Unlike traditional video cameras,event cameras capture asynchronous event streams in which each event encodes pixel location,triggers’timestamps,and the polarity of brightness changes.In this paper,we introduce a nov...Unlike traditional video cameras,event cameras capture asynchronous event streams in which each event encodes pixel location,triggers’timestamps,and the polarity of brightness changes.In this paper,we introduce a novel hypergraph-based framework for moving object classification.Specifically,we capture moving objects with an event camera,to perceive and collect asynchronous event streams in a high temporal resolution.Unlike stacked event frames,we encode asynchronous event data into a hypergraph,fully mining the high-order correlation of event data,and designing a mixed convolutional hypergraph neural network for training to achieve a more efficient and accurate motion target recognition.The experimental results show that our method has a good performance in moving object classification(e.g.,gait identification).展开更多
Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small obje...Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small object detection a complex and demanding task.One effective approach to overcome these issues is the integration of multimodal image data to enhance detection capabilities.This paper proposes a novel small object detection method that utilizes three types of multimodal image combinations,such as Hyperspectral-Multispectral(HSMS),Hyperspectral-Synthetic Aperture Radar(HS-SAR),and HS-SAR-Digital Surface Model(HS-SAR-DSM).The detection process is done by the proposed Jaccard Deep Q-Net(JDQN),which integrates the Jaccard similarity measure with a Deep Q-Network(DQN)using regression modeling.To produce the final output,a Deep Maxout Network(DMN)is employed to fuse the detection results obtained from each modality.The effectiveness of the proposed JDQN is validated using performance metrics,such as accuracy,Mean Squared Error(MSE),precision,and Root Mean Squared Error(RMSE).Experimental results demonstrate that the proposed JDQN method outperforms existing approaches,achieving the highest accuracy of 0.907,a precision of 0.904,the lowest normalized MSE of 0.279,and a normalized RMSE of 0.528.展开更多
This paper uses three size metrics,which are collectable during the design phase,to analyze the potentially confounding effect of class size on the associations between object-oriented（OO）metrics and maintainability...This paper uses three size metrics,which are collectable during the design phase,to analyze the potentially confounding effect of class size on the associations between object-oriented（OO）metrics and maintainability.To draw as many general conclusions as possible,the confounding effect of class size is analyzed on 127 C＋＋ systems and 113 Java systems.For each OO metric,the indirect effect that represents the distortion of the association caused by class size and its variance for individual systems is first computed.Then,a statistical meta-analysis technique is used to compute the average indirect effect over all the systems and to determine if it is significantly different from zero.The experimental results show that the confounding effects of class size on the associations between OO metrics and maintainability generally exist,regardless of whatever size metric is used.Therefore,empirical studies validating OO metrics on maintainability should consider class size as a confounding variable.展开更多
For classifying unknown 3-D objects into a set of predetermined object classes, a part-level object classification method based on the improved interpretation tree is presented. The part-level representation is implem...For classifying unknown 3-D objects into a set of predetermined object classes, a part-level object classification method based on the improved interpretation tree is presented. The part-level representation is implemented, which enables a more compact shape description of 3-D objects. The proposed classification method consists of two key processing stages： the improved constrained search on an interpretation tree and the following shape similarity measure computation. By the classification method, both whole match and partial match with shape similarity ranks are achieved; especially, focus match can be accomplished, where different key parts may be labeled and all the matched models containing corresponding key parts may be obtained. A series of experiments show the effectiveness of the presented 3-D object classification method.展开更多
An object oriented coal mining land cover classification method based on semantically meaningful image segmentation and image combination of GeoEye imagery and airborne laser scanning (ALS) data was presented. First, ...An object oriented coal mining land cover classification method based on semantically meaningful image segmentation and image combination of GeoEye imagery and airborne laser scanning (ALS) data was presented. First, DEM, DSM and nDSM (normalized Digital Surface Model, nDSM) were extracted from ALS data. The GeoEye imagery and DSM data were combined to create segmented objects based on neighbor regions merge method. Then 10 kinds of objects were extracted. Different kinds of vegetation objects, including crop, grass, shrub and tree, can be extracted by using NDVI and height value of nDSM. Water and coal pile field was extracted by using NDWI and the standard deviation of DSM method. Height differences also can be used to distinguish buildings from road and vacant land, and accurate building contour information can be extracted by using relationship of neighbor objects and morphological method. The test result shows that the total classification accuracy of the presented method is 90.78% and the kappa coefficient is 0.891 4.展开更多
Object-based classification differentiates forest gaps from canopies at large regional scale by using remote sensing data. To study the segmentation and classification processes of object-based forest gaps classificat...Object-based classification differentiates forest gaps from canopies at large regional scale by using remote sensing data. To study the segmentation and classification processes of object-based forest gaps classification at a regional scale, we sampled a natural secondary forest in northeast China at Maoershan Experimental Forest Farm.Airborne light detection and ranging(LiDAR; 3.7 points/m2) data were collected as the original data source and the canopy height model(CHM) and topographic dataset were extracted from the LiDAR data. The accuracy of objectbased forest gaps classification depends on previous segmentation. Thus our first step was to define 10 different scale parameters in CHM image segmentation. After image segmentation, the machine learning classification method was used to classify three kinds of object classes, namely,forest gaps, tree canopies, and others. The common support vector machine(SVM) classifier with the radial basis function kernel(RBF) was first adopted to test the effect of classification features(vegetation height features and some typical topographic features) on forest gap classification.Then the different classifiers(KNN, Bayes, decision tree,and SVM with linear kernel) were further adopted to compare the effect of classifiers on machine learning forest gaps classification. Segmentation accuracy and classification accuracy were evaluated by using Mo¨ller's method and confusion metrics, respectively. The scale parameter had a significant effect on object-based forest gap segmentation and classification. Classification accuracies at different scales revealed that there were two optimal scales(10 and 20) that provided similar accuracy, with the scale of 10 yielding slightly greater accuracy than 20. The accuracy of the classification by using combination of height features and SVM classifier with linear kernel was91% at the optimal scale parameter of 10, and it was highest comparing with other classification classifiers, such as SVM RBF(90%), Decision Tree(90%), Bayes(90%),or KNN(87%). The classifiers had no significant effect on forest gap classification, but the fewer parameters in the classifier equation and higher speed of operation probably lead to a higher accuracy of final classifications. Our results confirm that object-based classification can extract forest gaps at a large regional scale with appropriate classification features and classifiers using LiDAR data. We note, however, that final satisfaction of forest gap classification depends on the determination of optimal scale(s) of segmentation.展开更多
An objective analysis of tropical cyclone tracks is performed, with which the tracks of 131 tropical storms(TSs) in 1972-2011 are separated into three types that move west-, north- and northwestward, denoted as Types ...An objective analysis of tropical cyclone tracks is performed, with which the tracks of 131 tropical storms(TSs) in 1972-2011 are separated into three types that move west-, north- and northwestward, denoted as Types A, B and C, respectively. Type A(21 TSs and 16% of total) has the origin in the southwestern Bay of Bengal, with the TS in a unimodal distribution as its seasonal feature, occurring mainly in autumn; 18 of the 21 TSs(taking up 90%) land mostly on the western Bay coast(west of 85°E); 5% of Type-A TSs attains the wind speed of >42.7 to 48.9 m/s. Type A has little or no effect on Tibet. Type B(74 TSs, 56.6% of the total) has its preferable origin in the central Bay of Bengal, with the TS in a bimodal distribution as its seasonal pattern. This type denotes the travel in the north in spring,with the landfall of 67 of the 74 TSs(accounting for 91%) mainly on the middle coast of the Bay(85° to 95°E), and19% of the TSs reaching the wind velocity of >42.7 to 48.9 m/s, which exert great effect on Tibet and it is this TS track that gives strong precipitation on its way through this region. Type C(36 TSs, 27.5% of the total) has its main origin in the southern part of the bay, and these TSs are formed largely in autumn, moving in the northwest direction,and 23 of the 36 TSs(64%) land mostly on the western Bay coast, lasting for a longer time, with almost no impact upon Tibet.展开更多
Taking TM images, SPOT photos and DEM images as the basic information, this paper had not only put forward a kind of manual controlled computer-automatic extraction method, but also completed the task of extracting th...Taking TM images, SPOT photos and DEM images as the basic information, this paper had not only put forward a kind of manual controlled computer-automatic extraction method, but also completed the task of extracting the main types of ground objects in the Three Gorges Reservoir area under relatively high accuracy, after finishing such preprocessing tasks as correcting the topographical spectrum and synthesizing the data. Taking the specialized image analysis software-eCognition as the platform, the research achieved the goal of classifying through choosing samples, picking out the best wave bands, and producing the identifying functions. At the same time the extraction process partly dispelled the influence of such phenomena as the same thing with different spectrums, different things with the same spectrum, border transitions, etc. The research did certain exploration in the aspect of technological route and method of using automatic extraction of the remote sensing image to obtain the information of land cover for the regions whose ground objects have complicated spectrums.展开更多
Moving object extraction and classification are important problems in automated video surveillance systems. A background model based on region segmentation is proposed. An adaptive single Gaussian background model is ...Moving object extraction and classification are important problems in automated video surveillance systems. A background model based on region segmentation is proposed. An adaptive single Gaussian background model is used in the stable region with gradual changes, and a nonparametric model is used in the variable region with jumping changes. A generalized agglomerative scheme is used to merge the pixels in the variable region and fill in the small interspaces. A two-threshold sequential algorithmic scheme is used to group the background samples of the variable region into distinct Gaussian distributions to accelerate the kernel density computation speed of the nonparametric model. In the feature-based object classification phase, the surveillance scene is first partitioned according to the road boundaries of different traffic directions and then re-segmented according to their scene localities. The method improves the discriminability of the features in each partition. AdaBoost method is applied to evaluate the relative importance of the features in each partition respectively and distinguish whether an object is a vehicle, a single human, a human group, or a bike. Experimental results show that the proposed method achieves higher performance in comparison with the existing method.展开更多
A Support Vector Machine is used as a classifier to the automatic detection and recognition of underwater still objects. Discrimination between the objects can be transferred into different projection spaces by the pr...A Support Vector Machine is used as a classifier to the automatic detection and recognition of underwater still objects. Discrimination between the objects can be transferred into different projection spaces by the process of multi-field feature extraction. The multi-field feature vector includes time-domain, spectral, time-frequency distribution and bi-spectral features. Underwater target recognition can be considered as a problem of small sample recognition. SVM algorithm is appropriate to this kind of problems because of its outstanding generalizability. The SVM is contrasted with a Gaussian classifier and a k-nearest classifier in some experiments using real data of lake or sea trial. The experimental results indicate that SVM is better than the others two.展开更多
We have collected a sample of 70 BL Lacs （33 radio-selected BL Lacs and 37 X-ray selected BL Lacs） with multi-waveband data for investigating the classifying criteria of BL Lacertae Objects. For each source, we esti...We have collected a sample of 70 BL Lacs （33 radio-selected BL Lacs and 37 X-ray selected BL Lacs） with multi-waveband data for investigating the classifying criteria of BL Lacertae Objects. For each source, we estimate its luminosities in radio, optical and X-ray, the broad-band spectral index from radio to X-ray and the peak frequency of the synchrotron emission, and make a statistical analysis of the data obtained. Our main results are as follows： （1） The broad-band spectral index and the peak frequency have no correlation with the redshift, while they are inversely correlated with each other and they could be regarded as equivalent classifying criteria of BL Lac objects. （2） There are significant effects of the luminosity/redshift relation on the observed luminosity distribution in our sample, hence, if the radio luminosity is to be used as a classifying criterion of BL Lac objects, it should not be regarded as equivalent to the broad-band spectral index or the peak frequency. （3） Our resuits supply a specific piece of evidence for the suggestion that the use of luminosities always introduces a redshift bias to the data and show that the location of the peak frequency is not always linked to the luminosity of any wave band.展开更多
Object tracking is one of the killer applications for wireless sensor networks (WSN) in which the network of wireless sensors is assigned the task of tracking a particular object. The network employs the object tracki...Object tracking is one of the killer applications for wireless sensor networks (WSN) in which the network of wireless sensors is assigned the task of tracking a particular object. The network employs the object tracking techniques to continuously report the position of the object in terms of Cartesian coordinates to a sink node or to a central base station. A family tree of object tracking techniques has been prepared.In this paper we have summarized the object tracking techniques available so far in wireless sensor networks.展开更多
Inclusion of textures in image classification has been shown beneficial.This paper studies an efficient use of semivariogram features for object-based high-resolution image classification.First,an input image is divid...Inclusion of textures in image classification has been shown beneficial.This paper studies an efficient use of semivariogram features for object-based high-resolution image classification.First,an input image is divided into segments,for each of which a semivariogram is then calculated.Second,candidate features are extracted as a number of key locations of the semivariogram functions.Then we use an improved Relief algorithm and the principal component analysis to select independent and significant features.Then the selected prominent semivariogram features and the conventional spectral features are combined to constitute a feature vector for a support vector machine classifier.The effect of such selected semivariogram features is compared with those of the gray-level co-occurrence matrix(GLCM)features and window-based semivariogram texture features(STFs).Tests with aerial and satellite images show that such selected semivariogram features are of a more beneficial supplement to spectral features.The described method in this paper yields a higher classification accuracy than the combination of spectral and GLCM features or STFs.展开更多
Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and...Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and classification of ground objects at home and abroad. The research results of identification and classification of forest tree species, grassland and urban land features were summarized. Then the researches of classification methods were summarized. Finally the prospects of hyperspectral remote sensing in ground object identification and classification were prospected.展开更多
In this Paper, a classification method based on neural networks is presented for recognition of 3D objects. Indeed, the objective of this paper is to classify an object query against objects in a database, which leads...In this Paper, a classification method based on neural networks is presented for recognition of 3D objects. Indeed, the objective of this paper is to classify an object query against objects in a database, which leads to recognition of the former. 3D objects of this database are transformations of other objects by one element of the overall transformation. The set of transformations considered in this work is the general affine group.展开更多
An efficient convolution neural network(CNN) plays a crucial role in various visual tasks like object classification or detection, etc. The most common way to construct a CNN is stacking the same convolution block or ...An efficient convolution neural network(CNN) plays a crucial role in various visual tasks like object classification or detection, etc. The most common way to construct a CNN is stacking the same convolution block or complex connection. These approaches may be efficient but the parameter size and computation(Comp) have explosive growth. So we present a novel architecture called"DLA+", which could obtain the feature from the different stages, and by the newly designed convolution block, could achieve better accuracy, while also dropping the computation six times compared to the baseline. We design some experiments about classification and object detection. On the CIFAR10 and VOC data-sets, we get better precision and faster speed than other architecture. The lightweight network even allows us to deploy to some low-performance device like drone, laptop, etc.展开更多
The Smart City concept revolves around gathering real time data from citizen,personal vehicle,public transports,building,and other urban infrastructures like power grid and waste disposal system.The understandings obt...The Smart City concept revolves around gathering real time data from citizen,personal vehicle,public transports,building,and other urban infrastructures like power grid and waste disposal system.The understandings obtained from the data can assist municipal authorities handle assets and services effectually.At the same time,the massive increase in environmental pollution and degradation leads to ecological imbalance is a hot research topic.Besides,the progressive development of smart cities over the globe requires the design of intelligent waste management systems to properly categorize the waste depending upon the nature of biodegradability.Few of the commonly available wastes are paper,paper boxes,food,glass,etc.In order to classify the waste objects,computer vision based solutions are cost effective to separate out the waste from the huge dump of garbage and trash.Due to the recent developments of deep learning(DL)and deep reinforcement learning(DRL),waste object classification becomes possible by the identification and detection of wastes.In this aspect,this paper designs an intelligence DRL based recycling waste object detection and classification(IDRL-RWODC)model for smart cities.The goal of the IDRLRWODC technique is to detect and classify waste objects using the DL and DRL techniques.The IDRL-RWODC technique encompasses a twostage process namely Mask Regional Convolutional Neural Network(Mask RCNN)based object detection and DRL based object classification.In addition,DenseNet model is applied as a baseline model for the Mask RCNN model,and a deep Q-learning network(DQLN)is employed as a classifier.Moreover,a dragonfly algorithm(DFA)based hyperparameter optimizer is derived for improving the efficiency of the DenseNet model.In order to ensure the enhanced waste classification performance of the IDRL-RWODC technique,a series of simulations take place on benchmark dataset and the experimental results pointed out the better performance over the recent techniques with maximal accuracy of 0.993.展开更多
Statistical and contextual information are typically used to detect moving regions in image sequences for a fixed camera.In this paper,we propose a fast and stable linear discriminant approach based on Gaussian Single...Statistical and contextual information are typically used to detect moving regions in image sequences for a fixed camera.In this paper,we propose a fast and stable linear discriminant approach based on Gaussian Single Model(GSM)and Markov Random Field(MRF).The performance of GSM is analyzed first,and then two main improvements corresponding to the drawbacks of GSM are proposed:the latest filtered data based update scheme of the background model and the linear classification judgment rule based on spatial-temporal feature specified by MRF.Experimental results show that the proposed method runs more rapidly and accurately when compared with other methods.展开更多
Moments have been used in all sorts of object classification systems based on image. There are lots of moments studied by many researchers in the area of object classification and one of the most preference moments is...Moments have been used in all sorts of object classification systems based on image. There are lots of moments studied by many researchers in the area of object classification and one of the most preference moments is the Zernike moment. In this paper, the performance of object classification using the Zernike moment has been explored. The classifier based on neural networks has been used in this study. The results indicate the best performance in identifying the aggregate is at 91.4% with a ten orders of the Zernike moment. This encouraging result has shown that the Zernike moment is a suitable moment to be used as a feature of object classification systems.展开更多
基金the National Key Research and Development Program of China(No.2021ZD0112400)。
文摘Unlike traditional video cameras,event cameras capture asynchronous event streams in which each event encodes pixel location,triggers’timestamps,and the polarity of brightness changes.In this paper,we introduce a novel hypergraph-based framework for moving object classification.Specifically,we capture moving objects with an event camera,to perceive and collect asynchronous event streams in a high temporal resolution.Unlike stacked event frames,we encode asynchronous event data into a hypergraph,fully mining the high-order correlation of event data,and designing a mixed convolutional hypergraph neural network for training to achieve a more efficient and accurate motion target recognition.The experimental results show that our method has a good performance in moving object classification(e.g.,gait identification).
文摘Accurate detection of small objects is critically important in high-stakes applications such as military reconnaissance and emergency rescue.However,low resolution,occlusion,and background interference make small object detection a complex and demanding task.One effective approach to overcome these issues is the integration of multimodal image data to enhance detection capabilities.This paper proposes a novel small object detection method that utilizes three types of multimodal image combinations,such as Hyperspectral-Multispectral(HSMS),Hyperspectral-Synthetic Aperture Radar(HS-SAR),and HS-SAR-Digital Surface Model(HS-SAR-DSM).The detection process is done by the proposed Jaccard Deep Q-Net(JDQN),which integrates the Jaccard similarity measure with a Deep Q-Network(DQN)using regression modeling.To produce the final output,a Deep Maxout Network(DMN)is employed to fuse the detection results obtained from each modality.The effectiveness of the proposed JDQN is validated using performance metrics,such as accuracy,Mean Squared Error(MSE),precision,and Root Mean Squared Error(RMSE).Experimental results demonstrate that the proposed JDQN method outperforms existing approaches,achieving the highest accuracy of 0.907,a precision of 0.904,the lowest normalized MSE of 0.279,and a normalized RMSE of 0.528.
基金The National Natural Science Foundation of China(No.60425206,60633010)
文摘This paper uses three size metrics,which are collectable during the design phase,to analyze the potentially confounding effect of class size on the associations between object-oriented（OO）metrics and maintainability.To draw as many general conclusions as possible,the confounding effect of class size is analyzed on 127 C＋＋ systems and 113 Java systems.For each OO metric,the indirect effect that represents the distortion of the association caused by class size and its variance for individual systems is first computed.Then,a statistical meta-analysis technique is used to compute the average indirect effect over all the systems and to determine if it is significantly different from zero.The experimental results show that the confounding effects of class size on the associations between OO metrics and maintainability generally exist,regardless of whatever size metric is used.Therefore,empirical studies validating OO metrics on maintainability should consider class size as a confounding variable.
基金The National Basic Research Program of China(973Program)(No2006CB303105)the Research Foundation of Bei-jing Jiaotong University (NoK06J0170)
文摘For classifying unknown 3-D objects into a set of predetermined object classes, a part-level object classification method based on the improved interpretation tree is presented. The part-level representation is implemented, which enables a more compact shape description of 3-D objects. The proposed classification method consists of two key processing stages： the improved constrained search on an interpretation tree and the following shape similarity measure computation. By the classification method, both whole match and partial match with shape similarity ranks are achieved; especially, focus match can be accomplished, where different key parts may be labeled and all the matched models containing corresponding key parts may be obtained. A series of experiments show the effectiveness of the presented 3-D object classification method.
基金Project(2009CB226107)supported by the National Basic Research Program of China
文摘An object oriented coal mining land cover classification method based on semantically meaningful image segmentation and image combination of GeoEye imagery and airborne laser scanning (ALS) data was presented. First, DEM, DSM and nDSM (normalized Digital Surface Model, nDSM) were extracted from ALS data. The GeoEye imagery and DSM data were combined to create segmented objects based on neighbor regions merge method. Then 10 kinds of objects were extracted. Different kinds of vegetation objects, including crop, grass, shrub and tree, can be extracted by using NDVI and height value of nDSM. Water and coal pile field was extracted by using NDWI and the standard deviation of DSM method. Height differences also can be used to distinguish buildings from road and vacant land, and accurate building contour information can be extracted by using relationship of neighbor objects and morphological method. The test result shows that the total classification accuracy of the presented method is 90.78% and the kappa coefficient is 0.891 4.
基金financially supported by grant from National Natural Science Foundation of China(No.31300533)
文摘Object-based classification differentiates forest gaps from canopies at large regional scale by using remote sensing data. To study the segmentation and classification processes of object-based forest gaps classification at a regional scale, we sampled a natural secondary forest in northeast China at Maoershan Experimental Forest Farm.Airborne light detection and ranging(LiDAR; 3.7 points/m2) data were collected as the original data source and the canopy height model(CHM) and topographic dataset were extracted from the LiDAR data. The accuracy of objectbased forest gaps classification depends on previous segmentation. Thus our first step was to define 10 different scale parameters in CHM image segmentation. After image segmentation, the machine learning classification method was used to classify three kinds of object classes, namely,forest gaps, tree canopies, and others. The common support vector machine(SVM) classifier with the radial basis function kernel(RBF) was first adopted to test the effect of classification features(vegetation height features and some typical topographic features) on forest gap classification.Then the different classifiers(KNN, Bayes, decision tree,and SVM with linear kernel) were further adopted to compare the effect of classifiers on machine learning forest gaps classification. Segmentation accuracy and classification accuracy were evaluated by using Mo¨ller's method and confusion metrics, respectively. The scale parameter had a significant effect on object-based forest gap segmentation and classification. Classification accuracies at different scales revealed that there were two optimal scales(10 and 20) that provided similar accuracy, with the scale of 10 yielding slightly greater accuracy than 20. The accuracy of the classification by using combination of height features and SVM classifier with linear kernel was91% at the optimal scale parameter of 10, and it was highest comparing with other classification classifiers, such as SVM RBF(90%), Decision Tree(90%), Bayes(90%),or KNN(87%). The classifiers had no significant effect on forest gap classification, but the fewer parameters in the classifier equation and higher speed of operation probably lead to a higher accuracy of final classifications. Our results confirm that object-based classification can extract forest gaps at a large regional scale with appropriate classification features and classifiers using LiDAR data. We note, however, that final satisfaction of forest gap classification depends on the determination of optimal scale(s) of segmentation.
基金Specialized Project for Public Welfare Industries(GYHY201106005)Showcase Project for Novel Technology at China Meteorological Administration(CMATG2010M25)
文摘An objective analysis of tropical cyclone tracks is performed, with which the tracks of 131 tropical storms(TSs) in 1972-2011 are separated into three types that move west-, north- and northwestward, denoted as Types A, B and C, respectively. Type A(21 TSs and 16% of total) has the origin in the southwestern Bay of Bengal, with the TS in a unimodal distribution as its seasonal feature, occurring mainly in autumn; 18 of the 21 TSs(taking up 90%) land mostly on the western Bay coast(west of 85°E); 5% of Type-A TSs attains the wind speed of >42.7 to 48.9 m/s. Type A has little or no effect on Tibet. Type B(74 TSs, 56.6% of the total) has its preferable origin in the central Bay of Bengal, with the TS in a bimodal distribution as its seasonal pattern. This type denotes the travel in the north in spring,with the landfall of 67 of the 74 TSs(accounting for 91%) mainly on the middle coast of the Bay(85° to 95°E), and19% of the TSs reaching the wind velocity of >42.7 to 48.9 m/s, which exert great effect on Tibet and it is this TS track that gives strong precipitation on its way through this region. Type C(36 TSs, 27.5% of the total) has its main origin in the southern part of the bay, and these TSs are formed largely in autumn, moving in the northwest direction,and 23 of the 36 TSs(64%) land mostly on the western Bay coast, lasting for a longer time, with almost no impact upon Tibet.
基金Under the auspices of the Construction Committeeof Three GorgesR eservoirProject(No .SX [2002]00401) andChineseAcademy ofSciences(No .KZCX2-SW-319-01 )
文摘Taking TM images, SPOT photos and DEM images as the basic information, this paper had not only put forward a kind of manual controlled computer-automatic extraction method, but also completed the task of extracting the main types of ground objects in the Three Gorges Reservoir area under relatively high accuracy, after finishing such preprocessing tasks as correcting the topographical spectrum and synthesizing the data. Taking the specialized image analysis software-eCognition as the platform, the research achieved the goal of classifying through choosing samples, picking out the best wave bands, and producing the identifying functions. At the same time the extraction process partly dispelled the influence of such phenomena as the same thing with different spectrums, different things with the same spectrum, border transitions, etc. The research did certain exploration in the aspect of technological route and method of using automatic extraction of the remote sensing image to obtain the information of land cover for the regions whose ground objects have complicated spectrums.
基金supported by the Science and Technology Program of Zhejiang Province of China(2005C11001-02).
文摘Moving object extraction and classification are important problems in automated video surveillance systems. A background model based on region segmentation is proposed. An adaptive single Gaussian background model is used in the stable region with gradual changes, and a nonparametric model is used in the variable region with jumping changes. A generalized agglomerative scheme is used to merge the pixels in the variable region and fill in the small interspaces. A two-threshold sequential algorithmic scheme is used to group the background samples of the variable region into distinct Gaussian distributions to accelerate the kernel density computation speed of the nonparametric model. In the feature-based object classification phase, the surveillance scene is first partitioned according to the road boundaries of different traffic directions and then re-segmented according to their scene localities. The method improves the discriminability of the features in each partition. AdaBoost method is applied to evaluate the relative importance of the features in each partition respectively and distinguish whether an object is a vehicle, a single human, a human group, or a bike. Experimental results show that the proposed method achieves higher performance in comparison with the existing method.
基金Supported by the Major State Basic Research Development Program of China under Grant No. 5132103ZZT32.
文摘A Support Vector Machine is used as a classifier to the automatic detection and recognition of underwater still objects. Discrimination between the objects can be transferred into different projection spaces by the process of multi-field feature extraction. The multi-field feature vector includes time-domain, spectral, time-frequency distribution and bi-spectral features. Underwater target recognition can be considered as a problem of small sample recognition. SVM algorithm is appropriate to this kind of problems because of its outstanding generalizability. The SVM is contrasted with a Gaussian classifier and a k-nearest classifier in some experiments using real data of lake or sea trial. The experimental results indicate that SVM is better than the others two.
基金the National Natural Science Foundation of China
文摘We have collected a sample of 70 BL Lacs （33 radio-selected BL Lacs and 37 X-ray selected BL Lacs） with multi-waveband data for investigating the classifying criteria of BL Lacertae Objects. For each source, we estimate its luminosities in radio, optical and X-ray, the broad-band spectral index from radio to X-ray and the peak frequency of the synchrotron emission, and make a statistical analysis of the data obtained. Our main results are as follows： （1） The broad-band spectral index and the peak frequency have no correlation with the redshift, while they are inversely correlated with each other and they could be regarded as equivalent classifying criteria of BL Lac objects. （2） There are significant effects of the luminosity/redshift relation on the observed luminosity distribution in our sample, hence, if the radio luminosity is to be used as a classifying criterion of BL Lac objects, it should not be regarded as equivalent to the broad-band spectral index or the peak frequency. （3） Our resuits supply a specific piece of evidence for the suggestion that the use of luminosities always introduces a redshift bias to the data and show that the location of the peak frequency is not always linked to the luminosity of any wave band.
文摘Object tracking is one of the killer applications for wireless sensor networks (WSN) in which the network of wireless sensors is assigned the task of tracking a particular object. The network employs the object tracking techniques to continuously report the position of the object in terms of Cartesian coordinates to a sink node or to a central base station. A family tree of object tracking techniques has been prepared.In this paper we have summarized the object tracking techniques available so far in wireless sensor networks.
基金This work was supported by the National Natural Science Foundation of China[grant number 41101410]the Comprehensive Transportation Applications of High-resolution Remote Sensing program[grant number 07-Y30B10-9001-14/16]+1 种基金the Key Laboratory of Surveying Mapping and Geoinformation in Geographical Condition Monitoring[grant number 2014NGCM]the Science and Technology Plan of Sichuan Bureau of Surveying,Mapping and Geoinformation,China[grant number J2014ZC02].
文摘Inclusion of textures in image classification has been shown beneficial.This paper studies an efficient use of semivariogram features for object-based high-resolution image classification.First,an input image is divided into segments,for each of which a semivariogram is then calculated.Second,candidate features are extracted as a number of key locations of the semivariogram functions.Then we use an improved Relief algorithm and the principal component analysis to select independent and significant features.Then the selected prominent semivariogram features and the conventional spectral features are combined to constitute a feature vector for a support vector machine classifier.The effect of such selected semivariogram features is compared with those of the gray-level co-occurrence matrix(GLCM)features and window-based semivariogram texture features(STFs).Tests with aerial and satellite images show that such selected semivariogram features are of a more beneficial supplement to spectral features.The described method in this paper yields a higher classification accuracy than the combination of spectral and GLCM features or STFs.
文摘Hyperspectral remote sensing has become one of the research frontiers in ground object identification and classification. On the basis of reviewing the application of hyperspectral remote sensing in identification and classification of ground objects at home and abroad. The research results of identification and classification of forest tree species, grassland and urban land features were summarized. Then the researches of classification methods were summarized. Finally the prospects of hyperspectral remote sensing in ground object identification and classification were prospected.
文摘In this Paper, a classification method based on neural networks is presented for recognition of 3D objects. Indeed, the objective of this paper is to classify an object query against objects in a database, which leads to recognition of the former. 3D objects of this database are transformations of other objects by one element of the overall transformation. The set of transformations considered in this work is the general affine group.
基金supported by University Synergy Innovation Program of Anhui Province (No. GXXT-2019-007)Corporative Information Processing and Deep Mining for Intelligent Robot (No. JCYJ20170817155854115)+1 种基金Major Project for New Generation of AI (No.2018AAA0100400)Anhui Provincial Natural Science Foundation (No. 1908085MF206)。
文摘An efficient convolution neural network(CNN) plays a crucial role in various visual tasks like object classification or detection, etc. The most common way to construct a CNN is stacking the same convolution block or complex connection. These approaches may be efficient but the parameter size and computation(Comp) have explosive growth. So we present a novel architecture called"DLA+", which could obtain the feature from the different stages, and by the newly designed convolution block, could achieve better accuracy, while also dropping the computation six times compared to the baseline. We design some experiments about classification and object detection. On the CIFAR10 and VOC data-sets, we get better precision and faster speed than other architecture. The lightweight network even allows us to deploy to some low-performance device like drone, laptop, etc.
基金The authors extend their appreciation to the Deanship of Scientific Research at King Khalid University for funding this work under Grant Number(RGP 1/282/42)This research was funded by the Deanship of Scientific Research at Princess Nourah bint Abdulrahman University through the Fast-Track Research Funding Program。
文摘The Smart City concept revolves around gathering real time data from citizen,personal vehicle,public transports,building,and other urban infrastructures like power grid and waste disposal system.The understandings obtained from the data can assist municipal authorities handle assets and services effectually.At the same time,the massive increase in environmental pollution and degradation leads to ecological imbalance is a hot research topic.Besides,the progressive development of smart cities over the globe requires the design of intelligent waste management systems to properly categorize the waste depending upon the nature of biodegradability.Few of the commonly available wastes are paper,paper boxes,food,glass,etc.In order to classify the waste objects,computer vision based solutions are cost effective to separate out the waste from the huge dump of garbage and trash.Due to the recent developments of deep learning(DL)and deep reinforcement learning(DRL),waste object classification becomes possible by the identification and detection of wastes.In this aspect,this paper designs an intelligence DRL based recycling waste object detection and classification(IDRL-RWODC)model for smart cities.The goal of the IDRLRWODC technique is to detect and classify waste objects using the DL and DRL techniques.The IDRL-RWODC technique encompasses a twostage process namely Mask Regional Convolutional Neural Network(Mask RCNN)based object detection and DRL based object classification.In addition,DenseNet model is applied as a baseline model for the Mask RCNN model,and a deep Q-learning network(DQLN)is employed as a classifier.Moreover,a dragonfly algorithm(DFA)based hyperparameter optimizer is derived for improving the efficiency of the DenseNet model.In order to ensure the enhanced waste classification performance of the IDRL-RWODC technique,a series of simulations take place on benchmark dataset and the experimental results pointed out the better performance over the recent techniques with maximal accuracy of 0.993.
基金Project (No. 10577017) supported by the National Natural Science Foundation of China
文摘Statistical and contextual information are typically used to detect moving regions in image sequences for a fixed camera.In this paper,we propose a fast and stable linear discriminant approach based on Gaussian Single Model(GSM)and Markov Random Field(MRF).The performance of GSM is analyzed first,and then two main improvements corresponding to the drawbacks of GSM are proposed:the latest filtered data based update scheme of the background model and the linear classification judgment rule based on spatial-temporal feature specified by MRF.Experimental results show that the proposed method runs more rapidly and accurately when compared with other methods.
基金supported by the Ministry of Higher Education Malaysia under Fundamental Research Grant No.0719
文摘Moments have been used in all sorts of object classification systems based on image. There are lots of moments studied by many researchers in the area of object classification and one of the most preference moments is the Zernike moment. In this paper, the performance of object classification using the Zernike moment has been explored. The classifier based on neural networks has been used in this study. The results indicate the best performance in identifying the aggregate is at 91.4% with a ten orders of the Zernike moment. This encouraging result has shown that the Zernike moment is a suitable moment to be used as a feature of object classification systems.
基金Supported by National Natural Science Foundation ot China（60572100, 60673122）, Royal Society （U.K.） International Joint Projects 2006/R3-Cost Share with NSFC （60711130233）, Science Foundation of Shenzhen City （CXQ2008019, 200706）, and Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry （2008[890]）.
