- 标题
- 摘要
- 关键词
- 实验方案
- 产品
过滤筛选
- 2018
- Conditional Random Fields (CRF)
- Convolutional Neural Network (CNN)
- Fine Classification
- Airborne hyperspectral
- green tide
- Elegant End-to-End Fully Convolutional Network (E3FCN)
- deep learning
- remote sensing
- Moderate Resolution Imaging Spectroradiometer (MODIS)
- Optoelectronic Information Science and Engineering
- Ocean University of China
- Wuhan University
- Central South University
- Hubei University
-
[IEEE 2018 IEEE 7th International Conference on Photonics (ICP) - Kuah (2018.4.9-2018.4.11)] 2018 IEEE 7th International Conference on Photonics (ICP) - Passive Element Fault Analysis at the Last Mile of the FTTH Network in Malaysia
摘要： The optical transceiver performance for GPON FTTH (OLTTx and ONURx) for 300 live subscribers at legacy area has been monitored hourly for a duration of four months. The optical performance is then plotted against its faults due to the passive element at the last mile of the GPON FTTH network. The results show that the fault rate for aerial type of fiber distribution (FDP) is higher (1.3%) than building, street cabinet and underground FDP types. More than 65% of the faults are contributed by optical fibers (drop fibers) and the rest is due to fiber connectors. The optical power received (ONURx) spectrum shows three types of patterns before disconnection; sudden disconnection, fluctuation and degradation of ONURx. This study provides preliminary information necessary for developing the passive element fault prediction for GPON FTTH at the last mile of the network.
关键词： link loss,GPON FTTH network,optical transmit power: receive power,optical fiber,signals degradation,passive elements
更新于2025-09-23 15:22:29
-
A Wake-Up Receiver With a Multi-Stage Self-Mixer and With Enhanced Sensitivity When Using an Interferer as Local Oscillator
摘要： An ultra-low power wake-up receiver with an energy-detection passive-RF architecture uses a multi-stage self-mixer that has a better conversion gain than the conventional envelope detector. The self-mixer, co-designed with the RF matching network, optimizes the sensitivity and minimizes the power consumption of the receiver. A wake-up receiver prototype in 0.13-μm CMOS operates at 550 MHz, consumes 220 nW from 0.5 V, and achieves a sensitivity of ?56.4 dBm at a 400-kb/s chip rate using an 11-bit wake-up code. When a large interferer is present, the receiver operates in an interferer-enhanced mode, leveraging the interferer as a local oscillator to improve the sensitivity; in the presence of a ?43.5-dBm interferer, a ?63.6-dBm sensitivity is achieved while consuming 1.1 μW.
关键词： self-mixer,wake-up receivers (WuRXs),wake-up radios,low-power wide-area network (LPWAN),Envelope-detectors,low-power wireless
更新于2025-09-23 15:22:29
-
Deep Refinement Network for Natural Low-Light Image Enhancement in Symmetric Pathways
摘要： Due to the cost limitation of camera sensors, images captured in low-light environments often suffer from low contrast and multiple types of noise. A number of algorithms have been proposed to improve contrast and suppress noise in the input low-light images. In this paper, a deep refinement network, LL-RefineNet, is built to learn from the synthetical dark and noisy training images, and perform image enhancement for natural low-light images in symmetric—forward and backward—pathways. The proposed network utilizes all the useful information from the down-sampling path to produce the high-resolution enhancement result, where global features captured from deeper layers are gradually refined using local features generated by earlier convolutions. We further design the training loss for mixed noise reduction. The experimental results show that the proposed LL-RefineNet outperforms the comparative methods both qualitatively and quantitatively with fast processing speed on both synthetic and natural low-light image datasets.
关键词： deep refinement network,image enhancement,low-light image
更新于2025-09-23 15:22:29
-
VLC and D2D Heterogeneous Network Optimization: A Reinforcement Learning Approach Based on Equilibrium Problems with Equilibrium Constraints
摘要： The Radio Frequency (RF) spectrum crunch has triggered the harnessing of other sources of bandwidth, for which visible light is a promising candidate. Even though Visible Light Communication (VLC) ensures high capacity, coverage is limited. This necessitates the integration of VLC and Device-to-Device (D2D) technologies into heterogeneous networks. In particular, mobile users which are accessible by the VLC transmitters can relay data to mobile users which are not, by means of D2D communication. However, due to the distributed behaviors of mobile users, determining optimal data transmission routes from VLC transmitters to end mobile devices is a major challenge. In this paper, we propose a Reinforcement Learning (RL) based approach to determine multi-hop data transmission routes in an indoor VLC-D2D heterogeneous network. We obtain the rewards for the RL based method dynamically, by formulating the interactions between the mobile users relaying the data as an Equilibrium Problem with Equilibrium Constraints (EPEC) and using Alternating Direction Method of Multipliers (ADMM) to solve it. The proposed technique can achieve optimal data transmission routes in a distributed manner. Simulation results demonstrate the effectiveness of the proposed approach, showing that transmission routes with low delays and high capacities can be achieved through the learning algorithm.
关键词： Device-to-Device,Reinforcement Learning,Visible Light Communication,Alternating Direction Method of Multipliers,Heterogeneous Network,Equilibrium Problem with Equilibrium Constraints
更新于2025-09-23 15:22:29
-
Brain Storm Optimization Graph Theory (BSOGT) and Energy Resource Aware Virtual Network Mapping (ERVNM) for Medical Image System in Cloud
摘要： With the development of Internet and the make use of Internet for medical information, the demand for huge scale and reliable managing medical information has brought out the huge scale Internet data centers. This work that has been presented here highlights the structural lay out and formulation of the medical information model. The aim of presenting this to aid medical departments as well as workers to exchange information and integrate available resources that help facilitate the analysis to be conducted on the given information. Software here comprises of medical information and offers a comprehensive service structure that benefits medical data centers. VNM or Virtual Network Mapping (VNM) essentially relates to substrate network that involves the installation and structuring of on demand virtual machines. These however are subjective to certain limitations that are applicable in relation to latency, capacity as well as bandwidth. Data centers need to dynamically handle cloud workloads effectively and efficiently. Simultaneously, since the mapping of virtual and physical networks with several providers’ consumes more time along with energy. In order to resolve this issue, VNM has been mapped by making use of Graph Theory (GT) matching, a well-studied database topic. (i) Brain Storm Optimization Graph Theory (BSOGT) is introduced for modeling a virtual network request in the form of a GT with different resource constraints, and the substrate networks here is considered being a graph. For this graph the nodes and edges comprise of attributes that indicate their constraints. (ii) The algorithm that has been recently introduced executes graph decomposition into several topology patterns. Thereafter the BSOGT is executed to solve any issues that pertain to mapping. (iii) The model that has been presented here, ERVNM and the BSOGT are used with a specific mapping energy computation function.(iv) Issues pertaining to these are categorized as being those related to virtual network mapping as the ACGT and optimal solution are drawn by using effective integer linear programming. ACGT, pragmatic approach, as well as the precise and two-stage algorithms performance is evaluated by means of cloud Simulator environment. The results obtained from simulation indicate that the BSOGT algorithm attains the objectives of cloud service providers with respect to Acceptance ratio, mapping percentage, processing time as well as Convergence Time.
关键词： Virtual Network Mapping (VNM),Distributed cloud computing and optimization,Virtualization quality of services (Qos),Brain Storm Optimization Graph Theory (BSOGT)
更新于2025-09-23 15:22:29
-
Detection of Circulating Tumor Cells in Fluorescence Microscopy Images Based on ANN Classifier
摘要： Circulating tumor cells (CTCs) is a clinical biomarker for cancer metastasis. CTCs are cells circulating in the body of patients by being separated from primary cancer and entering into blood vessel. CTCs spread every positions in the body, and this is one of the cause of cancer metastasis. To analyze them, pathologists get information about metastasis without invasive test. CTCs test is conducted by analyzing the blood sample from patient. The fluorescence microscope generates a large number of images per each sample, and images contain a lot of cells. There are only a few CTCs in images and cells often have blurry boundaries. So CTCs identification is not an easy work for pathologists. In this paper, we develop an automatic CTCs identification method in fluorescence microscopy images. This proposed method has three section. In the first approach, we conduct the cell segmentation in images by using filtering methods. Next, we compute feature values from each CTC candidate region. Finally, we identify CTCs using artificial neural network algorithm. We apply the proposed method to 5895 microscopy images (7 samplesas), and evaluate the effectiveness of our proposed method by using leave-one-out cross validation. We achieve the result of performance tests, a true positive rate is 92.57% and false positive rate is 9.156%.
关键词： Fluorescence microscopy image,Artificial neural network,Feature analysis,Computer aided diagnosis,Circulating tumor cells
更新于2025-09-23 15:22:29
-
Mixed Pixel Decomposition Based on Extended Fuzzy Clustering for Single Spectral Value Remote Sensing Images
摘要： The presence of mixed pixels in remote sensing images is the major issue for accurate classification. In this paper, we have focused on two aspects of mixed pixel problem: firstly, to identify mixed pixels from an image and secondly to label them to their appropriate class. In phase I, extraction of mixed pixels has been performed from the RSI images-based super-pixel algorithm and RGB model by using fuzzy C-means (FCM). In phase II, the extracted mixed pixel from phase I has been decomposed to the appropriate class. This new proposed technique is the amalgamation of PSO-FCM (particle swarm optimization-fuzzy C-means) for clustering of mixed pixels and ANN-BPO (artificial neural network-biogeography-based particle swarm optimization) for the classification purpose. Experimental results reveal that the proposed method has improved the accuracy as compared to the existing techniques and succeeds in better classification of the remote sensing images.
关键词： Fuzzy C-means,BBO,Remote sensing images,Pure pixels,Mixed pixels,PSO,Neural network
更新于2025-09-23 15:22:29
-
Millimeter-wave home area network prospect with cost-effective RoF links
摘要： The growth towards the millimeter-wave band in the home area networks (HAN) leads to high data rate transmission to satisfy the new user services. Unfortunately, the transmission coverage in this band is limited to short distances because of the strong air absorption and obstacles such as walls. The effort is then focused on the extension of the network coverage of the wireless link in this band. Solutions based on multiple connected access points to optical fibers are useful methods to ensure wireless connectivity to the entire home. For HAN applications, radio-over-fiber (RoF) using intensity modulation and direct detection technique is the mostly favorite technology for the transmission of a broadband wireless signal because dealing with a cost-effective solution. We investigate in this paper the performance of such RoF-wireless architecture with low-cost optoelectronic modules through the error vector magnitude (EVM) metric. The RoF links investigated are a directly modulated VCSEL with integrated photoreceiver module, an electroabsorption-modulated laser with PIN photodiode and a Mach–Zehnder Modulator with PIN photodiode. A simulation approach based on equivalent electrical circuit models of photonic components is developed in ADS (Advanced Design System) by using a co-simulation technique that combines both analog and digital signals. The downlink channel of the complete transmission system including wireless channel and frequency conversion circuits to millimeter-wave (mm-wave) band is studied by simulation. The obtained results of EVM show good performances of cost-effective links with QPSK and 16-QAM modulation over a dynamic range of 15 dB.
关键词： Radio over fiber,Intensity modulation–direct detection,Home area network,Wireless channel
更新于2025-09-23 15:22:29
-
Modal transmission line theory of plane wave excited layered media with multiple conductive anisotropic sheets at the interfaces
摘要： A succinct and efficient transmission line formulation is presented for the analysis of plane wave excited planar multilayered uniaxial media comprising multiple conductive sheets at the interfaces. The conductive sheets may be anisotropic with a fully populated conductivity tensor, which results in the coupling the transverse-magnetic and transverse-electric waves. The transmission line formalism is thus extended to accommodate this hybrid mode of propagation. Three different algorithms for the solution of the transmission line network are developed and compared, including a numerically stable and efficient scattering matrix procedure. Numerical examples are included that validate the theoretical method.
关键词： Anisotropic conductive sheet,Terahertz,Transmission line network,Transition matrix,Graphene,Scattering matrix,Plane wave,Planar multilayer
更新于2025-09-23 15:22:29
-
Hyperspectral signature analysis using neural network for grade estimation of copper ore
摘要： The ever-increasing demand for the different metal and mineral resources from the earth’s subsurface has brought tremendous pressure on the geochemical laboratory for the growing countries. The success of any mining industry relies on the estimated values of ore grade in the mineral deposit. Hence, rapid assessment of ore grade is critical in daily schedule in mines operations. Commonly the assay value is determined by chemical analysis or X-Ray Fluorescence (XRF), which is one of the constrained by real-time grade estimation, duration of sample preparation and processing. Several researches carried out in exploration and revealed that hyperspectral technique is a promising tool for mineral identification and mapping. The goal of the present study is to determine the effectiveness of narrow band spectroscopy in Cu grade estimation. To achieve this, a multilayer feed-forward neural network model has been developed to establish a functional link between hyperspectral signature derived features with the copper grade. Altogether eight different types of features including absorption depth, band depth center, the area under the absorption curve, full width at half maxima were extracted from continuum removed spectra along with derivative reflectance features, e.g. band depth ratio, 1st and 2nd slopes from the hyperspectral profile. The dimensionality was reduced by applying Principal Component Analysis onto the extracted features. The first seven PCAs are then used as input vector of the ANN model. A five-fold cross-validation exercise is carried out for model performance. The high degree of correlation reveals that the PCA generated feature from hyperspectral data coupled with ANN model could be an alternative approach to predict the copper grade for the copper mine.
关键词： copper grade,ore grade estimation,spectral feature,K-Fold cross validation,principal component analysis,artificial neural network
更新于2025-09-23 15:22:29