In view of the weak ability of the convolutional neural networks to explicitly learn spatial invariance and the probabilistic loss of discriminative features caused by occlusion and background interference in pedestri...In view of the weak ability of the convolutional neural networks to explicitly learn spatial invariance and the probabilistic loss of discriminative features caused by occlusion and background interference in pedestrian re-identification tasks,a person re-identification method combining spatial feature learning and multi-granularity feature fusion was proposed.First,an attention spatial transformation network(A-STN)is proposed to learn spatial features and solve the problem of misalignment of pedestrian spatial features.Then the network was divided into a global branch,a local coarse-grained fusion branch,and a local fine-grained fusion branch to extract pedestrian global features,coarse-grained fusion features,and fine-grained fusion features,respectively.Among them,the global branch enriches the global features by fusing different pooling features.The local coarse-grained fusion branch uses an overlay pooling to enhance each local feature while learning the correlation relationship between multi-granularity features.The local fine-grained fusion branch uses a differential pooling to obtain the differential features that were fused with global features to learn the relationship between pedestrian local features and pedestrian global features.Finally,the proposed method was compared on three public datasets:Market1501,DukeMTMC-ReID and CUHK03.The experimental results were better than those of the comparative methods,which verifies the effectiveness of the proposed method.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
Rapid development of deepfake technology led to the spread of forged audios and videos across network platforms,presenting risks for numerous countries,societies,and individuals,and posing a serious threat to cyberspa...Rapid development of deepfake technology led to the spread of forged audios and videos across network platforms,presenting risks for numerous countries,societies,and individuals,and posing a serious threat to cyberspace security.To address the problem of insufficient extraction of spatial features and the fact that temporal features are not considered in the deepfake video detection,we propose a detection method based on improved CapsNet and temporal–spatial features(iCapsNet–TSF).First,the dynamic routing algorithm of CapsNet is improved using weight initialization and updating.Then,the optical flow algorithm is used to extract interframe temporal features of the videos to form a dataset of temporal–spatial features.Finally,the iCapsNet model is employed to fully learn the temporal–spatial features of facial videos,and the results are fused.Experimental results show that the detection accuracy of iCapsNet–TSF reaches 94.07%,98.83%,and 98.50%on the Celeb-DF,FaceSwap,and Deepfakes datasets,respectively,displaying a better performance than most existing mainstream algorithms.The iCapsNet–TSF method combines the capsule network and the optical flow algorithm,providing a novel strategy for the deepfake detection,which is of great significance to the prevention of deepfake attacks and the preservation of cyberspace security.展开更多
Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vi...Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vision, attracting the attention of many researchers. However, most HSI SR methods focus on the tradeoff between spatial resolution and spectral information, and cannot guarantee the efficient extraction of image information. In this paper, a multidimensional features network(MFNet) for HSI SR is proposed, which simultaneously learns and fuses the spatial,spectral, and frequency multidimensional features of HSI. Spatial features contain rich local details,spectral features contain the information and correlation between spectral bands, and frequency feature can reflect the global information of the image and can be used to obtain the global context of HSI. The fusion of the three features can better guide image super-resolution, to obtain higher-quality high-resolution hyperspectral images. In MFNet, we use the frequency feature extraction module(FFEM) to extract the frequency feature. On this basis, a multidimensional features extraction module(MFEM) is designed to learn and fuse multidimensional features. In addition, experimental results on two public datasets demonstrate that MFNet achieves state-of-the-art performance.展开更多
In the daily application of an iris-recognition-at-a-distance(IAAD)system,many ocular images of low quality are acquired.As the iris part of these images is often not qualified for the recognition requirements,the mor...In the daily application of an iris-recognition-at-a-distance(IAAD)system,many ocular images of low quality are acquired.As the iris part of these images is often not qualified for the recognition requirements,the more accessible periocular regions are a good complement for recognition.To further boost the performance of IAAD systems,a novel end-to-end framework for multi-modal ocular recognition is proposed.The proposed framework mainly consists of iris/periocular feature extraction and matching,unsupervised iris quality assessment,and a score-level adaptive weighted fusion strategy.First,ocular feature reconstruction(OFR)is proposed to sparsely reconstruct each probe image by high-quality gallery images based on proper feature maps.Next,a brand new unsupervised iris quality assessment method based on random multiscale embedding robustness is proposed.Different from the existing iris quality assess-ment methods,the quality of an iris image is measured by its robustness in the embedding space.At last,the fusion strategy exploits the iris quality score as the fusion weight to coalesce the complementary information from the iris and periocular regions.Extensive experi-mental results on ocular datasets prove that the proposed method is obviously better than unimodal biometrics,and the fusion strategy can significantly improve therecognition performance.展开更多
To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is ba...To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method（Gabor–NWSF and SRC）, abbreviated GNWSF–SRC. The proposed（GNWSF–SRC） method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.展开更多
The analysis of hidden spatial features is crucial for the improvement of hedonic regression models for analyzing the structure of land and housing prices. If critical variables representing the influence of spatial f...The analysis of hidden spatial features is crucial for the improvement of hedonic regression models for analyzing the structure of land and housing prices. If critical variables representing the influence of spatial features are omitted in the models, the residuals and the coefficients estimated usually exhibit some kind of spatial pattern. Hence, exploration of the relationship between the spatial patterns and the spatial features essentially leads to the discovery of omitted variables. The analyses in this paper were based on two exploratory approaches: one on the residual of a global regression model and the other on the geographically weighted regression (GWR) technique. In the GWR model, the regression coefficients are al- lowed to differ by location so more spatial patterns can be revealed. Comparison of the two approaches shows that they play supplementary roles for the detection of lot-associated variables and area-associated variables.展开更多
41 a （1961 - 2001） seasonal Z index series of 25 representative weather stations are investigated by virtue of EOF, FFT, continuous wavelet transformation （CWT） and orthogonai wavelet transformation （OWT）. It sh...41 a （1961 - 2001） seasonal Z index series of 25 representative weather stations are investigated by virtue of EOF, FFT, continuous wavelet transformation （CWT） and orthogonai wavelet transformation （OWT）. It shows that： （1） Fujian drought/flood （DF） has a significant 2 - 3a cycle for the periods 1965 - 1975 and 1990＇s; （2） the pattern, which represents the opposite DF trend between the southern and northem parts, has la and 3 - 4a cycles since the middle of 1980＇s; （3） EOF3, which denotes the reverse change between the middle-west region and other areas, has significant 1 - 2a cycle for the period from 1985 to 1998 and 9 - 13a cycle since 1980s; （4） there is an obvious drought trend for the last 40a （especially in the 1990＇s）, which is more outstanding in the south （east） than in the north （west）; （5） the 1960＇s and 1980＇s are in relatively wet phases and the 1970＇s and 1990＇s are in drought spells.展开更多
Based on temperature data of meteorological stations from 1971 to 2008 in Tibet,the temporal and spatial variation of maximum andminimum temperature in Tibet was analyzed.The results showed that both maximum temperatu...Based on temperature data of meteorological stations from 1971 to 2008 in Tibet,the temporal and spatial variation of maximum andminimum temperature in Tibet was analyzed.The results showed that both maximum temperature andminimum temperature increased distinctly,the warming amplitude of winter was the highest among the four seasons,and next came spring.The increment ofminimum temperature was visibly over that of maximum temperature,particularlyminimum temperature in winter with significant increment.For spatial variation,maximum temperature in most stations increased except particular stations,while theminimum temperature in all stations rose.In addition,the space variation law ofminimum temperature,being more obvious thanminimum temperature,increased from southeast to northwest with different spatial changes in various seasons.From decadal variation,both maximum andminimum temperature appeared increase from 1970s to the first eight years in the 21st century,and the rise ofminimum temperature was significant greater than maximum temperature.The increase of maximum andminimum temperature was the highest from 2001 to 2008,whereas the lowest in 1970s.展开更多
The results of an analysis of the temporal and spatial distribution of typhoon precipitation influencing Fujian from 1960 to 2005 show that typhoon precipitation in Fujian province occurs from May to November, with th...The results of an analysis of the temporal and spatial distribution of typhoon precipitation influencing Fujian from 1960 to 2005 show that typhoon precipitation in Fujian province occurs from May to November, with the most in August. There has been a decreasing trend since 1960. Typhoon precipitation gradually decreases from the coastal region to the northwestern mainland of Fujian and the maximum typhoon precipitation occurs in the northeast and the south of Fujian. Typhoon torrential rain is one of the extreme rainfall events in Fujian. High frequencies of typhoon torrential rain occur in the coastal and southwest regions of the province. With the impact of Fujian＇s terrain, typhoon precipitation occurs more easily to the east of the mountains than to the west. Atmospheric circulation at 500 hPa over Asia and sea surface temperature anomalies of the equatorial eastern Pacific are analyzed, with the finding that they are closely connected with the anomaly of typhoon precipitation influencing Fujian, possibly mainly by modulating the northbound track of typhoons via changing the atmosphere circulation to lead to the anomaly of typhoon precipitation over the province展开更多
Multi-Object Tracking(MOT)represents a fundamental but computationally demanding task in computer vision,with particular challenges arising in occluded and densely populated environments.While contemporary tracking sy...Multi-Object Tracking(MOT)represents a fundamental but computationally demanding task in computer vision,with particular challenges arising in occluded and densely populated environments.While contemporary tracking systems have demonstrated considerable progress,persistent limitations—notably frequent occlusion-induced identity switches and tracking inaccuracies—continue to impede reliable real-world deployment.This work introduces an advanced tracking framework that enhances association robustness through a two-stage matching paradigm combining spatial and appearance features.Proposed framework employs:(1)a Height Modulated and Scale Adaptive Spatial Intersection-over-Union(HMSIoU)metric for improved spatial correspondence estimation across variable object scales and partial occlusions;(2)a feature extraction module generating discriminative appearance descriptors for identity maintenance;and(3)a recovery association mechanism for refining matches between unassociated tracks and detections.Comprehensive evaluation on standard MOT17 and MOT20 benchmarks demonstrates significant improvements in tracking consistency,with state-of-the-art performance across key metrics including HOTA(64),MOTA(80.7),IDF1(79.8),and IDs(1379).These results substantiate the efficacy of our Cue-Tracker framework in complex real-world scenarios characterized by occlusions and crowd interactions.展开更多
A large database is desired for machine learning(ML) technology to make accurate predictions of materials physicochemical properties based on their molecular structure.When a large database is not available,the develo...A large database is desired for machine learning(ML) technology to make accurate predictions of materials physicochemical properties based on their molecular structure.When a large database is not available,the development of proper featurization method based on physicochemical nature of target proprieties can improve the predictive power of ML models with a smaller database.In this work,we show that two new featurization methods,volume occupation spatial matrix and heat contribution spatial matrix,can improve the accuracy in predicting energetic materials' crystal density(ρ_(crystal)) and solid phase enthalpy of formation(H_(f,solid)) using a database containing 451 energetic molecules.Their mean absolute errors are reduced from 0.048 g/cm~3 and 24.67 kcal/mol to 0.035 g/cm~3 and 9.66 kcal/mol,respectively.By leave-one-out-cross-validation,the newly developed ML models can be used to determine the performance of most kinds of energetic materials except cubanes.Our ML models are applied to predict ρ_(crystal) and H_(f,solid) of CHON-based molecules of the 150 million sized PubChem database,and screened out 56 candidates with competitive detonation performance and reasonable chemical structures.With further improvement in future,spatial matrices have the potential of becoming multifunctional ML simulation tools that could provide even better predictions in wider fields of materials science.展开更多
How to construct an appropriate spatial consistent measurement is the key to improving image retrieval performance. To address this problem, this paper introduces a novel image retrieval mechanism based on the family ...How to construct an appropriate spatial consistent measurement is the key to improving image retrieval performance. To address this problem, this paper introduces a novel image retrieval mechanism based on the family filtration in object region. First, we supply an object region by selecting a rectangle in a query image such that system returns a ranked list of images that contain the same object, retrieved from the corpus based on 100 images, as a result of the first rank. To further improve retrieval performance, we add an efficient spatial consistency stage, which is named family-based spatial consistency filtration, to re-rank the results returned by the first rank. We elaborate the performance of the retrieval system by some experiments on the dataset selected from the key frames of ＂TREC Video Retrieval Evaluation 2005 （TRECVID2005）＂. The results of experiments show that the retrieval mechanism proposed by us has vast major effect on the retrieval quality. The paper also verifies the stability of the retrieval mechanism by increasing the number of images from 100 to 2000 and realizes generalized retrieval with the object outside the dataset.展开更多
GML is becoming the de facto standard for electronic data exchange among the applications of Web and distributed geographic information systems. However, the conventional query languages （e. g. SQL and its extended v...GML is becoming the de facto standard for electronic data exchange among the applications of Web and distributed geographic information systems. However, the conventional query languages （e. g. SQL and its extended versions） are not suitable for direct querying and updating of GML documents. Even the effective approaches working well with XML could not guarantee good results when applied to GML documents. Although XQuery is a powerful standard query language for XML, it is not proposed for querying spatial features, which constitute the most important components in GML documents. We propose GQL, a query language specification to support spatial queries over GML documents by extending XQuery. The data model, algebra, and formal semantics as well as various spatial Junctions and operations of GQL are presented in detail.展开更多
Sanxingdui cultural relics are the precious cultural heritage of humanity with high values of history,science,culture,art and research.However,mainstream analytical methods are contacting and detrimental,which is unfa...Sanxingdui cultural relics are the precious cultural heritage of humanity with high values of history,science,culture,art and research.However,mainstream analytical methods are contacting and detrimental,which is unfavorable to the protection of cultural relics.This paper improves the accuracy of the extraction,location,and analysis of artifacts using hyperspectral methods.To improve the accuracy of cultural relic mining,positioning,and analysis,the segmentation algorithm of Sanxingdui cultural relics based on the spatial spectrum integrated network is proposed with the support of hyperspectral techniques.Firstly,region stitching algorithm based on the relative position of hyper spectrally collected data is proposed to improve stitching efficiency.Secondly,given the prominence of traditional HRNet(High-Resolution Net)models in high-resolution data processing,the spatial attention mechanism is put forward to obtain spatial dimension information.Thirdly,in view of the prominence of 3D networks in spectral information acquisition,the pyramid 3D residual network model is proposed to obtain internal spectral dimensional information.Fourthly,four kinds of fusion methods at the level of data and decision are presented to achieve cultural relic labeling.As shown by the experiment results,the proposed network adopts an integrated method of data-level and decision-level,which achieves the optimal average accuracy of identification 0.84,realizes shallow coverage of cultural relics labeling,and effectively supports the mining and protection of cultural relics.展开更多
To find disaster relevant social media messages,current approaches utilize natural language processing methods or machine learning algorithms relying on text only,which have not been perfected due to the variability a...To find disaster relevant social media messages,current approaches utilize natural language processing methods or machine learning algorithms relying on text only,which have not been perfected due to the variability and uncertainty in the language used on social media and ignoring the geographic context of the messages when posted.Meanwhile,a disaster relevant social media message is highly sensitive to its posting location and time.However,limited studies exist to explore what spatial features and the extent of how temporal,and especially spatial features can aid text classification.This paper proposes a geographic context-aware text mining method to incorporate spatial and temporal information derived from social media and authoritative datasets,along with the text information,for classifying disaster relevant social media posts.This work designed and demonstrated how diverse types of spatial and temporal features can be derived from spatial data,and then used to enhance text mining.The deep learning-based method and commonly used machine learning algorithms,assessed the accuracy of the enhanced text-mining method.The performance results of different classification models generated by various combinations of textual,spatial,and temporal features indicate that additional spatial and temporal features help improve the overall accuracy of the classification.展开更多
A method is proposed for the prospecting prediction of subsurface mineral deposits based on soil geochemistry data and a deep convolutional neural network model.This method uses three techniques(window offset,scaling,...A method is proposed for the prospecting prediction of subsurface mineral deposits based on soil geochemistry data and a deep convolutional neural network model.This method uses three techniques(window offset,scaling,and rotation)to enhance the number of training data for the model.A window area is used to extract the spatial distribution characteristics of soil geochemistry and measure their correspondence with the occurrence of known subsurface deposits.Prospecting prediction is achieved by matching the characteristics of the window area of an unknown area with the relationships established in the known area.This method can efficiently predict mineral prospective areas where there are few ore deposits used for generating the training dataset,meaning that the deep-learning method can be effectively used for deposit prospecting prediction.Using soil active geochemical measurement data,this method was applied in the Daqiao area,Gansu Province,for which seven favorable gold prospecting target areas were predicted.The Daqiao orogenic gold deposit of latest Jurassic and Early Jurassic age in the southern domain has more than 105 t of gold resources at an average grade of 3-4 g/t.In 2020,the project team drilled and verified the K prediction area,and found 66 m gold mineralized bodies.The new method should be applicable to prospecting prediction using conventional geochemical data in other areas.展开更多
Microphone array-based sound source localization(SSL)is widely used in a variety of occasions such as video conferencing,robotic hearing,speech enhancement,speech recognition and so on.The traditional SSL methods cann...Microphone array-based sound source localization(SSL)is widely used in a variety of occasions such as video conferencing,robotic hearing,speech enhancement,speech recognition and so on.The traditional SSL methods cannot achieve satisfactory performance in adverse noisy and reverberant environments.In order to improve localization performance,a novel SSL algorithm using convolutional residual network(CRN)is proposed in this paper.The spatial features including time difference of arrivals(TDOAs)between microphone pairs and steered response power-phase transform(SRPPHAT)spatial spectrum are extracted in each Gammatone sub-band.The spatial features of different sub-bands with a frame are combine into a feature matrix as the input of CRN.The proposed algorithm employ CRN to fuse the spatial features.Since the CRN introduces the residual structure on the basis of the convolutional network,it reduce the difficulty of training procedure and accelerate the convergence of the model.A CRN model is learned from the training data in various reverberation and noise environments to establish the mapping regularity between the input feature and the sound azimuth.Through simulation verification,compared with the methods using traditional deep neural network,the proposed algorithm can achieve a better localization performance in SSL task,and provide better generalization capacity to untrained noise and reverberation.展开更多
The diurnal temperature range(DTR) has decreased dramatically in recent decades, but it is not yet obvious whether the extreme values of DTR have also reduced. Based on the daily maximum and minimum temperature data o...The diurnal temperature range(DTR) has decreased dramatically in recent decades, but it is not yet obvious whether the extreme values of DTR have also reduced. Based on the daily maximum and minimum temperature data of 653 stations in China, a set of monthly indices of warm extremes, cold extremes, and DTR extremes in summer(June, July, August) and winter(December, January, February) were studied for spatial and temporal features during the period 1971–2013. Results show that the incidence of warm extremes has been increasing in most parts of China, while the opposite trend was found in the cold extremes for summer and winter months. Both increasing and decreasing trends of monthly DTR extremes were identified in China for both seasons. For high DTR extremes, decreasing trends were identified in northern China for both seasons, but increasing trends were found only in southern China in summer, while in winter, they were found in central China. Monthly low DTR extreme indices demonstrated consistent positive trends in summer and winter, while significant increases(P < 0.05) were identified for only a few stations.展开更多
In medical research and clinical diagnosis, automated or computer-assisted classification and retrieval methods are highly desirable to offset the high cost of manual classification and manipulation by medical experts...In medical research and clinical diagnosis, automated or computer-assisted classification and retrieval methods are highly desirable to offset the high cost of manual classification and manipulation by medical experts. To facilitate the decision-making in the health-care and the related areas, in this paper, a two-step content-based medical image retrieval algorithm is proposed. Firstly, in the preprocessing step, the image segmentation is performed to distinguish image objects, and on the basis of the ...展开更多
基金the Foshan Science and technology Innovation Team Project(No.FS0AA-KJ919-4402-0060)the National Natural Science Foundation of China(No.62263018)。
文摘In view of the weak ability of the convolutional neural networks to explicitly learn spatial invariance and the probabilistic loss of discriminative features caused by occlusion and background interference in pedestrian re-identification tasks,a person re-identification method combining spatial feature learning and multi-granularity feature fusion was proposed.First,an attention spatial transformation network(A-STN)is proposed to learn spatial features and solve the problem of misalignment of pedestrian spatial features.Then the network was divided into a global branch,a local coarse-grained fusion branch,and a local fine-grained fusion branch to extract pedestrian global features,coarse-grained fusion features,and fine-grained fusion features,respectively.Among them,the global branch enriches the global features by fusing different pooling features.The local coarse-grained fusion branch uses an overlay pooling to enhance each local feature while learning the correlation relationship between multi-granularity features.The local fine-grained fusion branch uses a differential pooling to obtain the differential features that were fused with global features to learn the relationship between pedestrian local features and pedestrian global features.Finally,the proposed method was compared on three public datasets:Market1501,DukeMTMC-ReID and CUHK03.The experimental results were better than those of the comparative methods,which verifies the effectiveness of the proposed method.
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%.
基金supported by the Fundamental Research Funds for the Central Universities under Grant 2020JKF101the Research Funds of Sugon under Grant 2022KY001.
文摘Rapid development of deepfake technology led to the spread of forged audios and videos across network platforms,presenting risks for numerous countries,societies,and individuals,and posing a serious threat to cyberspace security.To address the problem of insufficient extraction of spatial features and the fact that temporal features are not considered in the deepfake video detection,we propose a detection method based on improved CapsNet and temporal–spatial features(iCapsNet–TSF).First,the dynamic routing algorithm of CapsNet is improved using weight initialization and updating.Then,the optical flow algorithm is used to extract interframe temporal features of the videos to form a dataset of temporal–spatial features.Finally,the iCapsNet model is employed to fully learn the temporal–spatial features of facial videos,and the results are fused.Experimental results show that the detection accuracy of iCapsNet–TSF reaches 94.07%,98.83%,and 98.50%on the Celeb-DF,FaceSwap,and Deepfakes datasets,respectively,displaying a better performance than most existing mainstream algorithms.The iCapsNet–TSF method combines the capsule network and the optical flow algorithm,providing a novel strategy for the deepfake detection,which is of great significance to the prevention of deepfake attacks and the preservation of cyberspace security.
基金supported by the Fundamental Research Funds for the Provincial Universities of Zhejiang (No.GK249909299001-036)National Key Research and Development Program of China (No. 2023YFB4502803)Zhejiang Provincial Natural Science Foundation of China (No.LDT23F01014F01)。
文摘Due to the limitations of existing imaging hardware, obtaining high-resolution hyperspectral images is challenging. Hyperspectral image super-resolution(HSI SR) has been a very attractive research topic in computer vision, attracting the attention of many researchers. However, most HSI SR methods focus on the tradeoff between spatial resolution and spectral information, and cannot guarantee the efficient extraction of image information. In this paper, a multidimensional features network(MFNet) for HSI SR is proposed, which simultaneously learns and fuses the spatial,spectral, and frequency multidimensional features of HSI. Spatial features contain rich local details,spectral features contain the information and correlation between spectral bands, and frequency feature can reflect the global information of the image and can be used to obtain the global context of HSI. The fusion of the three features can better guide image super-resolution, to obtain higher-quality high-resolution hyperspectral images. In MFNet, we use the frequency feature extraction module(FFEM) to extract the frequency feature. On this basis, a multidimensional features extraction module(MFEM) is designed to learn and fuse multidimensional features. In addition, experimental results on two public datasets demonstrate that MFNet achieves state-of-the-art performance.
基金This work was supported by National Natural Science Foundation of China(Nos.62006225,61906199 and 62071468)the Strategic Priority Research Program of Chinese Academy of Sciences(CAS),China(No.XDA 27040700)sponsored by The Beijing Nova Program,China(Nos.Z201100006820050 and Z211100002121010).
文摘In the daily application of an iris-recognition-at-a-distance(IAAD)system,many ocular images of low quality are acquired.As the iris part of these images is often not qualified for the recognition requirements,the more accessible periocular regions are a good complement for recognition.To further boost the performance of IAAD systems,a novel end-to-end framework for multi-modal ocular recognition is proposed.The proposed framework mainly consists of iris/periocular feature extraction and matching,unsupervised iris quality assessment,and a score-level adaptive weighted fusion strategy.First,ocular feature reconstruction(OFR)is proposed to sparsely reconstruct each probe image by high-quality gallery images based on proper feature maps.Next,a brand new unsupervised iris quality assessment method based on random multiscale embedding robustness is proposed.Different from the existing iris quality assess-ment methods,the quality of an iris image is measured by its robustness in the embedding space.At last,the fusion strategy exploits the iris quality score as the fusion weight to coalesce the complementary information from the iris and periocular regions.Extensive experi-mental results on ocular datasets prove that the proposed method is obviously better than unimodal biometrics,and the fusion strategy can significantly improve therecognition performance.
基金supported by the National Natural Science Foundation of China(No.61275010)the Ph.D.Programs Foundation of Ministry of Education of China(No.20132304110007)+1 种基金the Heilongjiang Natural Science Foundation(No.F201409)the Fundamental Research Funds for the Central Universities(No.HEUCFD1410)
文摘To minimize the low classification accuracy and low utilization of spatial information in traditional hyperspectral image classification methods, we propose a new hyperspectral image classification method, which is based on the Gabor spatial texture features and nonparametric weighted spectral features, and the sparse representation classification method（Gabor–NWSF and SRC）, abbreviated GNWSF–SRC. The proposed（GNWSF–SRC） method first combines the Gabor spatial features and nonparametric weighted spectral features to describe the hyperspectral image, and then applies the sparse representation method. Finally, the classification is obtained by analyzing the reconstruction error. We use the proposed method to process two typical hyperspectral data sets with different percentages of training samples. Theoretical analysis and simulation demonstrate that the proposed method improves the classification accuracy and Kappa coefficient compared with traditional classification methods and achieves better classification performance.
基金Supported by the Special Coordination Funds for Promoting Sci-ence and Technology, and the Research Grant-In-Aid provided by the Ministry of Education, Culture, Sports, Science, and Technol-ogy, Japan
文摘The analysis of hidden spatial features is crucial for the improvement of hedonic regression models for analyzing the structure of land and housing prices. If critical variables representing the influence of spatial features are omitted in the models, the residuals and the coefficients estimated usually exhibit some kind of spatial pattern. Hence, exploration of the relationship between the spatial patterns and the spatial features essentially leads to the discovery of omitted variables. The analyses in this paper were based on two exploratory approaches: one on the residual of a global regression model and the other on the geographically weighted regression (GWR) technique. In the GWR model, the regression coefficients are al- lowed to differ by location so more spatial patterns can be revealed. Comparison of the two approaches shows that they play supplementary roles for the detection of lot-associated variables and area-associated variables.
基金Project from the Ministry of Science and Technology of China (2001DIB20116)open projectfor KLME of Nanjing Institute of Meteorology (KJS02108)
文摘41 a （1961 - 2001） seasonal Z index series of 25 representative weather stations are investigated by virtue of EOF, FFT, continuous wavelet transformation （CWT） and orthogonai wavelet transformation （OWT）. It shows that： （1） Fujian drought/flood （DF） has a significant 2 - 3a cycle for the periods 1965 - 1975 and 1990＇s; （2） the pattern, which represents the opposite DF trend between the southern and northem parts, has la and 3 - 4a cycles since the middle of 1980＇s; （3） EOF3, which denotes the reverse change between the middle-west region and other areas, has significant 1 - 2a cycle for the period from 1985 to 1998 and 9 - 13a cycle since 1980s; （4） there is an obvious drought trend for the last 40a （especially in the 1990＇s）, which is more outstanding in the south （east） than in the north （west）; （5） the 1960＇s and 1980＇s are in relatively wet phases and the 1970＇s and 1990＇s are in drought spells.
文摘Based on temperature data of meteorological stations from 1971 to 2008 in Tibet,the temporal and spatial variation of maximum andminimum temperature in Tibet was analyzed.The results showed that both maximum temperature andminimum temperature increased distinctly,the warming amplitude of winter was the highest among the four seasons,and next came spring.The increment ofminimum temperature was visibly over that of maximum temperature,particularlyminimum temperature in winter with significant increment.For spatial variation,maximum temperature in most stations increased except particular stations,while theminimum temperature in all stations rose.In addition,the space variation law ofminimum temperature,being more obvious thanminimum temperature,increased from southeast to northwest with different spatial changes in various seasons.From decadal variation,both maximum andminimum temperature appeared increase from 1970s to the first eight years in the 21st century,and the rise ofminimum temperature was significant greater than maximum temperature.The increase of maximum andminimum temperature was the highest from 2001 to 2008,whereas the lowest in 1970s.
基金Project from Natural Science Foundation of China (40775046)Project from Research Plan "973" (2006CB403601)
文摘The results of an analysis of the temporal and spatial distribution of typhoon precipitation influencing Fujian from 1960 to 2005 show that typhoon precipitation in Fujian province occurs from May to November, with the most in August. There has been a decreasing trend since 1960. Typhoon precipitation gradually decreases from the coastal region to the northwestern mainland of Fujian and the maximum typhoon precipitation occurs in the northeast and the south of Fujian. Typhoon torrential rain is one of the extreme rainfall events in Fujian. High frequencies of typhoon torrential rain occur in the coastal and southwest regions of the province. With the impact of Fujian＇s terrain, typhoon precipitation occurs more easily to the east of the mountains than to the west. Atmospheric circulation at 500 hPa over Asia and sea surface temperature anomalies of the equatorial eastern Pacific are analyzed, with the finding that they are closely connected with the anomaly of typhoon precipitation influencing Fujian, possibly mainly by modulating the northbound track of typhoons via changing the atmosphere circulation to lead to the anomaly of typhoon precipitation over the province
文摘Multi-Object Tracking(MOT)represents a fundamental but computationally demanding task in computer vision,with particular challenges arising in occluded and densely populated environments.While contemporary tracking systems have demonstrated considerable progress,persistent limitations—notably frequent occlusion-induced identity switches and tracking inaccuracies—continue to impede reliable real-world deployment.This work introduces an advanced tracking framework that enhances association robustness through a two-stage matching paradigm combining spatial and appearance features.Proposed framework employs:(1)a Height Modulated and Scale Adaptive Spatial Intersection-over-Union(HMSIoU)metric for improved spatial correspondence estimation across variable object scales and partial occlusions;(2)a feature extraction module generating discriminative appearance descriptors for identity maintenance;and(3)a recovery association mechanism for refining matches between unassociated tracks and detections.Comprehensive evaluation on standard MOT17 and MOT20 benchmarks demonstrates significant improvements in tracking consistency,with state-of-the-art performance across key metrics including HOTA(64),MOTA(80.7),IDF1(79.8),and IDs(1379).These results substantiate the efficacy of our Cue-Tracker framework in complex real-world scenarios characterized by occlusions and crowd interactions.
基金support from the Ministry of Education(MOE) Singapore Tier 1 (RG8/20)。
文摘A large database is desired for machine learning(ML) technology to make accurate predictions of materials physicochemical properties based on their molecular structure.When a large database is not available,the development of proper featurization method based on physicochemical nature of target proprieties can improve the predictive power of ML models with a smaller database.In this work,we show that two new featurization methods,volume occupation spatial matrix and heat contribution spatial matrix,can improve the accuracy in predicting energetic materials' crystal density(ρ_(crystal)) and solid phase enthalpy of formation(H_(f,solid)) using a database containing 451 energetic molecules.Their mean absolute errors are reduced from 0.048 g/cm~3 and 24.67 kcal/mol to 0.035 g/cm~3 and 9.66 kcal/mol,respectively.By leave-one-out-cross-validation,the newly developed ML models can be used to determine the performance of most kinds of energetic materials except cubanes.Our ML models are applied to predict ρ_(crystal) and H_(f,solid) of CHON-based molecules of the 150 million sized PubChem database,and screened out 56 candidates with competitive detonation performance and reasonable chemical structures.With further improvement in future,spatial matrices have the potential of becoming multifunctional ML simulation tools that could provide even better predictions in wider fields of materials science.
基金supported by National High Technology Research and Development Program of China (863 Program)(No.2007AA01Z416)National Natural Science Foundation of China (No.60773056)+1 种基金Beijing New Star Project on Science and Technology (No.2007B071)Natural Science Foundation of Liaoning Province of China (No.20052184)
文摘How to construct an appropriate spatial consistent measurement is the key to improving image retrieval performance. To address this problem, this paper introduces a novel image retrieval mechanism based on the family filtration in object region. First, we supply an object region by selecting a rectangle in a query image such that system returns a ranked list of images that contain the same object, retrieved from the corpus based on 100 images, as a result of the first rank. To further improve retrieval performance, we add an efficient spatial consistency stage, which is named family-based spatial consistency filtration, to re-rank the results returned by the first rank. We elaborate the performance of the retrieval system by some experiments on the dataset selected from the key frames of ＂TREC Video Retrieval Evaluation 2005 （TRECVID2005）＂. The results of experiments show that the retrieval mechanism proposed by us has vast major effect on the retrieval quality. The paper also verifies the stability of the retrieval mechanism by increasing the number of images from 100 to 2000 and realizes generalized retrieval with the object outside the dataset.
基金Funded by the Youth Chengguang Project of Science and Technology of Wuhan City of China(No.20045006071-16)
文摘GML is becoming the de facto standard for electronic data exchange among the applications of Web and distributed geographic information systems. However, the conventional query languages （e. g. SQL and its extended versions） are not suitable for direct querying and updating of GML documents. Even the effective approaches working well with XML could not guarantee good results when applied to GML documents. Although XQuery is a powerful standard query language for XML, it is not proposed for querying spatial features, which constitute the most important components in GML documents. We propose GQL, a query language specification to support spatial queries over GML documents by extending XQuery. The data model, algebra, and formal semantics as well as various spatial Junctions and operations of GQL are presented in detail.
基金supported by Light of West China(No.XAB2022YN10)Shaanxi Key Rsearch and Development Plan(No.2018ZDXM-SF-093)Shaanxi Province Key Industrial Innovation Chain(Nos.S2022-YF-ZDCXL-ZDLGY-0093,2023-ZDLGY-45).
文摘Sanxingdui cultural relics are the precious cultural heritage of humanity with high values of history,science,culture,art and research.However,mainstream analytical methods are contacting and detrimental,which is unfavorable to the protection of cultural relics.This paper improves the accuracy of the extraction,location,and analysis of artifacts using hyperspectral methods.To improve the accuracy of cultural relic mining,positioning,and analysis,the segmentation algorithm of Sanxingdui cultural relics based on the spatial spectrum integrated network is proposed with the support of hyperspectral techniques.Firstly,region stitching algorithm based on the relative position of hyper spectrally collected data is proposed to improve stitching efficiency.Secondly,given the prominence of traditional HRNet(High-Resolution Net)models in high-resolution data processing,the spatial attention mechanism is put forward to obtain spatial dimension information.Thirdly,in view of the prominence of 3D networks in spectral information acquisition,the pyramid 3D residual network model is proposed to obtain internal spectral dimensional information.Fourthly,four kinds of fusion methods at the level of data and decision are presented to achieve cultural relic labeling.As shown by the experiment results,the proposed network adopts an integrated method of data-level and decision-level,which achieves the optimal average accuracy of identification 0.84,realizes shallow coverage of cultural relics labeling,and effectively supports the mining and protection of cultural relics.
基金the funding support from the Vilas Associates Competition Award at University of Wisconsin-Madison(UW-Madison)the National Science Foundation[grant number 1940091].
文摘To find disaster relevant social media messages,current approaches utilize natural language processing methods or machine learning algorithms relying on text only,which have not been perfected due to the variability and uncertainty in the language used on social media and ignoring the geographic context of the messages when posted.Meanwhile,a disaster relevant social media message is highly sensitive to its posting location and time.However,limited studies exist to explore what spatial features and the extent of how temporal,and especially spatial features can aid text classification.This paper proposes a geographic context-aware text mining method to incorporate spatial and temporal information derived from social media and authoritative datasets,along with the text information,for classifying disaster relevant social media posts.This work designed and demonstrated how diverse types of spatial and temporal features can be derived from spatial data,and then used to enhance text mining.The deep learning-based method and commonly used machine learning algorithms,assessed the accuracy of the enhanced text-mining method.The performance results of different classification models generated by various combinations of textual,spatial,and temporal features indicate that additional spatial and temporal features help improve the overall accuracy of the classification.
基金funded by a pilot project entitled“Deep Geological Survey of Benxi-Linjiang Area”(1212011220247)of the 3D Geological Mapping and Deep Geological Survey of China Geological Survey。
文摘A method is proposed for the prospecting prediction of subsurface mineral deposits based on soil geochemistry data and a deep convolutional neural network model.This method uses three techniques(window offset,scaling,and rotation)to enhance the number of training data for the model.A window area is used to extract the spatial distribution characteristics of soil geochemistry and measure their correspondence with the occurrence of known subsurface deposits.Prospecting prediction is achieved by matching the characteristics of the window area of an unknown area with the relationships established in the known area.This method can efficiently predict mineral prospective areas where there are few ore deposits used for generating the training dataset,meaning that the deep-learning method can be effectively used for deposit prospecting prediction.Using soil active geochemical measurement data,this method was applied in the Daqiao area,Gansu Province,for which seven favorable gold prospecting target areas were predicted.The Daqiao orogenic gold deposit of latest Jurassic and Early Jurassic age in the southern domain has more than 105 t of gold resources at an average grade of 3-4 g/t.In 2020,the project team drilled and verified the K prediction area,and found 66 m gold mineralized bodies.The new method should be applicable to prospecting prediction using conventional geochemical data in other areas.
基金supported by Nature Science Research Project of Higher Education Institutions in Jiangsu Province under Grant No.21KJB510018National Nature Science Foundation of China (NSFC)under Grant No.62001215.
文摘Microphone array-based sound source localization(SSL)is widely used in a variety of occasions such as video conferencing,robotic hearing,speech enhancement,speech recognition and so on.The traditional SSL methods cannot achieve satisfactory performance in adverse noisy and reverberant environments.In order to improve localization performance,a novel SSL algorithm using convolutional residual network(CRN)is proposed in this paper.The spatial features including time difference of arrivals(TDOAs)between microphone pairs and steered response power-phase transform(SRPPHAT)spatial spectrum are extracted in each Gammatone sub-band.The spatial features of different sub-bands with a frame are combine into a feature matrix as the input of CRN.The proposed algorithm employ CRN to fuse the spatial features.Since the CRN introduces the residual structure on the basis of the convolutional network,it reduce the difficulty of training procedure and accelerate the convergence of the model.A CRN model is learned from the training data in various reverberation and noise environments to establish the mapping regularity between the input feature and the sound azimuth.Through simulation verification,compared with the methods using traditional deep neural network,the proposed algorithm can achieve a better localization performance in SSL task,and provide better generalization capacity to untrained noise and reverberation.
基金financially supported by the National Basic Research Development Program of China(Grant Nos.2011CB952001 and 2012CB95570001)the National Natural Science Foundation of China(Grant No.41301076)
文摘The diurnal temperature range(DTR) has decreased dramatically in recent decades, but it is not yet obvious whether the extreme values of DTR have also reduced. Based on the daily maximum and minimum temperature data of 653 stations in China, a set of monthly indices of warm extremes, cold extremes, and DTR extremes in summer(June, July, August) and winter(December, January, February) were studied for spatial and temporal features during the period 1971–2013. Results show that the incidence of warm extremes has been increasing in most parts of China, while the opposite trend was found in the cold extremes for summer and winter months. Both increasing and decreasing trends of monthly DTR extremes were identified in China for both seasons. For high DTR extremes, decreasing trends were identified in northern China for both seasons, but increasing trends were found only in southern China in summer, while in winter, they were found in central China. Monthly low DTR extreme indices demonstrated consistent positive trends in summer and winter, while significant increases(P < 0.05) were identified for only a few stations.
文摘In medical research and clinical diagnosis, automated or computer-assisted classification and retrieval methods are highly desirable to offset the high cost of manual classification and manipulation by medical experts. To facilitate the decision-making in the health-care and the related areas, in this paper, a two-step content-based medical image retrieval algorithm is proposed. Firstly, in the preprocessing step, the image segmentation is performed to distinguish image objects, and on the basis of the ...
