检索结果-南通市图书馆

Domain-specific feature elimination:multi-source domain adaptation for image classification

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2023年第4期17卷 171-179页

作者： Kunhong WU Fan JIA Yahong HAN College of Intelligence and Computing Tianjin UniversityTianjin 300350China Tianjin Key lab of machine learning Tianjin UniversityTianjin 300350China

Multi-source domain adaptation utilizes multiple source domains to learn the knowledge and transfers it to an unlabeled target *** address the problem,most of the existing methods aim to minimize the domain shift by auxiliary distribution alignment objectives,which reduces the effect of domain-specific ***,without explicitly modeling the domain-specific features,it is not easy to guarantee that the domain-invariant representation extracted from input domains contains domain-specific information as few as *** this work,we present a different perspective on MSDA,which employs the idea of feature elimination to reduce the influence of domain-specific *** design two different ways to extract domain-specific features and total features and construct the domain-invariant representations by eliminating the domain-specific features from total *** experimental results on different domain adaptation datasets demonstrate the effectiveness of our method and the generalization ability of our model.

关键词： multi-source domain adaptation generalization feature elimination

Optimized high order product quantization for approximate nearest neighbors search

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2020年第2期14卷 259-272页

作者： Linhao LI Qinghua HU School of Artificial Intelligence Hebei University of TechnologyTianjin 300401China Tianjin Key lab of machine learning School of Computer Science and TechnologyTianjin UniversityTianjin 300072China

Product quantization is now considered as an effective approach to solve the approximate nearest neighbor(ANN)search.A collection of derivative algorithms have been ***,the current techniques ignore the intrinsic high order structures of data,which usually contain helpful information for improving the computational *** this paper,aiming at the complex structure of high order data,we design an optimized technique,called optimized high order product quantization(O-HOPQ)for ANN *** O-HOPQ,we incorporate the high order structures of the data into the process of designing a more effective subspace decomposition *** a result,spatial adjacent elements in the high order data space are grouped into the same ***,O-HOPQ generates its spatial structured codebook,by optimizing the quantization *** from the structured codebook,the global optimum quantizers can be obtained effectively and *** results show that appropriate utilization of the potential information that exists in the complex structure of high order data will result in significant improvements to the performance of the product ***,the high order structure based approaches are effective to the scenario where the data have intrinsic complex structures.

关键词： product quantization high order structured data tensor theory approximate nearest neighbor search

Insect classification and detection in field crops using modern machine learning techniques

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Information Processing in Agriculture 2021年第3期8卷 446-457页

作者： Thenmozhi Kasinathan Dakshayani Singaraju Srinivasulu Reddy Uyyala machine learning and Data Analytics lab Centre of Excellence in Artificial IntelligenceDepartment of Computer ApplicationsNational Institute of TechnologyTiruchirappalliTiruchirappalli 620015Tamil NaduIndia

The agriculture sector has an immense potential to improve the requirement of food and supplies healthy and nutritious *** insect detection is a challenging task for farmers as a significant portion of the crops are damaged,and the quality is degraded due to the pest *** insect identification has the drawback of requiring well-trained tax-onomists to identify insects based on morphological features *** were conducted for classification on nine and 24 insect classes of Wang and Xie dataset using the shape features and applying machine learning techniques such as artificial neural net-works(ANN),support vector machine(SVM),k-nearest neighbors(KNN),naive bayes(NB)and convolutional neural network(CNN)*** paper presents the insect pest detec-tion algorithm that consists of foreground extraction and contour identification to detect the insects for Wang,Xie,Deng,and IP102 datasets in a highly complex *** 9-fold cross-validation was applied to improve the performance of the classification *** highest classification rate of 91.5%and 90%was achieved for nine and 24 class insects using the CNN *** detection performance was accomplished with less com-putation time for Wang,Xie,Deng,and IP102 datasets using insect pest detection *** comparison results with the state-of-the-art classification algorithms exhibited considerable improvement in classification accuracy,computation time perfor-mance while apply more efficiently in field crops to recognize the *** results of classification accuracy are used to recognize the crop insects in the early stages and reduce the time to enhance the crop yield and crop quality in agriculture.

关键词： Crop pest classification Crop insect detection Image processing machine learning Image segmentation

Efficient image representation for object recognition via pivots selection

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2015年第3期9卷 383-391页

作者： Bojun XIE Yi LIU HuiZHANG Jian YU Beijing Key lab of Traffic Data Analysis and Mining School of Computer and Information Technology Beijing Jiaotong University Beijing 100044 China Key lab of machine learning and Computational Intelligence College of Mathematics and Computer Science Hebei University Baoding 071000 China

Patch-level features are essential for achieving good performance in computer vision tasks. Besides well- known pre-defined patch-level descriptors such as scalein- variant feature transform （SIFT） and histogram of oriented gradient （HOG）, the kernel descriptor （KD） method [1] of- fers a new way to ＂grow-up＂ features from a match-kernel defined over image patch pairs using kernel principal compo- nent analysis （KPCA） and yields impressive results. In this paper, we present efficient kernel descriptor （EKD） and efficient hierarchical kernel descriptor （EHKD）, which are built upon incomplete Cholesky decomposition. EKD au- tomatically selects a small number of pivot features for gener- ating patch-level features to achieve better computational effi- ciency. EHKD recursively applies EKD to form image-level features layer-by-layer. Perhaps due to parsimony, we find surprisingly that the EKD and EHKD approaches achieved competitive results on several public datasets compared with other state-of-the-art methods, at an improved efficiency over KD.

关键词： efficient kernel descriptor efficient hierarchi-cal kernel descriptor incomplete Cholesky decomposition,patch-level features image-level features

Deep learning enabled localization for UAV autolanding

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Aeronautics 2021年第5期34卷 585-600页

作者： Minghui LI Tianjiang HU machine Intelligence and Collective RObotics(MICRO)lab Sun Yet-sen UniversityGuangzhou 510275China

This article concentrates on ground vision guided autonomous landing of a fixed-wing Unmanned Aerial Vehicle(UAV)within Global Navigation Satellite System(GNSS)denied *** deep learning models are developed and employed into image detection and its accuracy promoting for UAV autolanding,***,we design a target bounding box detection network Bbox Locate-Net to extract its image coordinate of the flying ***,the detected coordinate is fused into spatial localization with an extended Kalman filter ***,a point regression network Point Refine-Net is developed for promoting detection accuracy once the flying vehicle’s motion continuity is checked *** proposed approach definitely accomplishes the closed-loop mutual inspection of spatial positioning and image detection,and automatically improves the inaccurate coordinates within a certain *** results demonstrate and verify that our method outperforms the previous works in terms of accuracy,robustness and real-time ***,the newly developed Bbox Locate-Net attaches over 500 fps,almost five times the published state-of-the-art in this field,with comparable localization accuracy.

关键词： Deep learning Localization Safe landing Stereo vision UAV autolanding

Dual collaboration for decentralized multi-source domain adaptation

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2022年第12期23卷 1780-1794页

作者： Yikang WEI Yahong HAN College of Intelligence and Computing Tianjin UniversityTianjin 300350China Tianjin Key lab of machine learning Tianjin UniversityTianjin 300350China

The goal of decentralized multi-source domain adaptation is to conduct unsupervised multi-source domain adaptation in a data decentralization scenario. The challenge of data decentralization is that the source domains and target domain lack cross-domain collaboration during training. On the unlabeled target domain, the target model needs to transfer supervision knowledge with the collaboration of source models, while the domain gap will lead to limited adaptation performance from source models. On the labeled source domain, the source model tends to overfit its domain data in the data decentralization scenario, which leads to the negative transfer problem. For these challenges, we propose dual collaboration for decentralized multi-source domain adaptation by training and aggregating the local source models and local target model in collaboration with each other. On the target domain, we train the local target model by distilling supervision knowledge and fully using the unlabeled target domain data to alleviate the domain shift problem with the collaboration of local source models. On the source domain, we regularize the local source models in collaboration with the local target model to overcome the negative transfer problem. This forms a dual collaboration between the decentralized source domains and target domain, which improves the domain adaptation performance under the data decentralization scenario. Extensive experiments indicate that our method outperforms the state-of-the-art methods by a large margin on standard multi-source domain adaptation datasets.

关键词： Multi-source domain adaptation Data decentralization Domain shift Negative transfer

Hybrid Bayesian estimation tree learning with discrete and fuzzy labels

在线全文

学校读者我要写书评

暂无评论

Frontiers of Computer Science 2013年第6期7卷 852-863页

作者： Zengchang QIN Tao WAN Intelligent Computing and machine learning lab School of Automation Science and Electrical Engineering Beihang University Beijing 100191 China School of Biological Science and Medical Engineering Beihang University Beijing 100191 China Department of Biomedical Engineering Case Western Reserve University Cleveland OH 44106 USA

Classical decision tree model is one of the classical machine learning models for its simplicity and effectiveness in applications. However, compared to the DT model, probability estimation trees （PETs） give a better estimation on class probability. In order to get a good probability estimation, we usually need large trees which are not desirable with respect to model transparency. Linguistic decision tree （LDT） is a PET model based on label semantics. Fuzzy labels are used for building the tree and each branch is associated with a probability distribution over classes. If there is no overlap between neighboring fuzzy labels, these fuzzy labels then become discrete labels and a LDT with discrete labels becomes a special case of the PET model. In this paper, two hybrid models by combining the naive Bayes classifier and PETs are proposed in order to build a model with good performance without losing too much transparency. The first model uses naive Bayes estimation given a PET, and the second model uses a set of small-sized PETs as estimators by assuming the independence between these trees. Empirical studies on discrete and fuzzy labels show that the first model outperforms the PET model at shallow depth, and the second model is equivalent to the naive Bayes and PET.

关键词： fuzzy labels label semantics random set probability estimation tree mass assignment linguistic decision tree naive Bayes

Dynamic parameterized learning for unsupervised domain adaptation

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2023年第11期24卷 1616-1632页

作者： Runhua JIANG Yahong HAN College of Intelligence and Computing Tianjin UniversityTianjin 300350China Tianjin Key lab of machine learning Tianjin UniversityTianjin 300350China

Unsupervised domain adaptation enables neural networks to transfer from a labeled source domain to an unlabeled target domain by learning domain-invariant *** approaches achieve this by directly matching the marginal distributions of these two *** of them,however,ignore exploration of the dynamic trade-off between domain alignment and semantic discrimination learning,thus rendering them susceptible to the problems of negative transfer and outlier *** address these issues,we introduce the dynamic parameterized learning ***,by exploring domain-level semantic knowledge,the dynamic alignment parameter is proposed,to adaptively adjust the optimization steps of domain alignment and semantic discrimination ***,for obtaining semantic-discriminative and domain-invariant representations,we propose to align training trajectories on both source and target *** experiments are conducted to validate the effectiveness of the proposed methods,and extensive comparisons are conducted on seven datasets of three visual tasks to demonstrate their practicability.

关键词： Unsupervised domain adaptation Optimization steps Domain alignment Semantic discrimination

6G-Enabled Internet of Things:Vision,Techniques,and Open Issues

在线全文

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2022年第12期133卷 509-556页

作者： Mehdi Hosseinzadeh Atefeh Hemmati Amir Masoud Rahmani Pattern Recognition and machine learning lab Gachon UniversitySeongnam13120Korea Department of Computer Engineering Science and Research BranchIslamic Azad UniversityTehran***Iran Future Technology Research Center National Yunlin University of Science and TechnologyYunlin64002Taiwan

There are changes in the development of wireless technology systems every decade.6G(sixth generation)wireless networks improve on previous generations by increasing dependability,accelerating networks,increasing available bandwidth,decreasing latency,and increasing data transmission speed to standardize communication *** purpose of this article is to comprehend the current directions in 6G studies and their relationship to the Internet of Things(IoT).Also,this paper discusses the impacts of 6G on IoT,critical requirements and trends for 6G-enabled IoT,new service classes of 6G and IoT technologies,and current 6G-enabled IoT studies selected by the systematic literature review(SLR)method published from 2018 to *** addition,we present a technical taxonomy for the classification of 6G-enabled IoT,which includes self-organization systems,energy efficiency,channel assessment,and ***,according to the articles reviewed,we consider the evaluation factors in this domain,including data transmission,delay,energy consumption,and ***,we focus on open issues and future research challenges in 6G-enabled *** mention important future challenges and directions,we can point to migration,data storage,data resource,data security,data sharing,data offloading,availability,scalability,portability,user experience,reliability,authentication,and authorization.

关键词： Internet of Things 6G wireless network machine learning artificial intelligence