This paper studies the effect of breaking single-class building data into multi-class building data for semantic segmentation under end-to-end architecture such as UNet, UNet++, DeepLabV3, and DeepLabv3+. Although, th...This paper studies the effect of breaking single-class building data into multi-class building data for semantic segmentation under end-to-end architecture such as UNet, UNet++, DeepLabV3, and DeepLabv3+. Although, the already existing semantic segmentation methods for building detection work on the imagery of developed world, where the buildings are highly structured and there is a clearly distinguishable space present between the building instances, the same methods do not work as effectively on the developing world where there is often no clear differentiable spaces between instances of building thus reducing the number of detected instances. Hence as a noble approach, we have added building contours as new class along with building segmentation data, and detected the building contours and the inner building regions, hence giving the precise number of buildings existing in the input imagery especially in the convoluted areas where the boundary between the buildings are often hard to determine even for human eyes. Breaking down the building data into multi-class data increased the building detection precision and recall. This is useful in building detection where building instances are convoluted and are difficult for bare instance segmentation to detect all the instances.展开更多
Modern leather industries are focused on producing high quality leather products for sustaining the market com-petitiveness. However, various leather defects are introduced during various stages of manufacturing proce...Modern leather industries are focused on producing high quality leather products for sustaining the market com-petitiveness. However, various leather defects are introduced during various stages of manufacturing process such as material handling, tanning and dyeing. Manual inspection of leather surfaces is subjective and inconsistent in nature;hence machine vision systems have been widely adopted for the automated inspection of leather defects. It is neces-sary develop suitable image processing algorithms for localize leather defects such as folding marks, growth marks, grain off, loose grain, and pinhole due to the ambiguous texture pattern and tiny nature in the localized regions of the leather. This paper presents deep learning neural network-based approach for automatic localization and classifica-tion of leather defects using a machine vision system. In this work, popular convolutional neural networks are trained using leather images of different leather defects and a class activation mapping technique is followed to locate the region of interest for the class of leather defect. Convolution neural networks such as Google net, Squeeze-net, RestNet are found to provide better accuracy of classification as compared with the state-of-the-art neural network architectures and the results are presented.展开更多
自动安全换道是车辆实现无人驾驶的关键,为精确识别行驶车辆换道状态,保证行车安全,设计了一种基于多分类支持向量机(Multi-class Support Vector Machine,Multiclass SVM)的车辆换道识别模型。从NGSIM数据集中选取美国101公路车辆轨迹...自动安全换道是车辆实现无人驾驶的关键,为精确识别行驶车辆换道状态,保证行车安全,设计了一种基于多分类支持向量机(Multi-class Support Vector Machine,Multiclass SVM)的车辆换道识别模型。从NGSIM数据集中选取美国101公路车辆轨迹数据进行分类处理,并将车辆换道过程划分为车辆跟驰阶段、车辆换道准备阶段和车辆换道执行阶段。采用网格搜索结合粒子群优化算法(Grid Search-PSO)对SVM模型中惩罚参数C和核参数g进行寻优标定,利用多分类支持向量机换道识别模型对样本数据进行训练和测试,模型测试精度达97.68%。研究表明,模型能够很好地识别车辆在换道过程中的行为状态,为车辆换道阶段的研究提供支持。展开更多
Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine...Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine information from several data sources. In the centralized scheme, all information from several data sources is centralized to construct an input space. Then a multi-class Support Vector Machine classifier is trained. In the distributed schemes, the individual data sources are proc-essed separately and modelled by using the multi-class Support Vector Machine. Then new data fusion strategies are proposed to combine the information from the individual multi-class Support Vector Machine models. Our proposed fusion strategies take into account that an Support Vector Machine (SVM) classifier achieves classification by finding the optimal classification hyperplane with maximal margin. The proposed methods are applied for fault diagnosis of a diesel engine. The experimental results showed that almost all the proposed approaches can largely improve the diagnostic accuracy. The robustness of diagnosis is also improved because of the implementation of data fusion strategies. The proposed methods can also be applied in other fields.展开更多
The basic idea of multi-class classification is a disassembly method,which is to decompose a multi-class classification task into several binary classification tasks.In order to improve the accuracy of multi-class cla...The basic idea of multi-class classification is a disassembly method,which is to decompose a multi-class classification task into several binary classification tasks.In order to improve the accuracy of multi-class classification in the case of insufficient samples,this paper proposes a multi-class classification method combining K-means and multi-task relationship learning(MTRL).The method first uses the split method of One vs.Rest to disassemble the multi-class classification task into binary classification tasks.K-means is used to down sample the dataset of each task,which can prevent over-fitting of the model while reducing training costs.Finally,the sampled dataset is applied to the MTRL,and multiple binary classifiers are trained together.With the help of MTRL,this method can utilize the inter-task association to train the model,and achieve the purpose of improving the classification accuracy of each binary classifier.The effectiveness of the proposed approach is demonstrated by experimental results on the Iris dataset,Wine dataset,Multiple Features dataset,Wireless Indoor Localization dataset and Avila dataset.展开更多
Due to e-business' s variety of customers with different navigational patterns and demands, multiclass queuing network is a natural performance model for it. The open multi-class queuing network(QN) models are bas...Due to e-business' s variety of customers with different navigational patterns and demands, multiclass queuing network is a natural performance model for it. The open multi-class queuing network(QN) models are based on the assumption that no service center is saturated as a result of the combined loads of all the classes. Several formulas are used to calculate performance measures, including throughput, residence time, queue length, response time and the average number of requests. The solution technique of closed multi-class QN models is an approximate mean value analysis algorithm (MVA) based on three key equations, because the exact algorithm needs huge time and space requirement. As mixed multi-class QN models, include some open and some closed classes, the open classes should be eliminated to create a closed multi-class QN so that the closed model algorithm can be applied. Some corresponding examples are given to show how to apply the algorithms mentioned in this article. These examples indicate that multi-class QN is a reasonably accurate model of e-business and can be solved efficiently.展开更多
Based on the framework of support vector machines( SVM) using one-against-one( OAO) strategy, a new multi-class kernel method based on directed acyclic graph( DAG) and probabilistic distance is proposed to raise the m...Based on the framework of support vector machines( SVM) using one-against-one( OAO) strategy, a new multi-class kernel method based on directed acyclic graph( DAG) and probabilistic distance is proposed to raise the multi-class classification accuracies. The topology structure of DAG is constructed by rearranging the nodes' sequence in the graph. DAG is equivalent to guided operating SVM on a list,and the classification performance depends on the nodes' sequence in the graph. Jeffries-Matusita distance( JMD) is introduced to estimate the separability of each class,and the implementation list is initialized with all classes organized according to certain sequence in the list. To testify the effectiveness of the proposed method,numerical analysis is conducted on UCI data and hyperspectral data. Meanwhile,comparative studies using standard OAO and DAG classification methods are also conducted and the results illustrate better performance and higher accuracy of the proposed JMD-DAG method.展开更多
The accurate identification and classification of various power quality disturbances are keys to ensuring high-quality electrical energy. In this study, the statistical characteristics of the disturbance signal of wav...The accurate identification and classification of various power quality disturbances are keys to ensuring high-quality electrical energy. In this study, the statistical characteristics of the disturbance signal of wavelet transform coefficients and wavelet transform energy distribution constitute feature vectors. These vectors are then trained and tested using SVM multi-class algorithms. Experimental results demonstrate that the SVM multi-class algorithms, which use the Gaussian radial basis function, exponential radial basis function, and hyperbolic tangent function as basis functions, are suitable methods for power quality disturbance classification.展开更多
This paper proposes a new queuing model and adaptive scheduling scheme which realizes multi-class QoS mechanism under DiffServ architecture. The queuing model is composed of two parallel output subqueues, each output ...This paper proposes a new queuing model and adaptive scheduling scheme which realizes multi-class QoS mechanism under DiffServ architecture. The queuing model is composed of two parallel output subqueues, each output sub-queue adopts random drop algorithm by setting different buffer threshold for different class traffic, so it can provide multi-class QoS. The new proposed scheduling scheme which adaptively changes the parameter A can guarantee the performance target of high class traffic, in the mean time, improve the QoS of low classes traffic.展开更多
The requirement for guaranteed Quality of Service (QoS) have become very essential since there are numerous network base application is available such as video conferencing, data streaming, data transfer and many more...The requirement for guaranteed Quality of Service (QoS) have become very essential since there are numerous network base application is available such as video conferencing, data streaming, data transfer and many more. This has led to the multi-class switch architecture to cater for the needs for different QoS requirements. The introduction of threshold in multi-class switch to solve the starvation problems in loss sensitive class has increased the mean delay for delay sensitive class. In this research, a new scheduling architecture is introduced to improve mean delay in delay sensitive class when the threshold is active. The proposed architecture has been simulated under uniform and non-uniform traffic to show performance of the switch in terms of mean delay. The results show that the proposed architecture has achieved better performance as compared to Weighted Fair Queueing (WFQ) and Priority Queue (PQ).展开更多
A multi-class method is proposed based on Error Correcting Output Codes algorithm in order to get better performance of attack recognition in Wireless Sensor Networks. Aiming to enhance the accuracy of attack detectio...A multi-class method is proposed based on Error Correcting Output Codes algorithm in order to get better performance of attack recognition in Wireless Sensor Networks. Aiming to enhance the accuracy of attack detection, the multi-class method is constructed with Hadamard matrix and two-class Support Vector Machines. In order to minimize the complexity of the algorithm, sparse coding method is applied in this paper. The comprehensive experimental results show that this modified multi-class method has better attack detection rate compared with other three coding algorithms, and its time efficiency is higher than Hadamard coding algorithm.展开更多
The acoustic vibration signal of tank is disassembled into the sum of intrinsic mode function (IMF) by multi-resolution empirical mode decomposition (EMD) method. The instantaneous frequency is obtained, and featu...The acoustic vibration signal of tank is disassembled into the sum of intrinsic mode function (IMF) by multi-resolution empirical mode decomposition (EMD) method. The instantaneous frequency is obtained, and feature transformation matrix is figured out by class scatter matrix. Multi- dimensional scale energy vector is mapped into low-dimensional eigenvector, and classification extraction is realized. This method sufficiently separates of different sound target features. The test result indicates that it is effective.展开更多
数据流分类是数据流挖掘领域一项重要研究任务,目标是从不断变化的海量数据中捕获变化的类结构.目前,几乎没有框架可以同时处理数据流中常见的多类非平衡、概念漂移、异常点和标记样本成本高昂问题.基于此,提出一种非平衡数据流在线主...数据流分类是数据流挖掘领域一项重要研究任务,目标是从不断变化的海量数据中捕获变化的类结构.目前,几乎没有框架可以同时处理数据流中常见的多类非平衡、概念漂移、异常点和标记样本成本高昂问题.基于此,提出一种非平衡数据流在线主动学习方法(Online active learning method for imbalanced data stream,OALM-IDS).AdaBoost是一种将多个弱分类器经过迭代生成强分类器的集成分类方法,AdaBoost.M2引入了弱分类器的置信度,此类方法常用于静态数据.定义了基于非平衡比率和自适应遗忘因子的训练样本重要性度量,从而使AdaBoost.M2方法适用于非平衡数据流,提升了非平衡数据流集成分类器的性能.提出了边际阈值矩阵的自适应调整方法,优化了标签请求策略.将概念漂移程度融入模型构建过程中,定义了基于概念漂移指数的自适应遗忘因子,实现了漂移后的模型重构.在6个人工数据流和4个真实数据流上的对比实验表明,提出的非平衡数据流在线主动学习方法的分类性能优于其他5种非平衡数据流学习方法.展开更多
针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature...针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature,MSF)提取模块及高效的全局上下文信息融合(efficient global contextual information aggregation,EGCA)模块结合U型分割网络进行动静脉分类,抑制了倾向于背景的特征并增强了血管的边缘、交点和末端特征,解决了段内动静脉错误分类问题。此外,在U型网络的解码器部分加入3层深度监督,使浅层信息得到充分训练,避免梯度消失,优化训练过程。在2个公开的眼底图像数据集(DRIVE-AV,LES-AV)上,与3种现有网络进行方法对比,该模型的F1评分分别提高了2.86、1.92、0.81个百分点,灵敏度分别提高了4.27、2.43、1.21个百分点,结果表明所提出的模型能够很好地解决动静脉分类错误的问题。展开更多
文摘This paper studies the effect of breaking single-class building data into multi-class building data for semantic segmentation under end-to-end architecture such as UNet, UNet++, DeepLabV3, and DeepLabv3+. Although, the already existing semantic segmentation methods for building detection work on the imagery of developed world, where the buildings are highly structured and there is a clearly distinguishable space present between the building instances, the same methods do not work as effectively on the developing world where there is often no clear differentiable spaces between instances of building thus reducing the number of detected instances. Hence as a noble approach, we have added building contours as new class along with building segmentation data, and detected the building contours and the inner building regions, hence giving the precise number of buildings existing in the input imagery especially in the convoluted areas where the boundary between the buildings are often hard to determine even for human eyes. Breaking down the building data into multi-class data increased the building detection precision and recall. This is useful in building detection where building instances are convoluted and are difficult for bare instance segmentation to detect all the instances.
文摘Modern leather industries are focused on producing high quality leather products for sustaining the market com-petitiveness. However, various leather defects are introduced during various stages of manufacturing process such as material handling, tanning and dyeing. Manual inspection of leather surfaces is subjective and inconsistent in nature;hence machine vision systems have been widely adopted for the automated inspection of leather defects. It is neces-sary develop suitable image processing algorithms for localize leather defects such as folding marks, growth marks, grain off, loose grain, and pinhole due to the ambiguous texture pattern and tiny nature in the localized regions of the leather. This paper presents deep learning neural network-based approach for automatic localization and classifica-tion of leather defects using a machine vision system. In this work, popular convolutional neural networks are trained using leather images of different leather defects and a class activation mapping technique is followed to locate the region of interest for the class of leather defect. Convolution neural networks such as Google net, Squeeze-net, RestNet are found to provide better accuracy of classification as compared with the state-of-the-art neural network architectures and the results are presented.
文摘Multi-source multi-class classification methods based on multi-class Support Vector Machines and data fusion strategies are proposed in this paper. The centralized and distributed fusion schemes are applied to combine information from several data sources. In the centralized scheme, all information from several data sources is centralized to construct an input space. Then a multi-class Support Vector Machine classifier is trained. In the distributed schemes, the individual data sources are proc-essed separately and modelled by using the multi-class Support Vector Machine. Then new data fusion strategies are proposed to combine the information from the individual multi-class Support Vector Machine models. Our proposed fusion strategies take into account that an Support Vector Machine (SVM) classifier achieves classification by finding the optimal classification hyperplane with maximal margin. The proposed methods are applied for fault diagnosis of a diesel engine. The experimental results showed that almost all the proposed approaches can largely improve the diagnostic accuracy. The robustness of diagnosis is also improved because of the implementation of data fusion strategies. The proposed methods can also be applied in other fields.
基金supported by the National Natural Science Foundation of China(61703131 61703129+1 种基金 61701148 61703128)
文摘The basic idea of multi-class classification is a disassembly method,which is to decompose a multi-class classification task into several binary classification tasks.In order to improve the accuracy of multi-class classification in the case of insufficient samples,this paper proposes a multi-class classification method combining K-means and multi-task relationship learning(MTRL).The method first uses the split method of One vs.Rest to disassemble the multi-class classification task into binary classification tasks.K-means is used to down sample the dataset of each task,which can prevent over-fitting of the model while reducing training costs.Finally,the sampled dataset is applied to the MTRL,and multiple binary classifiers are trained together.With the help of MTRL,this method can utilize the inter-task association to train the model,and achieve the purpose of improving the classification accuracy of each binary classifier.The effectiveness of the proposed approach is demonstrated by experimental results on the Iris dataset,Wine dataset,Multiple Features dataset,Wireless Indoor Localization dataset and Avila dataset.
文摘Due to e-business' s variety of customers with different navigational patterns and demands, multiclass queuing network is a natural performance model for it. The open multi-class queuing network(QN) models are based on the assumption that no service center is saturated as a result of the combined loads of all the classes. Several formulas are used to calculate performance measures, including throughput, residence time, queue length, response time and the average number of requests. The solution technique of closed multi-class QN models is an approximate mean value analysis algorithm (MVA) based on three key equations, because the exact algorithm needs huge time and space requirement. As mixed multi-class QN models, include some open and some closed classes, the open classes should be eliminated to create a closed multi-class QN so that the closed model algorithm can be applied. Some corresponding examples are given to show how to apply the algorithms mentioned in this article. These examples indicate that multi-class QN is a reasonably accurate model of e-business and can be solved efficiently.
基金Sponsored by the National Natural Science Foundation of China(Grant No.61201310)the Fundamental Research Funds for the Central Universities(Grant No.HIT.NSRIF.201160)the China Postdoctoral Science Foundation(Grant No.20110491067)
文摘Based on the framework of support vector machines( SVM) using one-against-one( OAO) strategy, a new multi-class kernel method based on directed acyclic graph( DAG) and probabilistic distance is proposed to raise the multi-class classification accuracies. The topology structure of DAG is constructed by rearranging the nodes' sequence in the graph. DAG is equivalent to guided operating SVM on a list,and the classification performance depends on the nodes' sequence in the graph. Jeffries-Matusita distance( JMD) is introduced to estimate the separability of each class,and the implementation list is initialized with all classes organized according to certain sequence in the list. To testify the effectiveness of the proposed method,numerical analysis is conducted on UCI data and hyperspectral data. Meanwhile,comparative studies using standard OAO and DAG classification methods are also conducted and the results illustrate better performance and higher accuracy of the proposed JMD-DAG method.
文摘The accurate identification and classification of various power quality disturbances are keys to ensuring high-quality electrical energy. In this study, the statistical characteristics of the disturbance signal of wavelet transform coefficients and wavelet transform energy distribution constitute feature vectors. These vectors are then trained and tested using SVM multi-class algorithms. Experimental results demonstrate that the SVM multi-class algorithms, which use the Gaussian radial basis function, exponential radial basis function, and hyperbolic tangent function as basis functions, are suitable methods for power quality disturbance classification.
文摘This paper proposes a new queuing model and adaptive scheduling scheme which realizes multi-class QoS mechanism under DiffServ architecture. The queuing model is composed of two parallel output subqueues, each output sub-queue adopts random drop algorithm by setting different buffer threshold for different class traffic, so it can provide multi-class QoS. The new proposed scheduling scheme which adaptively changes the parameter A can guarantee the performance target of high class traffic, in the mean time, improve the QoS of low classes traffic.
文摘The requirement for guaranteed Quality of Service (QoS) have become very essential since there are numerous network base application is available such as video conferencing, data streaming, data transfer and many more. This has led to the multi-class switch architecture to cater for the needs for different QoS requirements. The introduction of threshold in multi-class switch to solve the starvation problems in loss sensitive class has increased the mean delay for delay sensitive class. In this research, a new scheduling architecture is introduced to improve mean delay in delay sensitive class when the threshold is active. The proposed architecture has been simulated under uniform and non-uniform traffic to show performance of the switch in terms of mean delay. The results show that the proposed architecture has achieved better performance as compared to Weighted Fair Queueing (WFQ) and Priority Queue (PQ).
文摘A multi-class method is proposed based on Error Correcting Output Codes algorithm in order to get better performance of attack recognition in Wireless Sensor Networks. Aiming to enhance the accuracy of attack detection, the multi-class method is constructed with Hadamard matrix and two-class Support Vector Machines. In order to minimize the complexity of the algorithm, sparse coding method is applied in this paper. The comprehensive experimental results show that this modified multi-class method has better attack detection rate compared with other three coding algorithms, and its time efficiency is higher than Hadamard coding algorithm.
文摘The acoustic vibration signal of tank is disassembled into the sum of intrinsic mode function (IMF) by multi-resolution empirical mode decomposition (EMD) method. The instantaneous frequency is obtained, and feature transformation matrix is figured out by class scatter matrix. Multi- dimensional scale energy vector is mapped into low-dimensional eigenvector, and classification extraction is realized. This method sufficiently separates of different sound target features. The test result indicates that it is effective.
文摘数据流分类是数据流挖掘领域一项重要研究任务,目标是从不断变化的海量数据中捕获变化的类结构.目前,几乎没有框架可以同时处理数据流中常见的多类非平衡、概念漂移、异常点和标记样本成本高昂问题.基于此,提出一种非平衡数据流在线主动学习方法(Online active learning method for imbalanced data stream,OALM-IDS).AdaBoost是一种将多个弱分类器经过迭代生成强分类器的集成分类方法,AdaBoost.M2引入了弱分类器的置信度,此类方法常用于静态数据.定义了基于非平衡比率和自适应遗忘因子的训练样本重要性度量,从而使AdaBoost.M2方法适用于非平衡数据流,提升了非平衡数据流集成分类器的性能.提出了边际阈值矩阵的自适应调整方法,优化了标签请求策略.将概念漂移程度融入模型构建过程中,定义了基于概念漂移指数的自适应遗忘因子,实现了漂移后的模型重构.在6个人工数据流和4个真实数据流上的对比实验表明,提出的非平衡数据流在线主动学习方法的分类性能优于其他5种非平衡数据流学习方法.
文摘针对由于血管类间具有强相似性造成的动静脉错误分类问题,提出了一种新的融合上下文信息的多尺度视网膜动静脉分类网络(multi-scale retinal artery and vein classification network,MCFNet),该网络使用多尺度特征(multi-scale feature,MSF)提取模块及高效的全局上下文信息融合(efficient global contextual information aggregation,EGCA)模块结合U型分割网络进行动静脉分类,抑制了倾向于背景的特征并增强了血管的边缘、交点和末端特征,解决了段内动静脉错误分类问题。此外,在U型网络的解码器部分加入3层深度监督,使浅层信息得到充分训练,避免梯度消失,优化训练过程。在2个公开的眼底图像数据集(DRIVE-AV,LES-AV)上,与3种现有网络进行方法对比,该模型的F1评分分别提高了2.86、1.92、0.81个百分点,灵敏度分别提高了4.27、2.43、1.21个百分点,结果表明所提出的模型能够很好地解决动静脉分类错误的问题。