检索结果-内蒙古大学图书馆

2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016

作者： Lei, Chenyi Liu, Dong Li, Weiping Zha, Zheng-Jun Li, Houqiang CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei230027 China

ISBN: (纸本)9781467388511

In many image-related tasks, learning expressive and discriminative representations of images is essential, and deep learning has been studied for automating the learning of such representations. Some user-centric tasks, such as image recommendations, call for effective representations of not only images but also preferences and intents of users over images. Such representations are termed hybrid and addressed via a deep learning approach in this paper. We design a dual-net deep network, in which the two subnetworks map input images and preferences of users into a same latent semantic space, and then the distances between images and users in the latent space are calculated to make decisions. We further propose a comparative deep learning (CDL) method to train the deep network, using a pair of images compared against one user to learn the pattern of their relative distances. The CDL embraces much more training data than naive deep learning, and thus achieves superior performance than the latter, with no cost of increasing network complexity. Experimental results with real-world data sets for image recommendations have shown the proposed dual-net network and CDL greatly outperform other stateof-the-art image recommendation solutions.

关键词： Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

High-Resolution Urban Flows Forecasting With Coarse-Grained Spatiotemporal Data

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第2期4卷 315-327页

作者： Xu, Zhenyi Kang, Yu Cao, Yang Hefei Comprehensive National Science Center Institute of Artificial Intelligence Hefei230088 China Department of Automation University of Science and Technology of China Hefei230026 China Chinese Academy of Sciences Key Laboratory of Technology in GeoSpatial Information Processing and Application System Beijing100192 China

Urban flow monitoring and forecasting systems play important roles in smart city management. However, due to the long-lasting and enormous deployment cost of ubiquitous traffic monitoring devices (e.g., loop detectors, traffic video detection), it is very difficult to predict flow in high-resolution (HR) with limited monitoring devices. The existing spatiotemporal network based methods usually predict the urban flow in the same spatial scale without considering the spatial correlations between coarse-grained and fine-grained urban flows. To tackle these issues, we propose a HR spatiotemporal transformer network (HRSTT) to predict fine-grained urban flow. Specifically, residual convolution units are employed to constructs the high-level features of three-view temporal (e.g., closeness, period, and trend) flow data. Then, a transformer block is designed to jointly learn the spatiotemporal dynamic features of each temporal flow with self-attention mechanism. For the external factors (e.g., holidays, weather conditions) are extracted by embedding dense networks, which are fused with high level coarse-grained flow feature maps with gated-fusion scheme. Finally, the coarse-grained fusion feature maps are transferred to the distributional upsampling module, generating the fined-grained flow map of target predicted time. Furthermore, the proposed model is evaluated with several baselines on real-world TaxiBJ datasets, demonstrating the state-of-the-art performance of our approach on the fine-grained urban flow forecasting problem. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Single Channel CSAR Moving Target Detection Algorithm Based on Low-rank Sparse Decomposition

Single Channel CSAR Moving Target Detection Algorithm Based ...

引用

2021 CIE International Conference on Radar, Radar 2021

作者： Zhang, Zhiguo Shen, Wenjie Yue, Xiaoyang Lin, Yun Hong, Wen Aerospace Information Research Institute Chinese Academy of Sciences Key Laboratory of Technology in Geospatial Information Processing and Application System Beijing China School of Information Science and Technology North China University of Technology Radar Monitoring Technology Laboratory Beijing China

ISBN: (纸本)9781665498142

With the goal of detecting moving targets, this paper proposes a new single channel Circular Synthetic Aperture Radar (CSAR) moving targets detection algorithm based on Low-rank Sparse Decomposition (LRSD). This algorithm utilizes the correlation among overlap subaperture logarithmic amplitude image sequences, the static clutter is regarded as low-rank component and the moving targets is considered as sparse component. Then, the background image sequence (without moving target) and the foreground image sequence (moving targets) can be extracted by the LRSD. Considering the unknown foreground image distribution, this paper proposes a Local Dynamic Threshold based on Otsu (LDTO) method independent of image distribution to detect moving targets. Finally, the experiment on the X-band airborne Gotcha Data demonstrates the effectiveness of the proposed algorithm. © 2021 IEEE.

关键词： circular synthetic aperture radar local dynamic threshold based on Otsu low rank sparse decomposition moving target detection

来源：评论

学校读者我要写书评

暂无评论

AZIMUTH AMBIGUITIES SUPPRESSION FOR MULTICHANNEL SAR IMAGING BASED ON L2,q REGULARIZATION: INITIAL RESULTS OF NON-SPARSE SCENARIO

AZIMUTH AMBIGUITIES SUPPRESSION FOR MULTICHANNEL SAR IMAGING...

引用

2021 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2021

作者： Liu, Mingqian Li, Jie Zhang, Zhe Zhang, Bingchen Wu, Yirong Aerospace Information Research Institute Chinese Academy of Sciences China Key Laboratory of Technology in Geospatial Information Processing and Application System China University of Chinese Academy of Sciences Beijing100190 China

The azimuth multichannel SAR is competent to achieve high-resolution and wide-swath (HRWS) imaging. For some spaceborne multichannel SAR systems, the pulse repetition frequency (PRF) of each channel at some beam positions is less than that of uniform sampling, hence leading to the azimuth ambiguities in the recovered images. In this paper, a novel azimuth ambiguity suppression method for multichannel SAR imaging based on L2,q regularization (0 2,q regularization from single-channel SAR system to multichannel SAR system and develop the proposed method. Finally, we demonstrate the effectiveness of the proposed method for non-sparse scenarios. Simulations and Gaofen-3 real data experiments are carried out to verify the validity of proposed method. ©2021 IEEE

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

AF2S: An Anchor-Free Two-Stage Tracker Based on a Strong SiamFC Baseline 16th

AF2S: An Anchor-Free Two-Stage Tracker Based on a Strong Sia...

引用

Workshops held at the 16th European Conference on Computer Vision, ECCV 2020

作者： He, Anfeng Wang, Guangting Luo, Chong Tian, Xinmei Zeng, Wenjun CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei Anhui China Microsoft Research Asia Beijing China

ISBN: (纸本)9783030682378

Siamese network based trackers have become a mainstream in visual object tracking. Recently, several high-performance multi-stage trackers have been proposed and some of them adopt SiamRPN for the first-stage region proposal. We argue that an anchor-based region proposal network is not necessary for the tracking task, as a tracker has a strong prior about the location and size of the target. In this paper, we propose a two-stage visual tracker which uses SiamFC for region proposal. SiamFC defines a bounding box by its center, which is a typical anchor-free (AF) network, so we dub our tracker AF2S. As the model size of SiamFC is only about 1/10 that of SiamRPN, AF2S results in a significantly lighter model than its SiamRPN-based counterparts. In the design of AF2S, we first build a strong AlexNet-based SiamFC baseline which improves the AUC on OTB-100 from 0.582 to 0.665. Further, we propose a position-sensitive convolutional layer which can be stacked after SiamFC backbone to increase the robustness of proposals without losing localization precision. Finally, a relation network is used for box refinement. Experimental results show that AF2S achieves the best performance on OTB-100 and VOT-18 among the state-of-the-art trackers which use AlexNet as backbone. On LaSOT-test, AF2S achieves an AUC of 0.480, which is among the first-tier performance even when trackers with more powerful backbone and much larger model size are considered. © 2020, Springer Nature Switzerland AG.

关键词： Tracking (position)

来源：评论

学校读者我要写书评

暂无评论

AUTOMATIC MODULATION RECOGNITION BASED ON SST AND DEEP LEARNING MODEL

AUTOMATIC MODULATION RECOGNITION BASED ON SST AND DEEP LEARN...

引用

IET International Radar Conference 2023, IRC 2023

作者： Xiang, Jianbing Lv, Xiaolei Han, Bing Zhong, Lihua Fang, Linlin Aerospace Information Research Institute Chinese Academy of Sciences Beijing China Key Laboratory of Technology in GeoSpatial Information Processing and Application System Aerospace Information Research Institute Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781839539954

As the electromagnetic environment in battlefields becomes increasingly complex, automatic modulation recognition for noncooperation radiation source signal is becoming vital and challenging. Most previous works were more likely to cause low recognition accuracy with high computational complexity in low signal-to-noise ratio (SNR), and only considered spatial or temporal attributes of signal which ignores the feature interaction. In this paper, we firstly apply synchrosqueezing transform (SST) to extract high resolution and robustness time-frequency feature as one of signal representation, and convert the signal into the in-phase/quadrature (I/Q) components as another signal representation, which enhances the diversity of signal representation. Then, we propose a deep learning model to achieve automatic recognition, which has two streams and each stream consists of the convolutional neural network (CNN), bistatic long short-term memory (BiLSTM) and temporal attention (TAT). It efficiently explores the feature from each signal representation and enhances the key features by the introduction of TAT, furthermore, it implements feature interaction to form a new feature matrix. Finally, experiments and analysis were conducted using the RadioML2016.10a dataset. The experimental results showed that compared to other algorithms, this method has higher robustness and perform better recognition accuracy under low SNR. © The Institution of Engineering & technology 2023.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

ERROR ESTIMATION and UNAMBIGUOUS RECONSTRUCTION for Chinese FIRST DUAL-CHANNEL SPACEBORNE SAR IMAGING

ERROR ESTIMATION and UNAMBIGUOUS RECONSTRUCTION for Chinese ...

引用

ISPRS geospatial Week 2017

作者： Jin, T. Qiu, X. Hu, D. Ding, C. University of Chinese Academy of Sciences Beijing100049 China Key Laboratory of Technology in Geospatial Information Processing and Application System Institute of Electronics Chinese Academy of Sciences Beijing100190 China

Multichannel synthetic aperture radar (SAR) is a significant breakthrough to the inherent limitation between high-resolution and wide-swath (HRWS) faced with conventional SAR. Error estimation and unambiguous reconstruction are two crucial techniques for obtaining high-quality imagery. This paper demonstrates the experimental results of the two techniques for Chinese first dualchannel spaceborne SAR imaging. The model of Chinese Gaofen-3 dual-channel mode is established and the mechanism of channel mismatches is first discussed. Particularly, we propose a digital beamforming (DBF) process composed of the subspace-based error estimation algorithm and the reconstruction algorithm before imaging. The results exhibit the effective suppression of azimuth ambiguities with the proposed DBF process, and indicate the feasibility of this technique for future HRWS SAR systems. © Authors 2017.

关键词： digital beamforming (DBF) high-resolution and wide-swath (HRWS) orthogonal subspace method (OSM) Synthetic aperture radar (SAR) unambiguous reconstruction

来源：评论

学校读者我要写书评

暂无评论

Unsupervised classification of polarimetric circular SAR images based on H/α/Multi-aperture polarimetric entropy 12

Unsupervised classification of polarimetric circular SAR ima...

引用

12th European Conference on Synthetic Aperture Radar, EUSAR 2018

作者： Hong, Wen Xue, Feiteng Lin, Yun Chen, Shiqiang Shen, Wenjie University of Chinese Academy of Sciences Beijing China Institute of Electronics Chinese Academy of Sciences Beijing China Key Laboratory of Technology in Geospatial Information Processing and Application System Beijing China

ISBN: (纸本)9783800746361

In conventional polarimetric synthetic aperture radar(PolSAR), targets are usually assumed isotropic and potential polarimetric variations across azimuth are unconsidered. As to circular SAR (CSAR), the azimuthal view angle is much larger. Polarimetric variations across azimuth are no longer negligible. H/α unsupervised classification method classifies data by polarimetric properties. H/α method can not distinguish anisotropic and isotropic targets which have same polarimetric properties when applied to polarimetric CSAR(Pol-CSAR) data. Recently multi-aperture polarimetric entropy(MAPE) was proposed in Ref. [10]. MAPE can be used to identify anisotropic and isotropic targets. In this paper, an unsupervised classification method based on H/α/MAPE is proposed. The proposed classification method classifies data with H/α/MAPE space. Anisotropic and isotropic targets which have same polarimetric properties are well classified into two classes by H/α/MAPE method. The effectiveness and advantages of H/α/MAPE method are demonstrated on Pol-CSAR data, acquired from the Institute of Electronics airborne CSAR system at P-band. © VDE VERLAG GMBH Â Berlin Â Offenbach.

关键词： Synthetic aperture radar

来源：评论

学校读者我要写书评

暂无评论

Deep supervised qantization by self-organizing map 25

Deep supervised qantization by self-organizing map

引用

25th ACM International Conference on Multimedia, MM 2017

作者： Wang, Min Zhou, Wengang Tian, Qi Pu, Junfu Li, Houqiang CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China China Computer Science Department University of Texas San Antonio United States

ISBN: (纸本)9781450349062

Approximate Nearest Neighbour (ANN) search is an important research topic in multimedia and computer vision fields. In this paper, we propose a new deep supervised quantization method by Self-Organizing Map (SOM) to address this problem. Our method integrates the Convolutional Neural Networks (CNN) and Self-Organizing Map into a unified deep architecture. The overall training objective includes supervised quantization loss and classification loss. With the supervised quantization loss, we minimize the differences on the maps between similar image pairs, and maximize the differences on the maps between dissimilar image pairs. By optimization, the deep architecture can simultaneously extract deep features and quantize the features into the suitable nodes in the Self-Organizing Map. The experiments on several public standard datasets prove the superiority of our approach over the existing ANN search methods. Besides, as a byproduct, our deep architecture can be directly applied to classification task and visualization with little modification, and promising performances are demonstrated on these tasks in the experiments. © 2017 ACM.

关键词： Self organizing maps

来源：评论

学校读者我要写书评

暂无评论

Linear distance preserving pseudo-supervised and unsupervised hashing 16

Linear distance preserving pseudo-supervised and unsupervise...

引用

24th ACM Multimedia Conference, MM 2016

作者： Wang, Min Zhou, Wengang Tian, Qi Zha, Zhengjun Li, Houqiang CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China China Computer Science Department University of Texas at San Antonio United States

ISBN: (纸本)9781450336031

With the advantage in compact representation and efficient comparison, binary hashing has been extensively investigated for approximate nearest neighbor search. In this paper, we propose a novel and general hashing framework, which simultaneously considers a new linear pair-wise distance preserving objective and point-wise constraint. The direct distance preserving objective aims to keep the linear relationships between the Euclidean distance and the Hamming distance of data points. Based on different pointwise constraints, we propose two methods to instantiate this framework. The first one is a pseudo-supervised hashing method, which uses existing unsupervised hashing methods to generate binary codes as pseudo-supervised information. The second one is an unsupervised hashing method, in which quantization loss is considered. We validate our framework on two large-scale datasets. The experiments demonstrate that our pseudo-supervised method achieves consistent improvement for the state-of-the-art unsupervised hashing methods, while our unsupervised method outperforms the state-of-the-art methods. ©2016 ACM.

关键词： Hamming distance

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：