precision of 0.6633 for attribute recognition tasks and a recall@10(recall@k is defined as the percentage of correct samples appearing in the top k retrieval results)of 39.18 for composed image retrieval task,satisfying user needs for freely searching for clothing through images and texts.展开更多
This paper presents an approach to improve medical image retrieval, particularly for brain tumors, by addressing the gap between low-level visual and high-level perceived contents in MRI, X-ray, and CT scans. Traditio...This paper presents an approach to improve medical image retrieval, particularly for brain tumors, by addressing the gap between low-level visual and high-level perceived contents in MRI, X-ray, and CT scans. Traditional methods based on color, shape, or texture are less effective. The proposed solution uses machine learning to handle high-dimensional image features, reducing computational complexity and mitigating issues caused by artifacts or noise. It employs a genetic algorithm for feature reduction and a hybrid residual UNet(HResUNet) model for Region-of-Interest(ROI) segmentation and classification, with enhanced image preprocessing. The study examines various loss functions, finding that a hybrid loss function yields superior results, and the GA-HResUNet model outperforms the HResUNet. Comparative analysis with state-of-the-art models shows a 4% improvement in retrieval accuracy.展开更多
With the development of information technology,the online retrieval of remote electronic data has become an important method for investigative agencies to collect evidence.In the current normative documents,the online...With the development of information technology,the online retrieval of remote electronic data has become an important method for investigative agencies to collect evidence.In the current normative documents,the online retrieval of electronic data is positioned as a new type of arbitrary investigative measure.However,study of its actual operation has found that the online retrieval of electronic data does not fully comply with the characteristics of arbitrary investigative measures.The root cause is its inaccurately defined nature due to analogy errors,an emphasis on the authenticity of electronic data at the cost of rights protection,insufficient effectiveness of normative documents to break through the boundaries of law,and superficial inconsistency found in the mechanical comparison with the nature of existing investigative measures causes.The nature of electronic data retrieved online should be defined according to different circumstances.The retrieval of electronic data disclosed on the Internet is an arbitrary investigative measure,and following procedural specifications should be sufficient.When investigators conceal their true identities and enter the cyberspace of the suspected crime through a registered account to extract dynamic electronic data for criminal activities,it is essentially a covert investigation in cyberspace,and they should follow the normative requirements for covert investigations.The retrieval of dynamic electronic data from private spaces is a technical investigative measure and should be implemented in accordance with the technical investigative procedures.Retrieval of remote“non-public electronic data involving privacy”is a mandatory investigative measure,and is essentially a search in the virtual space.Therefore,procedural specifications should be set in accordance with the standards of searching.展开更多
Video-text retrieval (VTR) is an essential task in multimodal learning, aiming to bridge the semantic gap between visual and textual data. Effective video frame sampling plays a crucial role in improving retrieval per...Video-text retrieval (VTR) is an essential task in multimodal learning, aiming to bridge the semantic gap between visual and textual data. Effective video frame sampling plays a crucial role in improving retrieval performance, as it determines the quality of the visual content representation. Traditional sampling methods, such as uniform sampling and optical flow-based techniques, often fail to capture the full semantic range of videos, leading to redundancy and inefficiencies. In this work, we propose CLIP4Video-Sampling: Global Semantics-Guided Multi-Granularity Frame Sampling for Video-Text Retrieval, a global semantics-guided multi-granularity frame sampling strategy designed to optimize both computational efficiency and retrieval accuracy. By integrating multi-scale global and local temporal sampling and leveraging the CLIP (Contrastive Language-Image Pre-training) model’s powerful feature extraction capabilities, our method significantly outperforms existing approaches in both zero-shot and fine-tuned video-text retrieval tasks on popular datasets. CLIP4Video-Sampling reduces redundancy, ensures keyframe coverage, and serves as an adaptable pre-processing module for multimodal models.展开更多
基金supported by the National Natural Science Foundation of China(Grant Nos.42030708,42375138,42030608,42105128,42075079)the Opening Foundation of Key Laboratory of Atmospheric Sounding,China Meteorological Administration(CMA),and the CMA Research Center on Meteorological Observation Engineering Technology(Grant No.U2021Z03),and the Opening Foundation of the Key Laboratory of Atmospheric Chemistry,CMA(Grant No.2022B02)。
文摘Aerosol optical depth(AOD)and fine particulate matter with a diameter of less than or equal to 2.5μm(PM_(2.5))play crucial roles in air quality,human health,and climate change.However,the complex correlation of AOD–PM_(2.5)and the limitations of existing algorithms pose a significant challenge in realizing the accurate joint retrieval of these two parameters at the same location.On this point,a multi-task learning(MTL)model,which enables the joint retrieval of PM_(2.5)concentration and AOD,is proposed and applied on the top-of-the-atmosphere reflectance data gathered by the Fengyun-4A Advanced Geosynchronous Radiation Imager(FY-4A AGRI),and compared to that of two single-task learning models—namely,Random Forest(RF)and Deep Neural Network(DNN).Specifically,MTL achieves a coefficient of determination(R^(2))of 0.88 and a root-mean-square error(RMSE)of 0.10 in AOD retrieval.In comparison to RF,the R^(2)increases by 0.04,the RMSE decreases by 0.02,and the percentage of retrieval results falling within the expected error range(Within-EE)rises by 5.55%.The R^(2)and RMSE of PM_(2.5)retrieval by MTL are 0.84 and 13.76μg m~(-3)respectively.Compared with RF,the R^(2)increases by 0.06,the RMSE decreases by 4.55μg m~(-3),and the Within-EE increases by 7.28%.Additionally,compared to DNN,MTL shows an increase of 0.01 in R^(2)and a decrease of 0.02 in RMSE in AOD retrieval,with a corresponding increase of 2.89%in Within-EE.For PM_(2.5)retrieval,MTL exhibits an increase of 0.05 in R^(2),a decrease of 1.76μg m~(-3)in RMSE,and an increase of 6.83%in Within-EE.The evaluation suggests that MTL is able to provide simultaneously improved AOD and PM_(2.5)retrievals,demonstrating a significant advantage in efficiently capturing the spatial distribution of PM_(2.5)concentration and AOD.
基金supported by the NationalNatural Science Foundation of China(No.61862041).
文摘Medical institutions frequently utilize cloud servers for storing digital medical imaging data, aiming to lower both storage expenses and computational expenses. Nevertheless, the reliability of cloud servers as third-party providers is not always guaranteed. To safeguard against the exposure and misuse of personal privacy information, and achieve secure and efficient retrieval, a secure medical image retrieval based on a multi-attention mechanism and triplet deep hashing is proposed in this paper (abbreviated as MATDH). Specifically, this method first utilizes the contrast-limited adaptive histogram equalization method applicable to color images to enhance chest X-ray images. Next, a designed multi-attention mechanism focuses on important local features during the feature extraction stage. Moreover, a triplet loss function is utilized to learn discriminative hash codes to construct a compact and efficient triplet deep hashing. Finally, upsampling is used to restore the original resolution of the images during retrieval, thereby enabling more accurate matching. To ensure the security of medical image data, a lightweight image encryption method based on frequency domain encryption is designed to encrypt the chest X-ray images. The findings of the experiment indicate that, in comparison to various advanced image retrieval techniques, the suggested approach improves the precision of feature extraction and retrieval using the COVIDx dataset. Additionally, it offers enhanced protection for the confidentiality of medical images stored in cloud settings and demonstrates strong practicality.
基金supported by the National Key R&D Program of China(No.2022ZD0118402)。
文摘Remote sensing cross-modal image-text retrieval(RSCIR)can flexibly and subjectively retrieve remote sensing images utilizing query text,which has received more researchers’attention recently.However,with the increasing volume of visual-language pre-training model parameters,direct transfer learning consumes a substantial amount of computational and storage resources.Moreover,recently proposed parameter-efficient transfer learning methods mainly focus on the reconstruction of channel features,ignoring the spatial features which are vital for modeling key entity relationships.To address these issues,we design an efficient transfer learning framework for RSCIR,which is based on spatial feature efficient reconstruction(SPER).A concise and efficient spatial adapter is introduced to enhance the extraction of spatial relationships.The spatial adapter is able to spatially reconstruct the features in the backbone with few parameters while incorporating the prior information from the channel dimension.We conduct quantitative and qualitative experiments on two different commonly used RSCIR datasets.Compared with traditional methods,our approach achieves an improvement of 3%-11% in sumR metric.Compared with methods finetuning all parameters,our proposed method only trains less than 1% of the parameters,while maintaining an overall performance of about 96%.
文摘In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.
基金the National Natural Science Foundation of China under Grant U22A2043.
文摘This paper investigates the adaptive fuzzy finite-time output-feedback fault-tolerant control (FTC) problemfor a class of nonlinear underactuated wheeled mobile robots (UWMRs) system with intermittent actuatorfaults. The UWMR system includes unknown nonlinear dynamics and immeasurable states. Fuzzy logic systems(FLSs) are utilized to work out immeasurable functions. Furthermore, with the support of the backsteppingcontrol technique and adaptive fuzzy state observer, a fuzzy adaptive finite-time output-feedback FTC scheme isdeveloped under the intermittent actuator faults. It is testifying the scheme can ensure the controlled nonlinearUWMRs is stable and the estimation errors are convergent. Finally, the comparison results and simulationvalidate the effectiveness of the proposed fuzzy adaptive finite-time FTC approach.
基金supported by the National Natural Science Foundation of China(Grant No.62102449)awarded to W.J.Wang.
文摘Blockchain has been widely used in finance,the Internet of Things(IoT),supply chains,and other scenarios as a revolutionary technology.Consensus protocol plays a vital role in blockchain,which helps all participants to maintain the storage state consistently.However,with the improvement of network environment complexity and system scale,blockchain development is limited by the performance,security,and scalability of the consensus protocol.To address this problem,this paper introduces the collaborative filtering mechanism commonly used in the recommendation system into the Practical Byzantine Fault Tolerance(PBFT)and proposes a Byzantine fault-tolerant(BFT)consensus protocol based on collaborative filtering recommendation(CRBFT).Specifically,an improved collaborative filtering recommendation method is designed to use the similarity between a node’s recommendation opinions and those of the recommender as a basis for determining whether to adopt the recommendation opinions.This can amplify the recommendation voice of good nodes,weaken the impact of cunningmalicious nodes on the trust value calculation,andmake the calculated resultsmore accurate.In addition,the nodes are given voting power according to their trust value,and a weight randomelection algorithm is designed and implemented to reduce the risk of attack.The experimental results show that CRBFT can effectively eliminate various malicious nodes and improve the performance of blockchain systems in complex network environments,and the feasibility of CRBFT is also proven by theoretical analysis.
基金supported in part by the National Natural Science Foundation of China under Grants 52025073 and 52107047in part by China Scholarship Council。
文摘In recent years,motor drive systems have garnered increasing attention due to their high efficiency and superior control performance.This is especially apparent in aerospace,marine propulsion,and electric vehicles,where high performance,efficiency,and reliability are crucial.The ability of the drive system to maintain long-term fault-tolerant control(FTC)operation after a failure is essential.The likelihood of inverter failures surpasses that of other components in the drive system,highlighting its critical importance.Long-term FTC operation ensures the system retains its fundamental functions until safe repairs or replacements can be made.The focus of developing a FTC strategy has shifted from basic FTC operations to enhancing the post-fault quality to accommodate the realities of prolonged operation post-failure.This paper primarily investigates FTC strategies for inverter failures in various motor drive systems over the past decade.These strategies are categorized into three types based on post-fault operational quality:rescue,remedy,and reestablishment.The paper discusses each typical control strategy and its research focus,the strengths and weaknesses of various algorithms,and recent advancements in FTC.Finally,this review summarizes effective FTC techniques for inverter failures in motor drive systems and suggests directions for future research.
基金supported by the National Natural Science of Foundation of China(41825011,42030608,42105128,and 42075079)the Opening Foundation of Key Laboratory of Atmospheric Sounding,the CMA and the CMA Research Center on Meteorological Observation Engineering Technology(U2021Z03).
文摘The Advanced Geosynchronous Radiation Imager(AGRI)is a mission-critical instrument for the Fengyun series of satellites.AGRI acquires full-disk images every 15 min and views East Asia every 5 min through 14 spectral bands,enabling the detection of highly variable aerosol optical depth(AOD).Quantitative retrieval of AOD has hitherto been challenging,especially over land.In this study,an AOD retrieval algorithm is proposed that combines deep learning and transfer learning.The algorithm uses core concepts from both the Dark Target(DT)and Deep Blue(DB)algorithms to select features for the machinelearning(ML)algorithm,allowing for AOD retrieval at 550 nm over both dark and bright surfaces.The algorithm consists of two steps:①A baseline deep neural network(DNN)with skip connections is developed using 10 min Advanced Himawari Imager(AHI)AODs as the target variable,and②sunphotometer AODs from 89 ground-based stations are used to fine-tune the DNN parameters.Out-of-station validation shows that the retrieved AOD attains high accuracy,characterized by a coefficient of determination(R2)of 0.70,a mean bias error(MBE)of 0.03,and a percentage of data within the expected error(EE)of 70.7%.A sensitivity study reveals that the top-of-atmosphere reflectance at 650 and 470 nm,as well as the surface reflectance at 650 nm,are the two largest sources of uncertainty impacting the retrieval.In a case study of monitoring an extreme aerosol event,the AGRI AOD is found to be able to capture the detailed temporal evolution of the event.This work demonstrates the superiority of the transfer-learning technique in satellite AOD retrievals and the applicability of the retrieved AGRI AOD in monitoring extreme pollution events.
文摘This study examines the database search behaviors of individuals, focusing on gender differences and the impact of planning habits on information retrieval. Data were collected from a survey of 198 respondents, categorized by their discipline, schooling background, internet usage, and information retrieval preferences. Key findings indicate that females are more likely to plan their searches in advance and prefer structured methods of information retrieval, such as using library portals and leading university websites. Males, however, tend to use web search engines and self-archiving methods more frequently. This analysis provides valuable insights for educational institutions and libraries to optimize their resources and services based on user behavior patterns.
基金Supported by Youth Fund of Jiangsu Provincial Natural Science Foundation of China(Grant No.BK20200423)National Natural Science Foundation of China(Grant Nos.52102435,52072175)+2 种基金Jiangsu Outstanding Youth Fund Project of China(Grant No.BK20220078)Jiangsu Provincial Key R&D Plan of China(Grant No.BE2022053)Postgraduate Research&Practice Innovation Program of NUAA of China(Grant No.xcxjh20220210)。
文摘This study presents a tracking and fault-tolerant controller architecture for uncertain steer-by-wire(SbW)systems using model predictive control in the presence of actuator malfunction and the nonlinear properties of tire lateral stiffness coefficients.By changing the internal model,the model predictive control(MPC)technique was used to achieve optimal tracking performance under the actuator output limitation variation problem and uncertain system parameters.System parameters and state estimates were simultaneously provided by the fault detection and isolation modules to detect actuator failure using the coupling estimation approach.The estimation accuracy was further improved by considering the replacement errors as virtual noise,which was also estimated during the estimation process.Simulation and experimental results demonstrate that the proposed fault-tolerant control technique can identify motor faults and conduct fault-tolerant control based on fault identification,showing good front-wheel steering angle tracking performance under both normal and fault conditions.
文摘The photovoltaic system is experiencing great growth in the production of electrical energy these days.It plays a vital role in the production of electrical energy in isolated towns.It is generally either stand-alone or connected to a network.The energy produced by the photovoltaic generator is in continuous form;the conversion from its continuous form to the alternating form requires a converter:the inverter.In order to improve the quality of the waveform,we moved from the classic solar inverter to multilevel inverters.These multilevel inverters are equipped with power switches which are required to withstand strong fluctuations in the voltage produced by the GPV(photovoltaic generator).It is obvious that the degradation of the inverter leads to a distortion of the wave quality.This article presents the simulation of the GPV-Chopper Boost-Inverter chain in fault-tolerant cascaded H-bridges in order to overcome the difficulties of voltage constraints experienced by power switches(IGBT:insulated gate bipolar transistor).The results of simulations carried out in Matlab/Simulink show good performance of the designed inverter model.
基金supported by the National Natural Science Foundation of China(61973015).
文摘For the multicopter with more than four rotors,the rotor fault information is unobservable,which limits the applica-tion of active fault-tolerant on multicopters.This paper applies an existing fault-tolerant control method for quadcopter to multi-copter with more than four rotors.Without relying on rotor fault information,this method is able to stabilize the multicopter with multiple rotor failures,which is validated on the hexacopter and octocopter using the hardware-in-the-loop simulations.Addi-tionally,the hardware-in-the-loop simulations demonstrate that a more significant tilt angle in flight will inhibit the maximum tolera-ble number of rotor failures of a multicopter.The more signifi-cant aerodynamic drag moment will make it difficult for the mul-ticopter to regain altitude control after rotor failure.
文摘Background A medical content-based image retrieval(CBIR)system is designed to retrieve images from large imaging repositories that are visually similar to a user′s query image.CBIR is widely used in evidence-based diagnosis,teaching,and research.Although the retrieval accuracy has largely improved,there has been limited development toward visualizing important image features that indicate the similarity of retrieved images.Despite the prevalence of 3D volumetric data in medical imaging such as computed tomography(CT),current CBIR systems still rely on 2D cross-sectional views for the visualization of retrieved images.Such 2D visualization requires users to browse through the image stacks to confirm the similarity of the retrieved images and often involves mental reconstruction of 3D information,including the size,shape,and spatial relations of multiple structures.This process is time-consuming and reliant on users'experience.Methods In this study,we proposed an importance-aware 3D volume visualization method.The rendering parameters were automatically optimized to maximize the visibility of important structures that were detected and prioritized in the retrieval process.We then integrated the proposed visualization into a CBIR system,thereby complementing the 2D cross-sectional views for relevance feedback and further analyses.Results Our preliminary results demonstrate that 3D visualization can provide additional information using multimodal positron emission tomography and computed tomography(PETCT)images of a non-small cell lung cancer dataset.
文摘This study introduces the Orbit Weighting Scheme(OWS),a novel approach aimed at enhancing the precision and efficiency of Vector Space information retrieval(IR)models,which have traditionally relied on weighting schemes like tf-idf and BM25.These conventional methods often struggle with accurately capturing document relevance,leading to inefficiencies in both retrieval performance and index size management.OWS proposes a dynamic weighting mechanism that evaluates the significance of terms based on their orbital position within the vector space,emphasizing term relationships and distribution patterns overlooked by existing models.Our research focuses on evaluating OWS’s impact on model accuracy using Information Retrieval metrics like Recall,Precision,InterpolatedAverage Precision(IAP),andMeanAverage Precision(MAP).Additionally,we assessOWS’s effectiveness in reducing the inverted index size,crucial for model efficiency.We compare OWS-based retrieval models against others using different schemes,including tf-idf variations and BM25Delta.Results reveal OWS’s superiority,achieving a 54%Recall and 81%MAP,and a notable 38%reduction in the inverted index size.This highlights OWS’s potential in optimizing retrieval processes and underscores the need for further research in this underrepresented area to fully leverage OWS’s capabilities in information retrieval methodologies.
文摘Crime scene investigation(CSI)image is key evidence carrier during criminal investiga-tion,in which CSI image retrieval can assist the public police to obtain criminal clues.Moreover,with the rapid development of deep learning,data-driven paradigm has become the mainstreammethod of CSI image feature extraction and representation,and in this process,datasets provideeffective support for CSI retrieval performance.However,there is a lack of systematic research onCSI image retrieval methods and datasets.Therefore,we present an overview of the existing worksabout one-class and multi-class CSI image retrieval based on deep learning.According to theresearch,based on their technical functionalities and implementation methods,CSI image retrievalis roughly classified into five categories:feature representation,metric learning,generative adversar-ial networks,autoencoder networks and attention networks.Furthermore,We analyzed the remain-ing challenges and discussed future work directions in this field.
基金National Natural Science Foundation of China(No.61971121)。
文摘Clothing attribute recognition has become an essential technology,which enables users to automatically identify the characteristics of clothes and search for clothing images with similar attributes.However,existing methods cannot recognize newly added attributes and may fail to capture region-level visual features.To address the aforementioned issues,a region-aware fashion contrastive language-image pre-training(RaF-CLIP)model was proposed.This model aligned cropped and segmented images with category and multiple fine-grained attribute texts,achieving the matching of fashion region and corresponding texts through contrastive learning.Clothing retrieval found suitable clothing based on the user-specified clothing categories and attributes,and to further improve the accuracy of retrieval,an attribute-guided composed network(AGCN)as an additional component on RaF-CLIP was introduced,specifically designed for composed image retrieval.This task aimed to modify the reference image based on textual expressions to retrieve the expected target.By adopting a transformer-based bidirectional attention and gating mechanism,it realized the fusion and selection of image features and attribute text features.Experimental results show that the proposed model achieves a mean precision of 0.6633 for attribute recognition tasks and a recall@10(recall@k is defined as the percentage of correct samples appearing in the top k retrieval results)of 39.18 for composed image retrieval task,satisfying user needs for freely searching for clothing through images and texts.
文摘This paper presents an approach to improve medical image retrieval, particularly for brain tumors, by addressing the gap between low-level visual and high-level perceived contents in MRI, X-ray, and CT scans. Traditional methods based on color, shape, or texture are less effective. The proposed solution uses machine learning to handle high-dimensional image features, reducing computational complexity and mitigating issues caused by artifacts or noise. It employs a genetic algorithm for feature reduction and a hybrid residual UNet(HResUNet) model for Region-of-Interest(ROI) segmentation and classification, with enhanced image preprocessing. The study examines various loss functions, finding that a hybrid loss function yields superior results, and the GA-HResUNet model outperforms the HResUNet. Comparative analysis with state-of-the-art models shows a 4% improvement in retrieval accuracy.
基金the phased research result of the Supreme People’s Procuratorate’s procuratorial theory research program“Research on the Governance Problems of the Crime of Aiding Information Network Criminal Activities”(Project Approval Number GJ2023D28)。
文摘With the development of information technology,the online retrieval of remote electronic data has become an important method for investigative agencies to collect evidence.In the current normative documents,the online retrieval of electronic data is positioned as a new type of arbitrary investigative measure.However,study of its actual operation has found that the online retrieval of electronic data does not fully comply with the characteristics of arbitrary investigative measures.The root cause is its inaccurately defined nature due to analogy errors,an emphasis on the authenticity of electronic data at the cost of rights protection,insufficient effectiveness of normative documents to break through the boundaries of law,and superficial inconsistency found in the mechanical comparison with the nature of existing investigative measures causes.The nature of electronic data retrieved online should be defined according to different circumstances.The retrieval of electronic data disclosed on the Internet is an arbitrary investigative measure,and following procedural specifications should be sufficient.When investigators conceal their true identities and enter the cyberspace of the suspected crime through a registered account to extract dynamic electronic data for criminal activities,it is essentially a covert investigation in cyberspace,and they should follow the normative requirements for covert investigations.The retrieval of dynamic electronic data from private spaces is a technical investigative measure and should be implemented in accordance with the technical investigative procedures.Retrieval of remote“non-public electronic data involving privacy”is a mandatory investigative measure,and is essentially a search in the virtual space.Therefore,procedural specifications should be set in accordance with the standards of searching.
文摘Video-text retrieval (VTR) is an essential task in multimodal learning, aiming to bridge the semantic gap between visual and textual data. Effective video frame sampling plays a crucial role in improving retrieval performance, as it determines the quality of the visual content representation. Traditional sampling methods, such as uniform sampling and optical flow-based techniques, often fail to capture the full semantic range of videos, leading to redundancy and inefficiencies. In this work, we propose CLIP4Video-Sampling: Global Semantics-Guided Multi-Granularity Frame Sampling for Video-Text Retrieval, a global semantics-guided multi-granularity frame sampling strategy designed to optimize both computational efficiency and retrieval accuracy. By integrating multi-scale global and local temporal sampling and leveraging the CLIP (Contrastive Language-Image Pre-training) model’s powerful feature extraction capabilities, our method significantly outperforms existing approaches in both zero-shot and fine-tuned video-text retrieval tasks on popular datasets. CLIP4Video-Sampling reduces redundancy, ensures keyframe coverage, and serves as an adaptable pre-processing module for multimodal models.
