Search

Labeling of the connected components is the key operation of the target recognition and segmentation in remote sensing images. The conventional connected-component labeling (CCL) algorithms for ordinary optical images are considered time-consuming in processing the remote sensing images because of the larger size. A dynamic run-length based CCL algorithm (DyRLC) is proposed in this paper for the large size, big granularity sparse remote sensing image, such as space debris images and ship images. In addition, the equivalence matrix method is proposed to help design the pre-processing method to accelerate the equivalence labels resolving. The result shows our algorithm outperforms 22.86% on execution time than the other algorithms in space debris image dataset. The proposed algorithm also can be implemented on the field programming logical array (FPGA) to enable the realization of the real-time processing on-board.

Many recent works have been proposed for face image editing by leveraging the latent space of pretrained GANs. However, few attempts have been made to directly apply them to videos, because 1) they do not guarantee temporal consistency, 2) their application is limited by their processing speed on videos, and 3) they cannot accurately encode details of face motion and expression. To this end, we propose a novel network to encode face videos into the latent space of StyleGAN for semantic face video manipulation. Based on the vision transformer, our network reuses the high-resolution portion of the latent vector to enforce temporal consistency. To capture subtle face motions and expressions, we design novel losses that involve sparse facial landmarks and dense 3D face mesh. We have thoroughly evaluated our approach and successfully demonstrated its application to various face video manipulations. Particularly, we propose a novel network for pose/expression control in a 3D coordinate system. Both qualitative and quantitative results have shown that our approach can significantly outperform existing single image methods, while achieving real-time (66 fps) speed.

Optical remote sensing ship target detection has become an essential means of ocean supervision, coastal defense, and frontier defense. Accurate, effective, fast, and real-time remote sensing data processing is the critical technology in this field. This paper proposes a real-time detection algorithm for moving targets in low-resolution wide-area remote sensing images, which includes four steps: pre-screening, simplified HOG feature identification, sequence correlation identification, and facilitated Yolo identification. It can effectively detect and track targets in low-resolution sequence data. Firstly, iterative morphological processing was used to improve the contrast of low-resolution ship target profile edge features compared with the sea surface background. Next, the target area after adaptive segmentation was used to eliminate false alarms. As a result, the invalid background information of extensive comprehensive data was quickly eliminated. Then, support vector machine classification of S-HOG feature was carried out for suspected targets, and interference such as islands and reefs, broken clouds, and waves were eliminated according to the shape characteristics of ship targets. The method of multi-frame data association and searching for adjacent target information between frames was adopted to eliminate the interference of static targets and broken clouds with similar contours. Finally, the sequential marks were further trained and learned, and further false alarm elimination was completed based on the clipped Yolo network. Compared with the traditional Yolo Tiny V2/V3 series network, this method had higher computational speed and better detection performance. The F1 number of detection results was increased by 3%, and the calculation time was reduced by 66%.

The design of reconfigurable multi-butterfly parallel radix-r FFT (Fast Fourier Transform) processors is proposed. FFT is widely used in signal processing, and the application needs real-time and high performance, while most of the traditional designs are limited to the power of two, which wastes the buffers and multipliers in big data. In response to the problem, we improve the parallel FFT algorithm with the design of reconfigurable control machine combined with buffer/multiplier, and the cost function with the input of radix/number/paddling number/time consuming is deduced. Constrained with the number of buffer and multipliers, the radix and number can be computed with the optimum cost function, and the resolution space of computing performance and hardware cost is presented. The proposed guarantees the real-time performance with better flexibility compared with the previous literature, and the comparison also suggests the effectiveness of the design.

This paper designs and implements an IP that supports the CAN 2.0 protocol and integrates it into the designed SoC. The upper layer software can be used to control and manage the CAN protocol. The SoC that we designed in this paper adopts AHB and APB bus architecture, which can easily integrate the IP of this bus. The processor core uses a five-stage pipeline to support the ArmV4 instruction set architecture. The multiplication unit uses pipeline processing. The forwarding unit and scoreboard unit module can solve and alleviate data dependence and improve instruction execution efficiency. The design has been verified by simulation and its function is correct. The comprehensive results of the target device Spartan-7 show that the design consumes less resources and can be implemented in FPGA. The design can be used in satellite-based integrated electronic solutions of small satellite to complete data transmission and information exchange between spacecraft electronic equipment.

In this paper, a CNN weight parameter quantization method which is suitable for FPGA has designed. By making the weight parameter logarithm based on 2, the multiplication of convolution is simplified to shift, which is easy to be realized in FPGA. Compared with the traditional direct quantization method, the quantization efficiency of the logarithm quantization method designed is improved greatly. If the bit width of the traditional quantization method is N that the quantization bit width of the log quantization method is changed to ceil(log2(n-1))+1,and the processing delay of the logarithm quantization method is better than that of the direct quantization method, especially in the case of small scale bit width processing. The advantages of occupying LUTs, FFs, BRAMs and DSP48es etc hardware resources are obvious, and the processing accuracy is similar to that of the direct quantization method, which is suitable for large scale parallel accelerated operation.

The existing navigation systems have lots of shortcomings, the lack of cross-line resolution lead to the low precision of terrain-aided navigation(TAN) systems. Satellite signals access is necessary for Global Navigation Satellite System (GNSS). Scene matching-based navigation systems require optical image matching, without all-day or all-weather features, and these navigation systems are also affected by seasonal changes. Synthetic aperture radar (SAR) systems cannot inverse platform attitude without three-dimensional localization. Combined with the TAN and scene matching-based navigation systems, a method of InSAR/INS-integrated navigation system based on the interferograms matching is proposed to solve the above problems. The proposed method will employ fundamental interferograms characteristics, which are sensitive to terrain and platform attitude, to provide highprecision navigation. InSAR/INS-integrated navigation system is an autonomous navigation system with high accuracy, and the efficiency, robustness, and reliability of this system are improved compared to the traditional navigation systems.

Ship detection in harbor region is an important aspect in Synthetic Aperture Radar (SAR) marine application research. Fast and accurate ship detection can improve the ability of automatic SAR images interpretation greatly. In this article, a method is proposed for ship detection in harbor region of SAR images based on sea-land segmentation. Firstly, the sea-land segmentation is realized based on SAR image, including binary operation, image dilation and filtering operation. Then, the CFAR ship detection algorithm is taken, including CFAR modeling, the calculation of detection threshold and ship feature analysis. Finally, the ship detection is realized in the detection area. In the experiment with the high resolution RADARSAT-1 image of Vancouver and three other TerraSAR images. Therefore, the experiment result shows that the algorithm we introduced is valid and accurate.

Aiming at the characteristics of target sparsity of remote sensing satellite image, the method for extracting and identifying the target of wide area remote sensing image on orbit is proposed, which is mainly realized by the technique of morphological matching & deep learning. The visual enhancement technology is used to extract the suspected target slices quickly and reduce the amount of processed data greatly, and the deep learning method is adopted to reduce the extraction false alarm rate greatly and realize the recognition and fineness of the target. In the system function and performance verification test, real-time and correct rate indicators were tested on 487 slices of GF-2 satellites with more than 64 scenes. The system can extract and identify 60 targets that the length is more than 100 meters per second. The extraction method based on traditional morphological matching targets has an efficiency of no more than 50%, and the recall rate reaches over 90%. The target classification and identification method based on deep learning increases the precision to over 95%. The system has high integration, excellent real-time performance and expansible, which can meet the real-time processing requirements of current remote sensing satellite payload.

This paper presents a lightweight IEEE 1149.5 module test and maintenance bus slave module interface core, by which we can access the on-chip buses to get the status information and error log. Our slave module interface core supports core class commands and read-write class command, which can be used to access the different ports. In this work, we utilize the streamlined design and efficient implementation, so we just need less resource on a FPGA. Our design has been simulated in Modelsim and verified on a Xilinx Virtex-5 device.

Two phase flow systems have been widely used in industrial engineering. Phase distribution characteristics are vital to the safety operation and optimization design of two phase flow systems. Lift force has been known as perpendicular to the bubbles’ moving direction, which is one of the mechanisms of interfacial momentum transfer. While most widely used lift force correlations, such as the correlation of Tomiyama et al. (2002), were obtained by experimentally tracking single bubble trajectories in high viscosity glycerol–water mixture, the applicability of these models into low viscosity fluid, such as water in nuclear engineering system, needs to be further evaluated. In the present paper, bubble in low viscosity fluid in shear flow was investigated in a full 3D numerical simulation and the volume of fluid (VOF) method was applied to capture the interface. The fluid parameter: fluid viscosity, bubble parameter: diameter and external flow parameters: shear stress magnitude and liquid velocity were examined. Comparing with bubble in high viscosity shear flow and bubble in low viscosity still flow, relative large bubble in low viscosity shear flow keep an oscillation way towards the moving wall and experienced a shape deformation process. The oscillation amplitude increased as the viscosity of fluid decreased. Small bubble migrated to the static wall in a line with larger migration velocity than that in high viscosity fluid and no deformation occurred. The shear stress triggered the oscillation behaviors while it had no direct influence with the behavior. The liquid velocity had no effect on lift coefficient in the present research range of 0 < V < 0.186 m/s.

A two-dimensional temporal and spatial correlated lognormal clutter sequence generation method based on FPGA is proposed. This method solves the problem of long generation time generating the sequence based on the DSP, due to the higher sampling rates or long data, which is difficult to meet real-time requirements of the data generated‥ In this paper, based on FPGA, using a one-dimensional temporal correlated Lognormal clutter sequence and one-dimensional Gaussian random sequence, using a single precision floating-point matrix multiplier and adders and multipliers of FPGA IP core, methods are proposed to generate two-dimensional temporal spatial correlated lognormal clutter sequence. One single precision floating-point matrix multiplier includes an input data buffer and output data buffer, a single-precision floating-point adder, a single-precision floating-point multiplier and data access control module. Two-dimensional temporal and spatial correlated lognormal clutter generated each step employs parallel processing method, the method avoids the problems of existing methods that instruction cycles execute serially and meet the requirements of real-time data generation.

Currently software defined radio (SDR) is implemented by using DSP only or by using DSP plus FPGA chip. For DSP processing speed of the bottleneck problem, a complete FPGA IF software radio implementation is proposed in this paper. Traditional solutions use only FPGA for digital up and down conversion, modulation and demodulation; design proposed in this paper based on the FPGA chip, a CPU processor for SDR is designed. By using this processor, the full realization of all digital signal processing algorithms on software radio is implemented by FGPA only. Because of the full use of FPGA to implement software radio, so that the system power consumption is smaller, to solve the needs of software radio for defined power. In this paper, based on the FPGA chip CPU design method, the proposed greatly reduces the difficulty of FPGA development; the use of C language and assembly language to develop digital signal processing on FPGA platform, greatly reducing system development time.

During severe accident in the nuclear power plant, a considerable amount of hydrogen can be generated by an active reaction of the fuel-cladding with steam within the pressure vessel which may be released into the containment of nuclear power plant. Hydrogen combustion may occur where there is sufficient oxygen, and the hydrogen release rates exceed 10% of the containment. During hydrogen combustion, detonation force and short term pressure may be produced. The production of these gas species can be detrimental to the structural integrity of the safety systems of the reactor and the containment. In 1979, the Three Mile Island (1979) accident occurred. This accident compelled experts and researchers to focus on the study of distribution of hydrogen inside the containment of nuclear power plant. However after the Fukushima Dai-ichi nuclear power plant accident (2011), the modeling of the gas behavior became important topic for scientists. For the stable and normal operation of the containment, it is essential to understand the behavior of hydrogen inside the containment of nuclear power plant in order to mitigate the occurrence of these types of accidents in the future. For this purpose, it is important to identify how burnable hydrogen clouds are produced in the containment of nuclear power plant. The combustion of hydrogen may occur in different modes based on geometrical complexity and gas composition. Reliable turbulence models must be used in order to obtain an accurate estimation of the concentration distribution as a function of time and other physical phenomena of the gas mixture. In this study, a small scale hydrogen-dispersion case is selected as a benchmark to address turbulence models. The computations are performed using HYDRAGON code developed by Department of Engineering Physics, Tsinghua University, China. HYDRAGON code is a three dimensional thermal-hydraulics analysis code. The purpose of this code is to predict the behavior of hydrogen gas and multiple gas species inside the containment of nuclear power plant during severe accident. This code mainly adopts CFD models and structural correlations used for wall flow resistance instead of using boundary layer at a wall. HYDROGAN code analyzes many processes such as hydrogen diffusion condensation, combustion, gas stratification, evaporation, mixing process. The main purpose of this research is to study the influence of turbulence models to the concentration distribution and to demonstrate the code thermal-hydraulic simulation capability during nuclear power plant accident. The calculated results of various turbulence models have different prediction values in different compartments. The results of k–ε turbulence model are in reasonable agreement as compared to the benchmark experimental data.

The paper presents the development of a sub-channel thermal-hydraulic analysis code named SUBCHAN. The code was originally developed to analyze super critical CANDU type reactor which has such characters as horizontal fuel channels, heavy water moderated, super critical light water cooled, any type of fuel bundle with or without thorium rods. Thermal-hydraulic model of SUBCHAN is based on four partial differential equations that describe the conservation of mass, energy and momentum vector in axial and lateral directions for the water liquid/vapor mixture. The heat transfer correlations and pressure drop correlations used in the SUBCHAN code are presented in this paper. The water properties package of the code is based on the Industrial Formulation 1997 for the Thermodynamic Properties of Water and Steam. The heat transfer correlation of super critical region is based on the experimental investigation of Xi'an Jiaotong University and Dyadyakin and Popov correlation. By calculating the TACR case, which is operating at 12.5 MPa pressure, compared with the results of ASSERT-PV code, the paper gets the conclusion that the development of the SUBCHAN code with super critical water property package is successful. Then the paper uses the SUBCHAN code to analyze CANDU-SCWR which is operating at 25.0 MPa pressure. The paper gets the conclusion that the SUBCHAN code can be used to analyze the steady state sub-channel thermal-hydraulic analysis of CANDU-SCWR fuel channel.

This paper presents a thorium fuel cycle scheme for the thorium-based advancednuclear energy system JANES). TANES that will use thorium resources is a homemade nuclear energy system that is fit for the situation of China and some other thorium-rich regions. By the reactor physics and fuel cycle analysis, the paper gets the conclusion that if TANES decreases lattice pitch (channel pitch) and increases driver fuel enrichment, TANES will provide a slightly negative coolant void reactivity coefficient to facilitate implementation of advanced automatic control technology.

Genome-wide RNAi screening has emerged as a powerful tool for loss-of-function studies that may lead to therapeutic target discovery for human malignancies in the era of personalized medicine. However, due to high false-positive and false-negative rates arising from noise of high-throughput measurements and off-target effects, powerful computational tools and additional knowledge are much needed to analyze and complement it. Availability of high-throughput genomic data including gene expression profiles, copy number variations from large-sampled primary patients and cell lines allows us to tackle underlying drivers causally associated with tumorigenesis or drug-resistance. In my dissertation, I have developed a framework to integrate functional RNAi screens with systems biology of cancer genomics to tailor potential therapeutics for reversal of drug-resistance or treatment of aggressive tumors. I developed a series of algorithms and tools to deconvolute, QC and post-analyze high-throughput shRNA screening data by next-generation sequencing technology (shSeq), particularly a novel Bayesian hierarchical modeling approach to integrate multiple shRNAs targeting the same gene, which outperforms existing methods. In parallel, I developed a systems biology algorithm, NetBID2, to infer disease drivers from high-throughput genomic data by reverse-engineering network and Bayesian inference, which is able to detect hidden drivers that traditional methods fail to find. Integrating NetBID2 with functional RNAi screens, I have identified known and novel driver-type therapeutic targets in various disease contexts. For example, I discovered that AKT1 is a driver for glucocorticoid (GC) resistance, a problem in the treatment of T-ALL. The inhibition of AKT1 was validated to reverse GC-resistance. Additionally, upon silencing predicted master regulators of GC resistance with shRNA screens, 13 out of 16 were validated to significantly overcome resistance. In breast cancer, I discovered that STAT3 is required for transformation of HER2+ breast cancer, an aggressive breast tumor subtype. The suppression of STAT3 was confirmed in vitro and in vivo to be an effective therapy for HER2+ breast cancer. Moreover, my analysis revealed that STAT3 silencing only works in ER- cases. Using my framework, I have also identified potential therapeutic targets for ABC or GCB-type DLBCL and subtype-based breast cancer that are currently being validated.