修车大队一品楼qm论坛51一品茶楼论坛,栖凤楼品茶全国楼凤app软件 ,栖凤阁全国论坛入口,广州百花丛bhc论坛杭州百花坊妃子阁

oe1(光电查) - 科学论文

过滤筛选

出版时间
  • 2018
研究主题
  • Conditional Random Fields (CRF)
  • Convolutional Neural Network (CNN)
  • Fine Classification
  • Airborne hyperspectral
应用领域
  • Optoelectronic Information Science and Engineering
机构单位
  • Wuhan University
  • Central South University
  • Hubei University
404 条数据
?? 中文（中国） ▾

  • Dynamic spectrum nonlinear modeling of VIS & NIR band based on RBF neural network for noninvasive blood component analysis to consider the effects of scattering

    摘要： Dynamic spectrum (DS) is expected to achieve non-invasive analysis of blood components by extracting the absorbance of arterial blood at multiple wavelengths. However, the nonlinearity caused by scattering of blood components is still a factor that limits the detection accuracy. According to the idea of “overlay modeling” in the “M + N” theory, theoretically, the consideration of nonlinear factor in modeling analysis can further improve the prediction accuracy of calibration model. But there is currently no recognized formula to describe this nonlinear relationship. In this paper, the ability of RBF neural network to approximate arbitrary nonlinear functions with arbitrary precision is used to approximate the nonlinear relationship between the spectrum and the component concentration from the perspective of ?tting. The calibration sets and prediction sets were randomly selected from the VIS & NIR band DS data of 231 volunteers, and 10 groups of modeling experiments were carried out. The results showed that compared with the conventional partial least squares (PLS) modeling method, the modeling indicators (correlation coe?cient (R) and root mean square error(RMSE)) of prediction set using radial basis function (RBF) neural network modeling have been signi?cantly improved. The modeling experiments suggest that the nonlinearity caused by scattering should not be ignored in DS non-invasive blood component analysis. By ?tting the nonlinear relationship, RBF neural network can better re?ect the actual mapping relationship between the spectrum and the component concentration, because it can not only consider the general part (Linear factor) in DS, but also the details (Nonlinear factor), which can e?ectively improve the accuracy of the non-invasive blood component analysis based on DS.

    关键词： Scattering,Nonlinearity,RBF neural network,Dynamic spectrum (DS),Non-invasive detection

    更新于2025-09-23 15:23:52

  • [IEEE 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC) - Chongqing (2018.6.27-2018.6.29)] 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC) - Research on Single-Frame Super-Resolution Reconstruction Algorithm for Low Resolution Cell Images Based on Convolutional Neural Network

    摘要： At the problem of low resolution and low contrast of cell images collected by lens-less cell imaging system, a novel cell super-resolution reconstruction network (CSRN) based on convolutional neural network is proposed. First, the cell image is collected by lens-less cell imaging system, and then the cell image is down-sampled by bicubic interpolation to obtain low-resolution cell image. Then, the low-resolution cell image is input into the CSRN network for super-resolution reconstruction. The experimental results show that the proposed CSRN network can effectively improve the resolution and contrast of cell images, and the reconstruction effect is better than that of the traditional bicubic interpolation and SRCNN network.

    关键词： convolutional neural network,cell image,super-resolution,lens-less imaging

    更新于2025-09-23 15:23:52

  • An image segmentation method of a modified SPCNN based on human visual system in medical images

    摘要： An image segmentation method of a modified simplified pulse-coupled neural network (MSPCNN) based on human visual system (HVS) is proposed for medical images. The method successfully determines the stimulus input of the MSPCNN according to the characteristics of PCNN and HVS. In order to accomplish the goal, we attempt to deduce the sub-intensity range of central neurons firing by introducing neighboring firing matrix Q and calculating intensity distribution range based on a new MSPCNN(NMSPCNN), and then reveal the way how sub-intensity range parameter Sint generates the stimulus input Sioij closer to HVS. Besides, we try to substitute the above stimulus input into the MSPCNN to extract more suitable lesions for medical images. In contrast to prevalent PCNN models, the MSPCNN has higher segmentation accuracy rates and lower computational complexity because of the parameter setting method. Finally, the proposed method comparing with the state-of-the-art methods has a better performance, presenting the overall metric OEM with MIAS of 0.8784, DDSM of 0.8606 and gallstones of 0.8585.

    关键词： Sub-intensity Range,Modified Simplified Pulse-coupled Neural Network,Image Segmentation,Stimulus Input,Human Visual System

    更新于2025-09-23 15:23:52

  • Neural network-based hybrid signal processing approach for resolving thin marine protective coating by terahertz pulsed imaging

    摘要： A novel approach was presented to enhance the capability of resolving thin coating layers using terahertz pulsed imaging (TPI) based on a neural network-based hybrid signal procession method, which is of great significance for in-line painting applications. In the present work, Terahertz detected signals were obtained by numerical simulation using finite difference time domain (FDTD) method. Models of marine protective coatings with different coating structures were calculated and analyzed. Different signal pre-processing techniques, including Fourier deconvolution, Fast Fourier Transform and wavelet analysis, were employed on the terahertz signals respectively to obtain various signal features. The processed signal was subsequently adopted as the input vectors for a neural network (NN). The optimization procedure for determining the architecture of neural network was investigated and the evaluated results obtained by the different networks were compared. Furthermore, the predicted results of thinner coating layer obtained by multiple-regression analysis method and BP network prediction method respectively were compared. The analysis demonstrated that the best prediction performance was achieved by neural network technique combined with wavelet analysis. Therefore, the hybrid signal processing approach could be recommended for terahertz non-destructive testing applications of marine protective coating.

    关键词： Terahertz pulsed imaging,Non-destructive testing,Neural network,Thin marine protective coating

    更新于2025-09-23 15:23:52

  • [IEEE 2018 IEEE 3rd Optoelectronics Global Conference (OGC) - Shenzhen (2018.9.4-2018.9.7)] 2018 IEEE 3rd Optoelectronics Global Conference (OGC) - High Speed Novel Hybrid Modulation Technique of Visible Light Communication Based on Artificial Neural Network Equalizer

    摘要： Visible light communication (VLC) which realizes data transmission and universal illumination simultaneously has attracted much attention recently. However, the transmission rate of the VLC remains low due to the low bandwidth performance and inter-symbol interference (ISI). Therefore, a hybrid approach using pulse amplitude modulation and pulse width modulation in conjunction with an artificial neural network (ANN) equalizer is proposed, which can theoretically increase the transmission rate by 4 times compared with the traditional way, and provide variable brightness to realize the integration of data transmission and illumination control. In addition, an artificial neural network equalizer is proposed to undo the effects of ISI, considering that the bandwidth of the LED is only 3MHz. Without the ANN equalizer, the maximum transmission rate of the proposed hybrid modulation link only reaches 36 Mbps under the condition of no signal processing; however, with the ANN equalizer, the transmission speed can up to 2.6 Gbps. The proposed system not only achieves a genuine combination of data transmission and control illumination levels, but also realizes a high data rate with less complexity.

    关键词： code division multiple access,pulse amplitude modulation,visible light communication,artificial neural network equalizer

    更新于2025-09-23 15:23:52

  • SCN: Switchable Context Network for Semantic Segmentation of RGB-D Images

    摘要： Context representations have been widely used to profit semantic image segmentation. The emergence of depth data provides additional information to construct more discriminating context representations. Depth data preserves the geometric relationship of objects in a scene, which is generally hard to be inferred from RGB images. While deep convolutional neural networks (CNNs) have been successful in solving semantic segmentation, we encounter the problem of optimizing CNN training for the informative context using depth data to enhance the segmentation accuracy. In this paper, we present a novel switchable context network (SCN) to facilitate semantic segmentation of RGB-D images. Depth data is used to identify objects existing in multiple image regions. The network analyzes the information in the image regions to identify different characteristics, which are then used selectively through switching network branches. With the content extracted from the inherent image structure, we are able to generate effective context representations that are aware of both image structures and object relationships, leading to a more coherent learning of semantic segmentation network. We demonstrate that our SCN outperforms state-of-the-art methods on two public datasets.

    关键词： Context representation,convolutional neural network (CNN),RGB-D images,semantic segmentation

    更新于2025-09-23 15:23:52

  • A CNN With Multiscale Convolution and Diversified Metric for Hyperspectral Image Classification

    摘要： Recently, researchers have shown the powerful ability of deep methods with multilayers to extract high-level features and to obtain better performance for hyperspectral image classification. However, a common problem of traditional deep models is that the learned deep models might be suboptimal because of the limited number of training samples, especially for the image with large intraclass variance and low interclass variance. In this paper, novel convolutional neural networks (CNNs) with multiscale convolution (MS-CNNs) are proposed to address this problem by extracting deep multiscale features from the hyperspectral image. Moreover, deep metrics usually accompany with MS-CNNs to improve the representational ability for the hyperspectral image. However, the usual metric learning would make the metric parameters in the learned model tend to behave similarly. This similarity leads to obvious model’s redundancy and, thus, shows negative effects on the description ability of the deep metrics. Traditionally, determinantal point process (DPP) priors, which encourage the learned factors to repulse from one another, can be imposed over these factors to diversify them. Taking advantage of both the MS-CNNs and DPP-based diversity-promoting deep metrics, this paper develops a CNN with multiscale convolution and diversified metric to obtain discriminative features for hyperspectral image classification. Experiments are conducted over four real-world hyperspectral image data sets to show the effectiveness and applicability of the proposed method. Experimental results show that our method is better than original deep models and can produce comparable or even better classification performance in different hyperspectral image data sets with respect to spectral and spectral–spatial features.

    关键词： deep metric learning,determinantal point process (DPP),image classification,multiscale features,Convolutional neural network (CNN),hyperspectral image

    更新于2025-09-23 15:23:52

  • FPGA-Based Implementation of an Artificial Neural Network for Measurement Acceleration in BOTDA Sensors

    摘要： In recent years, using distributed fiber-optic sensors based on Brillouin scattering, for monitoring pipelines, tunnels, and other constructional structures have gained huge popularity. However, these sensors have a low signal-to-noise ratio (SNR), which usually increases their measurement error. To alleviate this issue, ensemble averaging is used which improves the SNR but in return increases the measurement time. Reducing the noise by averaging requires hundreds or thousands of scans of the optical fiber; hence averaging is usually responsible for a large percent of the entire system latency. In this paper, we propose a novel method based on artificial neural network for SNR enhancement and measurement acceleration in distributed fiber-optic sensors based on the Brillouin scattering. Our method takes the noisy Brillouin spectrums and improves their SNR by 20 dB, which reduces the measurement time significantly. It also improves the accuracy of the Brillouin frequency shift estimation process and its latency by more than 50% in comparison with the state-of-the-art software and hardware solutions.

    关键词： Artificial neural network (ANN),digital signal processing,optical fibers,curve fitting,field-programmable gate arrays (FPGAs)

    更新于2025-09-23 15:23:52

  • Estimation of spatiotemporal PM1.0 distributions in China by combining PM2.5 observations with satellite aerosol optical depth

    摘要： Particulates smaller than 1.0 μm (PM1.0) have strong associations with public health and environment, and considerable exposure data should be obtained to understand the actual environmental burden. This study presented a PM1.0 estimation strategy based on the generalised regression neural network model. The proposed strategy combined ground-based observations of PM2.5 and satellite-derived aerosol optical depth (AOD) to estimate PM1.0 concentrations in China from July 2015 to June 2017. Results indicated that the PM1.0 estimates agreed well with the ground-based measurements with an R2 of 0.74, root mean square error of 19.0 μg/m3 and mean absolute error of 11.4 μg/m3 as calculated with the tenfold cross-validation method. The diurnal estimation performance displayed remarkable single-peak variation with the highest R2 of 0.80 at noon, and the seasonal estimation performance showed that the proposed method could effectively capture high-pollution events of PM1.0 in winter. Spatially, the most polluted areas were clustered in the North China Plain, where the average estimates presented a bimodal distribution during daytime. In addition, the quality of satellite-derived AOD, the robustness of the interpolation algorithm and the proportion of PM1.0 in PM2.5 were confirmed to affect the estimation accuracy of the proposed model.

    关键词： Himawari-8,PM1.0,Neural network,Air pollution,Aerosol optical depth

    更新于2025-09-23 15:23:52

  • DeepLens

    摘要： We aim to generate high resolution shallow depth-of-field (DoF) images from a single all-in-focus image with controllable focal distance and aperture size. To achieve this, we propose a novel neural network model comprised of a depth prediction module, a lens blur module, and a guided upsampling module. All modules are differentiable and are learned from data. To train our depth prediction module, we collect a dataset of 2462 RGB-D images captured by mobile phones with a dual-lens camera, and use existing segmentation datasets to improve border prediction. We further leverage a synthetic dataset with known depth to supervise the lens blur and guided upsampling modules. The effectiveness of our system and training strategies are verified in the experiments. Our method can generate high-quality shallow DoF images at high resolution, and produces significantly fewer artifacts than the baselines and existing solutions for single image shallow DoF synthesis. Compared with the iPhone portrait mode, which is a state-of-the-art shallow DoF solution based on a dual-lens depth camera, our method generates comparable results, while allowing for greater flexibility to choose focal points and aperture size, and is not limited to one capture setup.

    关键词： Neural Network,Shallow Depth of Field

    更新于2025-09-23 15:23:52