- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2018 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) - Las Vegas, NV (2018.4.8-2018.4.10)] 2018 IEEE Southwest Symposium on Image Analysis and Interpretation (SSIAI) - Underwater Image Restoration using Deep Networks to Estimate Background Light and Scene Depth
摘要： Images taken underwater often suffer color distortion and low contrast because of light scattering and absorption. An underwater image can be modeled as a blend of a clear image and a background light, with the relative amounts of each determined by the depth from the camera. In this paper, we propose two neural network structures to estimate background light and scene depth, to restore underwater images. Experimental results on synthetic and real underwater images demonstrate the effectiveness of the proposed method.
关键词： depth estimation,image restoration,convolutional neural networks,Underwater images
更新于2025-09-23 15:21:21
-
Software Implementation of 10G-EPON Upstream Physical-Layer Processing for Flexible Access Systems
摘要： This paper summarizes our studies on the software implementation of passive optical network (PON) physical-layer (PHY) processing to maximize the flexibility of the optical line terminal (OLT) in access systems, and demonstrates the softwarization of complete 10G-EPON upstream PHY of an optical system for the first time. Softwarization based on general-purpose hardware is more difficult to implement successfully than the application-specific integrated circuits (ASICs) used by dedicated hardware due to the marginal performance. Our work utilizes general-purpose graphic processing units (GPUs) as the implementation device as they offer significant computation performance. Our key advances that yield the softwarization of PON PHY processing, the GPU direct transfer technique and a low-complexity algorithm, are described. In addition, this paper describes the parallel implementation method of PON PHY decoding for upstream transmission. Demonstration results show that the proposed algorithm together with implementation satisfy the 10G-EPON standard by achieving 10.3125-Gbps real-time processing.
关键词： SDN,NFV,Access networks,GPU,PHY coding and scrambler
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Ship Detection Based on Deep Convolutional Neural Networks for Polsar Images
摘要： In this paper, we proposed a ship detection method based on deep convolutional neural networks for PolSAR images. The proposed ship detector firstly segments PolSAR images into sub-samples using a sliding window of fixed size to effectively extract translational-invariant spatial features. Further, the modified faster region based convolutional neural network (Faster-RCNN) method is utilized to realize ship detection for ships with different sizes and fusion the detection result. Finally, the proposed method was validated using real measured NASA/JPL AIRSAR datasets by comparing the performance with the modified constant false alarm rate (CFAR) detector. The comparison results demonstrate the validity and generality of the proposed detection algorithm.
关键词： Deep convolutional neural networks,polarimetric synthetic aperture radar (PolSAR),ship detection
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Data Augmentation Method of SAR Image Dataset
摘要： Large-scale high-quality, standardized, measurable and accurate data is the key to promote the progress of the algorithm in the radar remote sensing. Data scaling is a widespread technology that increases the size of a labeled training set dataset through specific data transformations. Synthetic Aperture Radar (SAR) image simulators based on computer-aided mapping models play an important role in SAR applications such as automatic target recognition and image interpretation, but the accuracy of this simulator is due to geometric errors and simplification of electromagnetic calculations. In order to achieve a SAR image datasets with the known target and azimuth angles, we can generate the desired image directly from a known image database. We can realize the augmentation of SAR image data set through linear synthesis and Generative Adversarial Networks, which can generate SAR images for the specified azimuth.
关键词： generative adversarial networks,SAR image,linear synthesis
更新于2025-09-23 15:21:21
-
COMPACT ROTMAN LENS STRUCTURE CONFIGURATIONS TO SUPPORT MILLIMETER WAVE DEVICES
摘要： The development of modern communication devices for the latest technologies such as 5G has brought the millimeter wave technology into the spotlight because it offers higher data rates and bandwidth. Since highly directional transmissions are necessary for communication in these frequencies due to high path loss and atmospheric absorption, the use of adaptive antennas is inevitable. Rotman lenses have long been used as analog beam forming networks to support linear array antennas for electronic scanning. Their broad bandwidth and planar structure make them ideal for a variety of applications. However, their overall dimensions can be prohibitive especially for large scan angles. In this paper, we propose a few compact configurations that reduce the overall dimensions of Rotman lens as much as 50% without degrading its performance.
关键词： compact configurations,Rotman lens,millimeter wave,beam forming networks,electronic scanning
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Oil-Palm Tree Detection in Aerial Images Combining Deep Learning Classifiers
摘要： Palm oil is the largest vegetable oil in the world in terms of produced volume, and 75% of global production is used for food and cooking purposes. Sustainable management of the producing areas calls for the frequent assessment of field conditions. In this paper, we investigate an automatic algorithm based on deep learning that is capable to build an inventory of individual oil-palm trees using aerial color images collected by unmanned aerial vehicles. The idea consists of combining the outputs of two independent convolutional neural networks, trained on partially distinct subsets of samples and different spatial scales to capture coarse and fine details of image patches. The estimated posterior probabilities are combined by simple averaging as to improve detection accuracy and estimate the confidence for each individual detection. Non-maxima suppression removes weak detections. Experiments at three commercial oil-palm tree plantations sites aged two, four, and 16 years in Northern Brazil revealed overall detection accuracies in the range 91.2–98.8% using orthomosaics of decimeter spatial resolution. The proposed approach can be a useful component of a forest monitoring system based on remote sensing.
关键词： convolutional neural networks,classification,Tree counting,remote sensing,forest inventory
更新于2025-09-23 15:21:21
-
[IEEE IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Valencia, Spain (2018.7.22-2018.7.27)] IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium - Cloud Detection from RGB Color Remote Sensing Images with Deep Pyramid Networks
摘要： Cloud detection from remotely observed data is a critical pre-processing step for various remote sensing applications. In particular, this problem becomes even harder for RGB color images, since there is no distinct spectral pattern for clouds, which is directly separable from the Earth surface. In this paper, we adapt a deep pyramid network (DPN) to tackle this problem. For this purpose, the network is enhanced with a pre-trained parameter model at the encoder layer. Moreover, the method is able to obtain accurate pixel-level segmentation and classification results from a set of noisy labeled RGB color images. In order to demonstrate the superiority of the method, we collect and label data with the corresponding cloud/non-cloudy masks acquired from low-orbit Gokturk-2 and RASAT satellites. The experimental results validates that the proposed method outperforms several baselines even for hard cases (e.g. snowy mountains) that are perceptually difficult to distinguish by human eyes.
关键词： Cloud Detection,Deep Pyramid Networks
更新于2025-09-23 15:21:21
-
[Springer Theses] Quantum Confined Excitons in 2-Dimensional Materials || Introduction: 2d-Based Quantum Technologies
摘要： Framed within the growing giant of quantum technologies, with billions in expenditure world-wide and rapidly growing, we harness the exciting physics and technological promise of 2-dimensional materials to create atomically-thin quantum devices capable of emitting single photons and capturing single spins. In this thesis we present the alliance of 2d and quantum information technology - the first steps towards hybrid light-matter quantum networks set in a low power, scalable on-chip platform.
关键词： quantum technologies,single spins,single photons,quantum networks,2-dimensional materials
更新于2025-09-23 15:21:21
-
[Lecture Notes in Computer Science] Pattern Recognition and Computer Vision Volume 11259 (First Chinese Conference, PRCV 2018, Guangzhou, China, November 23-26, 2018, Proceedings, Part IV) || Asymmetric Two-Stream Networks for RGB-Disparity Based Object Detection
摘要： Currently, most methods of object detection are monocular-based. However, due to the sensitivity to color, these methods can not handle many hard samples. With the depth information, disparity maps are helpful to get over this problem. In this paper, we propose the asymmetric two-stream networks for RGB-Disparity based object detection. Our method consists of two networks, Disparity Representations Mining Network (DRMN) and Muti-Modal Detection Network (MMDN), to combine RGB and disparity data for more accurate detection. Unlike normal two-stream networks, our model is asymmetric because of the di?erent capacity of RGB and disparity data. We are the ?rst to propose a deep learning based framework utilizing only binocular information for object detection. The experiment results on KITTI and our proposed BPD dataset demonstrate that our method can achieve a signi?cant increase in performance e?ciently and get the state-of-the-art.
关键词： Two-stream networks,Object detection,RGBD data
更新于2025-09-23 15:21:21
-
[IEEE 2018 10th IAPR Workshop on Pattern Recognition in Remote Sensing (PRRS) - Beijing (2018.8.19-2018.8.20)] 2018 10th IAPR Workshop on Pattern Recognition in Remote Sensing (PRRS) - Road Map Update from Satellite Images by Object Segmentation and Change Analysis
摘要： This paper studies to detect the change of road network from remote sensing images. Our purpose is to apply the method for practical usages, such as navigation map updating, road construction supervision, disaster survey, and so on. The proposed approach assumes that there is an outdated road map and the updating job is performed by detecting new road network and comparing the changes. The deep convolution network is utilized for precisely segmenting road areas. An image registration and correction procedure is performed to unify the spatial coordinate reference between the old map and the new road detection results. Then, we modify and standardize the extracted road segments, and apply it to determine the road variation of different periods. Experiments show that, the proposed method successfully identifies road changes, which is useful for fast map update in remote areas.
关键词： road change detection,road region extraction,image segmentation,deep convolutional networks
更新于2025-09-23 15:21:21