检索结果-内蒙古大学图书馆

Real-time correlation tracking via joint model compression and transfer

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Wang, Ning Zhou, Wengang Song, Yibing Ma, Chao Li, Houqiang CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System Department of Electronic Engineering and Information Science University of Science and Technology of China Hefei China Tencent AI Lab Shenzhen China MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai China

Correlation filters (CF) have received considerable attention in visual tracking because of their computational efficiency. Leveraging deep features via off-the-shelf CNN models (e.g., VGG), CF trackers achieve state-of-the-art performance while consuming a large number of computing resources. This limits deep CF trackers to be deployed to many mobile platforms on which only a single-core CPU is available. In this paper, we propose to jointly compress and transfer off-the-shelf CNN models within a knowledge distillation framework. We formulate a CNN model pretrained from the image classification task as a teacher network, and distill this teacher network into a lightweight student network as the feature extractor to speed up CF trackers. In the distillation process, we propose a fidelity loss to enable the student network to maintain the representation capability of the teacher network. Meanwhile, we design a tracking loss to adapt the objective of the student network from object recognition to visual tracking. The distillation process is performed offline on multiple layers and adaptively updates the student network using a background-aware online learning scheme. Extensive experiments on five challenging datasets demonstrate that the lightweight student network accelerates the speed of state-of-the-art deep CF trackers to real-time on a single-core CPU while maintaining almost the same tracking accuracy. Copyright © 2019, The Authors. All rights reserved.

关键词： Object recognition

Electromagnetic Simulation Aided SAR Target Classification Via Deep Domain Adaptation

学校读者我要写书评

暂无评论

Electromagnetic Simulation Aided SAR Target Classification V...

IEEE International Conference on Radar

作者： Xiaoling Lv Xiaolan Qiu Wenming Yu Aerospace Information Research Institute Chinese Academy of Sciences School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing China Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Suzhou Aerospace Information Research Institute Suzhou China State key Laboratory of Millimeter Waves Southeast University Nanjing China

ISBN: (纸本)9781665468893

Convolutional neural networks (CNNs) have made tremendous success in optical images classification recently. However, in synthetic aperture radar (SAR) target classification, it is difficult to annotate a large amount of real SAR images to train CNNs. Sufficient annotated images can be easily obtained through simulation, but the disparity between the simulated images and the real images makes them difficult to directly apply to the real images classification. In this paper, we propose a model that integrates multi-kernel maximum mean discrepancy (MK-MMD) and domain-adversarial training to alleviate this problem. Simulated SAR images with annotation and unlabeled real SAR images are used to train our model. First, we use domain-adversarial training to prompt the model to extract domain-invariant features. Then, the MK-MMD between the hidden representations of simulated images and real images is reduced to narrow domain discrepancy. Experimental results on the real SAR dataset demonstrate that our method effectively solves the domain shift problem and improves the classification accuracy.

关键词： Training Adaptation models Optical fiber networks Feature extraction Optical imaging Radar polarimetry Data models

Integrating Regularization and PnP Priors for SAR Image Reconstruction Using Multi-agent Consensus Equilibrium

学校读者我要写书评

暂无评论

Integrating Regularization and PnP Priors for SAR Image Reco...

International Workshop on Compressed Sensing Theory and its applications to Radar, Sonar and Remote Sensing (CoSeRa)

作者： Yizhe Fan Mingqian Liu Mingxiao Shao Bingchen Zhang Yirong Wu Chinese Academy of Sciences Aerospace Information Research Institute Beijing China Key Laboratory of Technology in Geo-spatial Information Processing and Application System School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Aeronautical Remote Sensing Center Chinese Academy of Sciences

ISBN: (数字)9798350365504

ISBN: (纸本)9798350365511

Multi-agent consensus equilibrium mechanism is a generalization of popular used PnP-ADMM method and composite regularization in computational sensing. We propose a novel SAR image sparse reconstruction method based on MACE mechanism. High-quality SAR images necessitate low ambiguity, high SNR, and low coherent speckle. We therefore integrate regularization priors and PnP priors for azimuth ambiguity suppression, sparsity inducing and despeckling. The proposed method’s performance is evaluated through experiments on both simulated and QILU-1 satellite SAR data.

关键词： TV Azimuth Spaceborne radar Satellite broadcasting Reconstruction algorithms Radar imaging Speckle Radar polarimetry Image reconstruction Signal to noise ratio

Rethinking the key Factors for the Generalization of Remote Sensing Stereo Matching Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jiang, Liting Wang, Feng Zhang, Wenyi Li, Peifeng You, Hongjian Xiang, Yuming Aerospace Information Research Institute Chinese Academy of Sciences Beijing100094 China The Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing100190 China Key Laboratory of Target Cognition and Application Technology Chinese Academy of Sciences Beijing100190 China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100049 China

Stereo matching, a critical step of 3D reconstruction, has fully shifted towards deep learning due to its strong feature representation of remote sensing images. However, ground truth for stereo matching task relies on expensive airborne LiDAR data, thus making it difficult to obtain enough samples for supervised learning. To improve the generalization ability of stereo matching networks on cross-domain data from different sensors and scenarios, in this paper, we dedicate to study key training factors from three perspectives. (1) For the selection of training dataset, it is important to select data with similar regional target distribution as the test set instead of utilizing data from the same sensor. (2) For model structure, cascaded structure that flexibly adapts to different sizes of features is preferred. (3) For training manner, unsupervised methods generalize better than supervised methods, and we design an unsupervised early-stop strategy to help retain the best model with pre-trained weights as the basis. Extensive experiments are conducted to support the previous findings, on the basis of which we present an unsupervised stereo matching network with good generalization performance. We release the source code and the datasets at https://***/Elenairene/RKF_RSSM to reproduce the results and encourage future work. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

A Pylon Detection Method Based on Faster R-CNN in High-Resolution SAR Images

学校读者我要写书评

暂无评论

A Pylon Detection Method Based on Faster R-CNN in High-Resol...

Asian and Pacific Conference on Synthetic Aperture Radar (APSAR)

作者： Lv Xiaoling Jiankun Chen Xiaolan Qiu Chinese Academy of Sciences School of Electronic Electrical and Communication Engineering Aerospace Information Research Institute University of Chinese Academy of Sciences Beijing China Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Aerospace Information Research Institute Chinese Academy of Sciences Beijing China

ISBN: (纸本)9781728173344

SyntheticAperture Radar (SAR) is an active microwave imaging system, which can provide all-time and allweather imaging with a wide observation range. In sub-meter high-resolution SAR images, man-made metallic targets, such as pylons, can be easily detected and their details can be obtained. Therefore, pylon detection using high-resolution SAR images has its unique advantages. To solve the problem that traditional pylon detection methods in SAR images require the manual design of feature extractors and perform poorly in terms of realtime requirements, an effective and fast framework for pylon detection in SAR images is proposed in this paper. It combines Faster R-CNN and sliding window algorithm to detect pylons in large-scene SAR images. Besides, it effectively improves detection accuracy by using data augmentation. The proposed method can detect up to 97.6% on the test set and 80% in largescene GF-3 SAR images. The average time cost of detection is 0. 06s for a slice image and 90s for a large-scale SAR image. Experiments demonstrate good performance in detection rate and instantaneity.

关键词： Costs Poles and towers Manuals Radar imaging Feature extraction Microwave theory and techniques Radar polarimetry

Multipath Imaging for Vehicle Targets in Non-Los Urban SAR

学校读者我要写书评

暂无评论

Multipath Imaging for Vehicle Targets in Non-Los Urban SAR

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Yixin Wang Xiaolan Qiu Xuejiao Wen Chinese Academy of Sciences Key Laboratory of Technology in Geo-Spatial Information Processing and Application Systems Beijing China Chinese Academy of Sciences Aerospace Information Research Institute Beijing China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing China National Key Laboratory of Microwave Imaging Technology Aerospace Information Research Institute Chinese Academy of Sciences Beijing China Laboratory of Spatial Information Intelligent Processing System Suzhou Aerospace Information Research Institute Suzhou China University of Science and Technology of China Hefei China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Integrating Synthetic Aperture Radar (SAR) imaging with unmanned aerial vehicles (UAVs) plays a crucial role in urban area surveillance and situational awareness, benefiting from the low cost, small size, and high flexibility of SAR carried by drones. However, the dense arrangement of high-rise buildings in urban environments creates Urban Canyons and numerous visual blind spots due to occlusion, which weakens the perception capability of SAR. Additionally, SAR imaging results of moving vehicles on the roads between buildings result in severe defocusing due to their non-cooperative motion. In this paper, we establish a vehicle signal model for SAR imaging with UAVs that considers the vehicle body’s translation and the wheels’ rotation. The range history modulation and imaging characteristics of returns caused by translation and micro-motion are derived. Simulation results validate the correctness of the theoretical analysis, and the proposed theory helps explain SAR imaging results, providing support for high-precision focusing and three-dimensional imaging of non-line-of-sight (NLOS) SAR images.

关键词： Visualization Surveillance Simulation Roads Urban areas Buildings Wheels

UNSUPERVISED STEREO MATCHING NETWORK FOR VHR REMOTE SENSING IMAGES BASED ON ERROR PREDICTION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Jiang, Liting Xiang, Yuming Wang, Feng You, Hongjian Aerospace Information Research Institute Chinese Academy of Sciences China Key Laboratory of Technology in Geo-spatial Information Processing and Application System Chinese Academy of Sciences Beijing100190 China School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing101408 China

Stereo matching in remote sensing has recently garnered increased attention, primarily focusing on supervised learning. However, datasets with ground truth generated by expensive airbone Lidar exhibit limited quantity and diversity, constraining the effectiveness of supervised networks. In contrast, unsupervised learning methods can leverage the increasing availability of very-high-resolution (VHR) remote sensing images, offering considerable potential in the realm of stereo matching. Motivated by this intuition, we propose a novel unsupervised stereo matching network for VHR remote sensing images. A light-weight module to bridge confidence with predicted error is introduced to refine the core model. Robust unsupervised losses are formulated to enhance network convergence. The experimental results on US3D and WHU-Stereo datasets demonstrate that the proposed network achieves superior accuracy compared to other unsupervised networks and exhibits better generalization capabilities than supervised models. Our code will be available at https://***/Elenairene/CBEM. Copyright © 2024, The Authors. All rights reserved.

关键词： Contrastive Learning

Synchronization of interconnected embedded systems via timer interrupts

学校读者我要写书评

暂无评论

Synchronization of interconnected embedded systems via timer...

International Conference on Control, Automation, Robotics and Vision (ICARCV)

作者： Fangyuan Li Jiahu Qin Shaoshuai Mou Yu Kang Department of Automation University of Science and Technology of China Hefei China School of Aeronautics and Astronautics Purdue University West Lafayette Indiana United States State Key Laboratory of Fire Science University of Science and Technology of China Hefei China Chinese Academy of Sciences Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Beijing China

ISBN: (纸本)9781509035502

Some applications of the interconnected embedded systems such as sensor networks rely on all nodes in the network to execute certain tasks simultaneously. To meet this demand for simultaneity, a multi-timer model based fully distributed task synchronization algorithm is proposed in this paper. In this multi-timer model, each node containing an embedded system is characterized by a timer. The microcontrollers (MCUs) within the interconnected embedded systems are switched to the assigned tasks by timer interrupts. Each timer decides when to trigger interrupts by only using the information from its neighbors. Task synchronization is realized by using the proposed synchronization algorithm. Some simulation examples are presented in the end to verify the effectiveness of the proposed synchronization algorithm.

关键词： Embedded systems Synchronization Clocks Nickel Switches Algorithm design and analysis Time-frequency analysis

W-net: Simultaneous segmentation of multi-anatomical retinal structures using a multi-task deep neural network

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhao, Hongwei Peng, Chengtao Liu, Lei Li, Bin School of Information Science and Technology University of Science and Technology of China Hefei Anhui230022 China Department of Precision Machinery and Instrumentation University of Science and Technology of China HefeiAnhui230022 China CAS Key Laboratory of Technology in Geo-spatial Information Processing and Application System University of Science and Technology of China Hefei Anhui230026 China

Segmentation of multiple anatomical structures is of great importance in medical image analysis. In this study, we proposed a W-net to simultaneously segment both the optic disc (OD) and the exudates in retinal images based on the multi-task learning (MTL) scheme. We introduced a class-balanced loss and a multi-task weighted loss to alleviate the imbalanced problem and to improve the robustness and generalization property of the W-net. We demonstrated the effectiveness of our approach by applying five-fold cross-validation experiments on two public datasets e_ophtha_EX and DiaRetDb1. We achieved F1-score of 94.76% and 95.73% for OD segmentation, and 92.80% and 94.14% for exudates segmentation. To further prove the generalization property of the proposed method, we applied the trained model on the DRIONS-DB dataset for OD segmentation and on the MESSIDOR dataset for exudate segmentation. Our results demonstrated that by choosing the optimal weights of each task, the MTL based W-net outperformed separate models trained individually on each task. Code and pre-trained models will be available at: https://***/FundusResearch/MTL_for_OD_and_***. Copyright © 2020, The Authors. All rights reserved.

关键词： Deep neural networks