Search

Liver tumor is one of the most deadly cancers around the world. Since the traditional liver tumor segmentation method is not only time-consuming but also low-precision, a precise and automatic liver tumor segmentation model is urgently needed in clinical practice. To address these above, we present a novel Multi-scale Pyramid Attention network for liver tumor segmentation. This proposed method replaces the traditional convolution network into pyramid convolution, which simultaneously can process the input data and capture different levels of detail features. Moreover, the Squeeze-and-Excitation (SE) block is introduced to obtain the channel relevance between feature maps. We conduct extensive experiments to validate MPA-Net on the MICCAI 2017 Liver Tumor Segmentation Challenge dataset. A large number of experiments demonstrate that the MPA-Net achieves better performance than other state-of-the-art methods. The dice per case score and dice global score of liver tumor segmentation are 0.771 and 0.826.

Automatic segmentation of liver from CT volumes is essential for diagnosis and treatment of liver cancer. In order to achieve accurate segmentation of liver, we proposed a novel 3D residual UNet with context collection module (CCM). The proposed CCM can capture the long-range correlations between all pixels, thereby producing rich and dense contextual information with full-image dependency. We applied CCM at the end of contraction path of UNet to ensure the features fed into expanding path rich in contextual information. Therefore, our model is more accurate and reliable for liver edge segmentation. After extensive experiments, our model finally achieved a state-of-the-art result as Dice per case score of 0.961 and Dice global score of 0.963.

With the rapid development of Internet technology, especially the massive unstructured data generated by various Internet applications generated by WEB2.0 technology and the high concurrent data read and write requirements, traditional relational databases have exposed large pressures such as disk IO, Weak scalability and many other insurmountable shortcomings, but some new databases and non-relational databases provide Internet applications with high concurrency, high read and write, and high flexibility database support services. This paper studies the principle and technical characteristics of MongoDB database fragmentation technology and automatic load balancing technology, analyzes and studies efficient fragmentation key selection algorithms, and proposes an automatic fragmentation key decision method based on database operation scenario analysis.

We propose a design of TE mode magneto-optical (MO) circulator based on the asymmetric slot waveguide with the multimode interference (MMI) principle. The nonreciprocal phase shift (NRPS) of two high-order TE modes in the MO waveguide lead to different phase difference between them when light propagates in the opposite direction. In-plane asymmetric structure makes the device to work in TE polarization. The crosstalk is less than 50 dB with 1.75 dB insertion loss at the wavelength of 1550 nm. A 14 nm broadband device design based on this structure is demonstrated, which shows insertion loss below 3 dB and crosstalk less than 20 dB.

A deep learning-based wavelength controllable forward prediction and inverse design model of nanophotonic devices is proposed. Both the target time-domain and wavelength-domain information can be utilized simultaneously, which enables multiple functions, including power splitter and wavelength demultiplexer, to be implemented efficiently and flexibly., Comment: Accepted by ECOC2020

All-dielectric metasurfaces have attracted great research interest due to their low loss and versatility to control light. In this paper, we report an active all dielectric metasurface based on Si/VO 2 hybrid meta-atoms operating in the long wave infrared (LWIR) wavelength range. We show that by judiciously designing the location of VO 2 layer in the meta-atom, the magnetic dipole resonance can be independently and reconfigurably turned on/off upon phase transition of the VO 2 material, leaving the electric dipole resonance intact. Our work paves the way for independent control of different Mie resonances in all dielectric metasurfaces for infrared camouflage, radiation control and photodetection applications.

With the rapid development of big data, the scale, dimensions, diversity and sparsity of high-dimensional data restrict the effectiveness of traditional clustering algorithms. This paper mainly focuses on high-dimensional data clustering. Starting from the traditional K-means clustering algorithm and subspace clustering algorithm based on self-representation model, an improved algorithm is designed and implemented based on the existing clustering algorithm in this paper. The improved algorithm has better clustering quality by combining the "distance optimization method" and the "density method" to determine the initial clustering center. The feasibility and effectiveness of improved algorithm are verified through simulation experiments.

The replica technology in cloud storage can not only maintain the high availability of the system, but also improve the overall performance of the system. This paper analyzes the limitations of the existing replica placement strategy in heterogeneous environment, and proposes a reasonable replica placement improvement strategy based on the comprehensive performance evaluation value of the node. The experimental results show that the improved replica placement strategy can make the replica distribution more reasonable and balanced on the premise of ensuring the overall availability of the system.

The protection of endangered animals is of great significance to the sustainable development of the earth. This paper analyzes the survival mode of the endangered species South China Tiger by constructing three groups of prey competition models. The sensitivity and error of the model were analyzed by changing the parameters such as the life and weight of the South China Tiger after the model was established. The results show that the model is robust and provides ideas and solutions for protecting South China Tigers, and can be extended to a wider range of applications.

The recent trends of saliency modeling suggest that deep learning based saliency models are viable tools as they achieve outstanding results. However, they require large training times and expensive hardware. In this study, we present a bottom up saliency model that exploit deep features from a pre-trained deep convolutional neural network. We evaluate deep features of a variety of activations from seven state-of-the-art deep convolutional neural networks, including 35 implementations of the proposed bottom up saliency model. All implementations are evaluated over a popular dataset using three evaluation metrics. The experimental results suggest that deep features of a network trained for object classification can be used for saliency modeling. While the ranking of the saliency models is varied across scores, majority of the GoogLeNet based implementations outperformed all other implementations.

To support mobile users in contributing sensing data for making urban management decisions, in ShanghaiGrid, unified data gathering operations are to be performed. For citywide coverage, public vehicles accept data from surrounding users and hand over to computing center through wireless base stations (BSs) deployed in the city. Meanwhile, several among the vehicles are hired as relays, which assist gathering from others with multicopy and multihop forwarding towards the BSs. However, the budget shared by deploying BSs and hiring relays is limited. We explore how to decide BS deployment and relay- based forwarding for efficient gathering under the budget. The challenge lies in the great uncertainty about collection opportunities of candidate locations and vehicles in future gathering processes. In this paper, we present an empirical approach for the problem. To tackle the challenges, we characterize collection performance as function of temporal data paths towards each candidate, and formulate the problem as a multiobjective optimization problem. To solve it, we reveal regular relations between the candidates and estimate expected importance of them with large set of real vehicular traces; and develop an algorithmic framework for BS deployment and corresponding forwarding strategy. Extensive trace-driven simulations demonstrate the efficacy of the approach.

This paper discusses the time-constrained data delivery problem in vehicular ad hoc networks (VANETs). The unique characteristics of the network present great challenges to the issue. First, there are no always-connected forwarding routes between vehicles. Second, there is an intrinsic tradeoff between communication cost and delivery quality. Third, there is great uncertainty about vehicular mobilities. Exploiting vehicular trajectories, we present a constructive approach called TNC to tackle the challenges. In TNC, contact-based data forwarding and communication connections through the mobile network are incorporated. TNC first predicts the time-stamped inter-vehicle contacts and establishes expected contact graph, based on which it then computes the configuration for enhancing the connectivity of the network while introducing minimum number of mobile communication connections. Extensive simulations based on three real vehicular traces collected from 2,000 taxis and 1,400 buses in Shanghai, and 2,500 taxis in Shenzhen have been conducted and results demonstrate the efficacy of our approach.

In this work, we report our recent progress on monolithic integration of magneto-optical oxide thin films and nonreciprocal photonic devices on silicon. We demonstrate strong Faraday rotation of − 6000 deg/cm at 1550 nm wavelength in silicon integrated Ce 1.5 Y 1.5 Fe 5 O 12 films, which exceeds the Faraday rotation of epitaxial CeiY 2 Fe 5 Oi 2 thin films on garnet substrates. We also present the design of a broadband optical isolator using silicon/CeYIG/silicon MMI structure. The device shows 20 dB isolation bandwidth of 1.6 nm, and insertion of 0.817 dB at 1550 nm wavelength, with a very compact device footprint of 310.42 μm. The fabrication tolerance and influence of a YIG seed layer on the device performance are also discussed.

Surface roughness has significate impacts on the land surface microwave emission and is an important process that should be considered in the development of soil moisture retrieval algorithm. The surface roughness can be descripted by two parameters: rms height (h) and correlation length (l), with a distribution function either in Gaussian or exponential form. These two parameters can be observed at field scale (1m ∼ 10 m) by using roughmeters, but they are hard to measure at the scale of satellite footprint (∼ 40 km). Currently, in the most of current soil moisture retrieval algorithm, the surface roughness parameters are either set as constant values or obtained from limited in situ observation. Consequently, the space heterogeneity of land surface roughness is not well handled. In this study, we developed a systemic method to estimate surface roughness parameters at satellite footprint scale and apply it to retrieve surface soil moisture over the Tibetan Plateau from AMSR-E/AMSR2 observation. At the first step, we optimized the surface roughness parameter over the Tibetan Plateau with using land surface model simulation and time series of microwave observation. At the same time, based on the accurate soil temperature moisture observations obtained at the Naqu dense soil network, surface roughness parameters were calibrated by using a radiative transfer model in which AIME was used. Finally, a regression relationship was developed between the optimized parameters and calibrated parameters and applied to correct the optimized parameter over whole Tibetan Plateau. With the new surface roughness parameters set, soil moisture over Tibetan Plateau was retrieved from AMSR-E/AMSR2 observation and validated at other two dense soil network: Maqu and Yadong. The study is carrying out now, more details will be reported in the conference.

In this study, we present a new magneto-optical surface plasmon resonance sensor based on magnetic oxide and noble metal thin films. In particial, we demonstrate enhanced index sensing figure of merit (FOM) in a waveguide coupled bimetallic MOSPR waveguide structure. Analytic and numerical calculations demonstrate the multilayer stack of prism/TiN/SiO 2 /Y 3 Fe 5 O 12 (YIG)/Ce 1 Y 2 Fe 5 O 12 (CeYIG)/Au has six fold higher figure of merit than Au/Co/Au structure at the wavelength of 650 nm. An optimized thickness of CeYIG 45 nm and Au 10 nm with a FOM of xxx is achieved for high sensitivity index sensing. A substantial enhancement of transverse kerr effect (TMOKE) and the index sensitivity was experimentally verified for the proposed device.

Existing noise metrics and modeling for noise induced hearing loss (NIHL) have limitations on prediction of gradually developing NIHL (GDHL) caused by high-level occupational noise. In this study, two auditory fatigue models have been proposed for prediction of GDHL in human. The generalized mammalian auditory model incorporating dual resonant nonlinear (DRNL) filter is used to obtain the velocity distribution on basilar membrane (BM). The velocities of BM as the input loads quantitatively reflect the mechanical fatigue failure of organ of Corti. In addition, experimental human hearing loss data are used to validate the effectiveness of the proposed auditory fatigue models. The regression analysis results show that both auditory fatigue models demonstrate high correlations with human hearing loss data.

Noise induced hearing loss (NIHL) as one of major avoidable occupational related health issues has been studied for decades. In order to accurately evaluate NIHL, several metrics that are based on the response of human auditory system have been developed. Excitation pattern (EP), as the responses of basilar membrane (BM) in cochlea, is correlated with auditory fatigue, which is considered as one fundamental mechanism of NIHL. Therefore, research on understanding EP of BM and its effect on human hearing loss is needed. In this study, two EP based models, referring as loudness EP and velocity EP, are proposed based on applying two typical auditory filters, rounded-exponential (ROEX) filter and dual resonant nonlinear (DRNL) filter, respectively. Moreover, two EP based noise metrics are developed to evaluate the hazardous caused by different types of noises (i.e., Gaussian and impact noises). The results show that both EP models can reflect the BM response to the input noise. For the assessment of Gaussian noise, two EP models are comparable, whereas the velocity EP is more accurate than the loudness EP on analysis of impact noise. The loudness EP may underestimate the hazardous caused by impact noise at high frequency.

Kidney stone is among the common painful disorders of the urinary system. The presence of stones in the kidney refers to Nephrolithiasis. For nephrolithiasis and kidney stones detection, a C-arm tomographic technique was investigated in this paper to generate three dimensional kidney structural information. A series of two dimensional (2D) x-ray projection images were acquired by moving the x-ray tube to different positions around the object over limited angular view with C-arm gantry. The results of our experiments were obtained by using a kidney phantom with low radiation dosage. Reconstruction algorithms based on distance driven (DD) method were developed for C-arm tomographic technique. Computer simulation study was done with simulated tomographic project images of spherical objects to evaluate the performance of each reconstruction algorithm. Preliminary results showed the capability to generate volume information of the kidney for nephrolithiasis and kidney stone detection.

In this article, evaluation index system was constructed based on logistics capability by AHP method and coefficient of variation. Principle analysis about the supply chain logistics was carried out. The system can be used as reference of assessing supply chain logistics revenue. First of all, looking for the best performance of the supply chain logistics efficiency index between logistics and supply chain logistics capability, mode of action, mechanism of action, the core content by AHP method. That is the logistics capability of supply chain logistics. Secondly, in the experimental simulation part, weights about capacity of logistics elements and logistics operation were calculated by coefficient of variation. Logistics capability will be eventually as an evaluation index of supply chain logistics assessment system that maximize efficiency.

The rapid development of computer networks, the time of the entire logistics process substantially reduced, and reduced logistics costs. For practical research in computer network technology in supply chain logistics, this article as a starting point of four indicators, namely computer network technology in the construction of logistics information, the positive role of computer network technology logistics information technology, computer network technology in modern logistics application trends, improve the level of information logistics business logistics enterprises start to actively promote weight logistics enterprise information management.

Nephrolithiasis can be a painful problem due to presence of kidney stones. In this paper, we investigated a C-arm tomographic technique with digital detector for nephrolithiasis and kidney stones detection. The studies were based on a C-arm system to provide two-dimensional (2D) x-ray projection images with a single sweep of both x-ray source and detector over limited angular view that is less than 180°. Our experiments were done with kidney phantom and low radiation dosage. Datasets of 21 two-dimensional projection views of a kidney phantom with embedded kidney stones over view angle of 40° were acquired. Tomographic reconstruction methods were developed for C-arm technique. Computer simulation was investigated to compare results in both spatial and frequency domains. Preliminary results demonstrated that the capability of our C-arm tomographic technique to generate volume information for nephrolithiasis and kidney stone detection. As a traditional three-dimensional (3D) nephrolithiasis and kidney stones detection technique, computed tomography (CT) provides detailed cross-sectional images as well as 3D structure of kidney from moving the x-ray beam in a circle around the body. However, CT scans of the kidney may include risks of allergy to contrast dye. It also includes relatively higher exposure to radiation which is more than regular x-rays. Our C-arm tomographic technique has great potentials to provide detailed 3D information, especially for patients who need to get x-ray scanning repletion after undergoing extracorporeal shockwave lithotripsy to track stone migration and fragmentation to provide the size and the location of kidney stones with low exposure of radiation compare with computed tomography (CT) scans.

Statistical relationships between human vision and natural scene statistics have been addressed. Contrast is fundamental image property. We proposed that infant visual exploration and learning development could be analyzed using contrast entropy minimization. Ten infants viewed 16 naturalistic images. Fixations of the infants were recorded. Infant gaze patterns are compared between individual infants. Infants are compared to the fixations predicted by the contrast entropy minimization. The results show that infant tends to have different gaze patterns, and infant entropy drops gradually as fixations are made.

In this project, we present a data acquisition and analysis study of impulse noise based on wavelet transform for military applications. Impulse noise is a type of highly transient signal widely experienced in military fields (e.g., an intense blast wave). The wavelet transform has been used to analyze signals of impact noise and vibrations, and it showed superior advantages on analysis of transient signals compared to the fast Fourier transform and the short-time Fourier transform. This study focuses on analysis of A-wave type impulse noise in the T-F domain using the continual wavelet transforms. Three different wavelets (i.e., Morlet, Mexican hat, and Meyer wavelets) were investigated and compared based on theoretical analysis and applications to experimental generated impulse noise signals. The underlying theory of continual wavelet transform was given and the temporal and spectral resolutions of different wavelets were theoretically analyzed. The results on singularity detection of the impulse noise showed the Mexican hat wavelet could better reflect the signal oscillations. Furthermore, the similarity of signals between the impulse noise and wavelets functions was investigated in time and frequency domain. The results showed the waveform of Mexican hat wavelets is more similar to the impulse noise signal than the other two wavelets. In summary, although all of three wavelets can represent detailed features of impulse noise in the T-F domain, the Mexican hat wavelets show obvious advantages over the Morlet and Meyer wavelets. The results of this study provide a possible strategy to design special wavelets for impulse noise detection and analysis.

This paper represents our recent experimental measurement study of the complex noise in industrial fields, using a novel acoustic detection system and wavelet transform algorithms. Noise induced hearing loss (NIHL) continues to be one of the most prevalent occupational hazards in the United States. Number of research on NIHL showed a complex noise could produce more hearing loss than an energy-equivalent continuous or impulsive noise alone. Many workplaces in varied industries are subjected to the high level complex noise (i.e., high-level impulsive noise mixed with continuous Gaussian noise). The current noise measurement guidelines and devices (e.g., conventional sound level meters) are based on the equal energy hypothesis (EEH), which states that loss of hearing by exposure to noise is proportional to the total acoustic energy of the exposure. However, the EEH does not accurately rate the impulsive noise and the complex noise. Therefore, the conventional sound level meter may not be able to accurately assess the complex noise in industrial fields. In this project, a new waveform profile based noise measurement system has been developed for evaluation of the high level complex noise in industrial fields. The system consists of four ½″ condenser microphones, and it can simultaneously detect and record four waveforms of the complex noise with high sampling rate (125 KHz). In addition, a wavelet transform based signal analysis algorithm has been modified and implemented to characterize the complex noise. Pilot field measurements have been conducted in selected local coal mining fields (e.g., wet coal preparation plant and dry coal handling plant) using the developed system. The preliminary results showed that the system successfully detected and recorded waveforms of complex noise in industrial fields. The modified algorithm can decomposed the complex noise signals and display the detailed features in the time-frequency joint domain. The key parameters of complex noise can be determined, and the hazardous complex noise in industrial fields can be identified. In addition, when measuring the equivalent A-weighted averaged sound pressure level, the developed system is comparable to a conventional sound level meter.

When a vehicle is running, its measured speed can reflect the operation status of the vehicle sufficiently. In practice, therefore, it is very important to estimate the speed. In this paper, FPGA is used to realize full hardware vehicle speed estimator. The longitudinal velocity of the vehicle can be achieved by a Kalman estimator, which is described by Verilog language. The output data from the eight degrees of freedom vehicle model are regarded as input data of the function simulation in ModelSim. The single precision floating-point binary number is adopted to expand the data range and improve data accuracy. The curves related to the actual speed and estimated speed are analyzed in Matlab, and various performance indicators of the estimator can be checked.

This paper presents a secure application model based on trust algorithm, called T-SAM, designed to establish a reasonable secure application infrastructure, dynamically support comprehensive reasonable assessment for the trust level of services, choose out of secure services from the many services in order to achieve the uniform management and the security services of the application. T-SAM meets the end users individual application requirement", " effectively inhibits the malicious service and constructs the complete secure application model based on trust algorithm.

This research categorized U-turn at Signalized Intersections into mainly four modes, designating them as Mode A,B,C,D: Mode A, at the Downstream Section of the Stop Line; Mode B, at Upstream Section of the Stop Line; Mode C, at the Right Side of the Ingress; Mode D, at the Upstream Section of the Transition Zone. We first evaluate their respective efficiency level, and then set up an efficiency evaluation index system. Three indexes, saturation of left turn lane, average queuing length and average delay of signal control, are standardized by modeling and simulating. The AHP method is then applied to study the integrated index for U-turn evaluation on efficiency. After theoretical arithmetic and simulating, the result shows that Mode D shows a significant advantage considering efficiency factors in each condition. This research may support urban traffic design both theoretically and technically.

Cracks of hydraulic turbine unit are very dangerous to the safety of power station. The crack signals are usually contaminated by strong background noise. Thus extraction of the crack signals is studied in the paper. Independent component analysis (ICA) technique is used to extract the crack signal mixed with white Gaussian noise and operating background noise of the hydraulic turbine unit. It concludes that the method can separate the wanted signal. The denoising result is hardly affected by input signal-to-noise ratios (SNRs) and frequency ranges of signals. Furthermore, the extracted signal has small nuances with the real signal in wave shape. As a result, ICA technique is a better method to extract weak signals.

In order to acquire more accurate crop yield information, the global optimization algorithm SCE-UA was used to integrate leaf area index derived from remote sensing with crop growth model EPIC to simulate regional summer maize yield and field management information in Huanghuaihai Plain in China. The results showed that the mean relative error of estimated summer maize yield was 4.37% and RMSE was 0.44t/ha. Compared with the actual field observation data, the mean relative error of simulated sowing date, plant density and net nitrogen fertilization application rate was 1.85%, −7.78% and −10.60% respectively. These above simulated results could meet need of accuracy of crop growth simulation and yield estimation at regional scale. It was proved that integrating remotely sensed LAI with EPIC model based on SCE-UA for simulating regional summer maize yield and field management information was feasible and reliable.

Assimilating external data into crop growth model to improve accuracy of crop growth monitoring and yield estimation has been being a research hotspot in recent years. In this paper, the global optimization algorithm SCE-UA (Shuffled Complex Evolution method - University of Arizona) was used to integrate remotely sensed leaf area index (LAI) with crop growth model EPIC to simulate regional yield, sowing date, plant density and net nitrogen fertilizer application rate of summer maize in Huanghuaihai Plain. The final results showed that average relative error of estimated summer maize yield was 4.37% and RMSE was 0.44t/ha. Meanwhile, compared with actual observation and investigation data, average relative error of simulated sowing date, plant density and net N fertilization application rate was 1.85%, −7.78% and −10.60% respectively. These above accuracy of simulated results could meet the need of crop monitoring at regional scale. It was proved that integrating remotely sensed LAI with EPIC model based on global optimization algorithm SCE-UA for simulation of crop growth condition and crop yield was feasible.

As the intensification of global competition and the complexity of manufacturing products, enterprise manufacturing information change more and more complex. To effective manage complex manufacturing information in product manufacturing process, based on holonic theory, the manufacturing information integration management framework based on holonic is established. By means of the characteristic analysis of this framework, the manufacturing information cooperation mechanism based on holonic and the manufacturing information communication mechanism based on holonic can be constructed for effectively share and transfer complex manufacturing information. Finally, the simulation system of the manufacturing information integration management framework based on holonic can be demonstrated and validated by Flexsim software. It shows the proposed method can satisfy the requirements of complex manufacturing information integration management in modern manufacturing mode, and it can provide a support for effectively quality management.