1. Topics of the Workshop

This workshop will focus on the joint research work on basic theory, key technologies and application development in optoelectronic characteristics of targets and environment, optoelectronic detection technology and optoelectronic information and image processing areas.

■ Optoelectronic Characteristics of Targets and Environment: wave propagation characteristics under complex environment, optical characteristics of targets and environment, the electromagnetic scattering characteristics of targets and environment.

■ Optoelectronic Detection Technology: infrared detection technology, multi/high-spectral detection technology and laser radar detection technology.

■ Optoelectronic Information and Image Processing: infrared image processing technology, hyperspectral imagery processing and optoelectronic information processing.

2. Workshop Committee

Sponsor

School of Physics and Optoelectronic Engineering, Xidian University (XDU), China

Co-organizer

Office of International Cooperation & Exchange, Xidian University (XDU), China

Workshop Chairman

Chein-I Chang, IEEE Life Fellow, Fellow of SPIE. Professor of Department of Computer Science and Electrical Engineering at theUniversity of Maryland, Baltimore County (UMBC)

Guo Lixin, Director of International Academic Platform for Optoelectronic Perception Science and Technology under Complex Environment (OPCE), Xidian University (XDU), China

3. Workshop date and Venue

June 14th-17th in theAcademic Report Hall in the Third Floor of annex building of Library, North Campus of Xidian University

4. Invited Talks

(1). Prof. Chein-I Chang：IEEE Life Fellow, Fellow of SPIE. Professor with Department of Computer Science and Electrical Engineering at the University of Maryland, USA

Talk Title: Hyperspectral Image Classification: A Statistical Detection Theory Approach

Abstract: This talk presents a statistical detection theory approach to hyperspectral image (HSI) classification which is quite different from many conventional approaches reported in the HSI classification literature. It interprets a multi-class classification problem as a multi-target detection problem in such a way that the well-established statistical detection theory can be readily applicable to solving classification problems. In particular, it introduces two types of classification, a priori classification and a posteriori classification, which can be considered as counterparts of Bayes detection and maximum a posteriori (MAP) detection respectively in detection theory. Accordingly, detection probability and false alarm probability can be also translated to classification rate and false classification rate derived from a confusion classification matrix for classification. To evaluate the effectiveness of a posteriori classification a new a posteriori classification measure, to be called precision rate (PR), is also introduced by MAP classification in contrast to overall accurate (OA) that has been used for Bayes classification. The experimental results provide evidence that a priori classifier as Bayes classifier which performs well in terms of OA does not necessarily perform well as a posteriori classifier in terms of PR. That is, PR is the only criterion that can be used as a posteriori classification measure to evaluate how well a classifier performs.

(2). Prof. Tapan K. Sarkar, IEEE Life Fellow, Syracuse University, USA

Talk Title: Generation of Non-Minimum Phase Response of Electromagnetic Systems Using Amplitude only Data

Abstract: The objective of this paper is to illustrate how to generate the nonminimum phase response of an electromagnetic system from amplitude only data. It is important to note that the phase reconstruction from amplitude only data yields a nonunique solution as a linear phase term can be added to the system response without distorting its amplitude response since the linear phase term only adds a delay.
Reconstruction of phase from amplitude-only data is an important problem. For minimum phase systems, the reconstruction of phase from amplitude-only data is relatively straight forward as the phase response is given by the Hilbert transform of the log of the magnitude (termed cepstrum) of the amplitude. If the system is not minimum phase (i.e., when some of the zeros of the transfer function may be on the right half-plane), then the above integral relationship does not hold. Most electromagnetic systems have a nonminimum phase response. Hence, this methodology developed for acoustic systems where almost all systems are minimum phase have very little use for the practical electromagnetic problems. However, there is a more general result of the Hilbert transform, which is based on causality. This latter result is valid for any nonminimum phase system. We utilize the principle of causality to carry out a nonminimum phase realizations. The principle of causality implies that the function f(t) = 0 for t < 0, and is nonzero otherwise. It is important at the onset to point out that the phase realization (be it minimum or nonminimum phase) is not a unique problem. A linear-phase term may be added to any phase function without altering its amplitude spectrum. This is because the addition of a linear phase to the phase of the transfer function with a uniform amplitude is equivalent to a pure delay in the time domain. Since we are dealing with linear-shift invariant systems (as the response of the system is the same independent of the time origin), changing the impulse response of the system by a time shift does not alter the transfer function of the original system, except that the phase spectrum is modified by a linear-phase function. The slope of this linear-phase function is equivalent to the time delay. Also, the amplitude spectrum of the transfer function is unaltered by providing a delay to the impulse response of the system at hand.

(3). Prof. Chen Zhining, IEEE Fellow

Talk Title: Metantennas: from Metamaterial Physics to Antenna Technology

Abstract: Metamaterials or Metasurfaces have been one of the hottest research topics in electromagnetic community, in particular, antenna engineering. This talk reports the latest progress in the applications of metasurfaces in 5G and B5G antenna engineering (metantennas in short) at microwave/millimeter-wave bands. First, the new requirements for 5G antennas are introduced and then the concepts of metasurfaces are briefed. After that, the recent progress in the research and development of microwave metantennas is updated with two metasurface-based design examples.

(4). Prof. Roberto Graglia,IEEE Fellow, Politecnico di Torino, DET Department, Italy

Talk Title: High-Order Modeling for Computational Electromagnetics

Abstract: This presentation provides an overview of the last decade advances in computational electromagnetics concerning the development and use of high-order models for Moment Method and Finite Element Method applications. Various two- and three-dimensional high-order divergence- and curl-conforming vector bases used for the solution of differential and integral equations are compared and considered before presenting basis functions of either substitutive or additive kind able to model vertex, edge, and corner singularities. The implementation problems and the advantages provided by use of these higher-order models are discussed in detail thereby presenting several results.

(5). Prof. Shen Zhongxiang, IEEE Fellow, Nanyang Technological University, Singapore.

Talk Title: Radar Cross-Section Enhancement Techniques

Abstract: In this talk, I will first point out a few important applications that require radar cross-section (RCS) enhancement and then briefly review the existing RCS enhancement techniques.  After that, we will consider the RCS enhancement of three objects: a thin plate, cylindrical objects and sphere. Physical principles will be explained for the RCS enhancement of each object. Simulation and measurement results are presented to demonstrate the effectiveness of the proposed RCS enhancement techniques. Some on-going research work on the topics will also be briefly introduced in the end.

(6). Prof. Chung Pau-Choo, IEEE Fellow, Cheng Kung University, Taiwan, China

Talk Title: Deep Learning Models for Medical Image Analysis

Abstract: Recent advancement of image understanding with deep learning neural networks has brought great attraction to those in image analysis into the focus of deep learning networks. While researchers on video/image analysis have jumped on the bandwagon of deep learning networks, medical image analyzers certainly is the coming followers. The characteristics of medical images are extremely different from those of photos and video images. The application of medical image analysis is also much more critical. For achieving the best effectiveness and feasibility of medical image analysis with deep learning approaches, several issues have been taken cared of. In this talk we will give a brief overview of the recent development of deep learning models and their applications in medical image analysis. Several issues in regard of the data preparation, techniques, and clinic applications will also be discussed.

(7). Prof. Ken-ichi Ueda, University of Electro-Communications, Japan

Talk Title: High Power Fiber Laser, History and Future

Abstract: We developed 1kW output fiber laser for the first time in the world in 2002. Before our work, a fiber laser amplifier was recognized as a optical amplifier device for optical communication.  Today, high power fiber laser is the most powerful in the laser processing market.  I introduce a brief history and future potential of high power laser technology.  A fiber laser is only one optical device to produce high output power under the spatial mode control. Future of coherent beam combining technique and photonic bandgap fiber application will be discussed.

(8). Prof. Gerard Gouesbet,Professor of Rouen University and Research Director in CNRS, France

Talk Title: Generalized Lorenz-Mie theories and mechanical effects of laser light

Abstract: On the occasion of Dr. Arthur Ashkin’s receipt of the 2018 Nobel Prize in physics for his pioneering work in optical levitation and manipulation, a story in the review concerns the first experimental validations of generalized Lorenz-Mie theories using optical levitation experiments as well as extended mechanical effects of laser light will be delivered in this talk.

Among the many works of Arthur Ashkin, many have been devoted to optical tweezers, optical levitation and optical manipulation of macroscopic particles (“macroscopic”being here to beunderstood as opposed to atoms or molecules). From a theoretical point of view, these experiments have been studied in the framework of two limiting regimes, namely Rayleigh regime for small size parameter and ray optics for large size parameter. The generalized Lorenz-Mie theory (GLMT, and more generally GLMTs) bridges the gap between these two regimes. The present talk will reviews GLMTs and mechanical effects of laser light, in Rouen where the GLMT had originally been built, but also worldwide.

(9). Prof. Ali A. Eftekharn，Georgia Institute of Technology，USA

Talk Title: Enabling hybrid optoelectronic material platforms

Abstract: In the past few years Si-photonic devices have been rapidly growing as the solution for different applications such as high-speed-optical interconnections and Lidar. Nevertheless, silicon by itself does not provide all of the required optoelectronic functionalities for different application. In this talk, we review alternative material platforms and hybrid material platform and devices based on heterogeneous integration of different materials, which enable functionalities beyond what can be achieved in single-material platforms.

(10). Prof. Zeev Zalevsky, BAR-ILAN University, Israel

Talk Title: Light based Remote Infra through Ultra Sounds Extraction: Remote Photonic Bio-Sensing and Diseases Diagnosis

Abstract: I will present a photonic technological sensing platform that can be used for remote, continuous and simultaneous sensing of many biomedical parameters as well as for serving as highly directional hearing aid device in which sounds’ frequencies from infra all the way through ultra can be sensed (depending on the sampling bandwidth of the used camera). The technology is based upon illuminating a light scattering surface (skin) with a laser and then using a camera with its special optics to perform temporal and spatial tracking of the back scattered secondary speckle patterns in order to have nano-metric accurate estimation of the movement of the back-reflecting surface. If the back-reflecting surface is a skin located close to main blood arteries, then multi-functional bio-medical monitoring can be realized. If the surface is close to our neck or head, then a directional communication channel can be established and used as e.g. an improved in its directionality hearing aid device.
The proposed technology was already applied for remote and continuous estimation of vital signs such as heart beats (rate and sound), respiration (rate and sound) and blood pressure; of hemodynamic sensing (characteristics of blood flow to various tissues in the body including the brain); blood hematology (e.g. estimation of alcohol and glucose concentrations in blood stream as well as blood coagulation and oximetry) and more. Specifically, the proposed sensor was directly applied for diagnosis of glaucoma (via measurement of intra-ocular pressure), breast cancer, melanoma, otitis and injuries in vocal cords.
 

(11). Prof. Yang Jian, Chalmers university of technology, Sweden

Talk Title: Challenges of Antenna Systems for 5G communication and Chalmers solutions

Abstract: 5G era is approaching and we now see some versions of commercial 5G networks starting to go live at different parts of the world. The 5G communication technology will bring new experience and new challenges including wider bandwidth, higher data rate, greater capacity, higher security, and lower latency. The key enabling 5G technology includes novel multiple access strategies, ultra-dense networking, all-spectrum access, massive MIMO, full digital beam forming or hybrid beam forming. This talk will present some Chalmers solutions to the above challenges, including gap waveguide technology and capped Bowtie technology.

(12). Prof. Yoav Y. Schechner, Israel Institute of Technology, Israel

Talk Title: Scattering as Key to Three dimensional Tomography: from Medical Imaging to Spaceborne Atmospheric Sensing

Abstract: Computed tomography (CT) yields 3D imaging and recovery  of volumetric objects.
Traditional CT has been based on linear models, in which scattering was often considered a disturbance. As computers become stronger, they enable solving more complex, nonlinear inverse problems. We leverage this capability, to formulate and solve CT which intentionally includes all scattering events. Scattering then serves as a source of useful information. We show two applications for this approach. In medical X-ray CT, scattering helps to estimate the chemical decomposition per voxel, and significant dose reduction per quality of recovery. The approach further facilitates next-generation medical CT scanners.  Scattering is also dominant in atmospheric radiative transfer of visible light. Hence, we develop scattering-based CT for passive 3D sensing of scatterer distributions in the atmosphere (aerosols, clouds). This approach is planned for an upcoming space mission (CloudCT).

(13). Prof. Gao Yongsheng, Griffith University, Australia

Talk Title: Vision Perception for Driving Farming Productivity

Abstract: Smart farming 4.0 studies new knowledge, technologies and devices for automation in agriculture and aquaculture, early detection of pest and plant disease, automatic species identification, plant phenomics, better water resource management, land environment monitoring, costal environment monitoring, marine life surveillance, etc. In this talk, he will introduce some of their work on automation in agriculture and aquaculture, faster grading and packing, species and cultivar identification, pest and disease recognition at Environmental Informatics @Griffith and ARC Industrial Transformation Research Hub for Driving Farming Productivity and Disease Prevention, including recognition without detection, large image database retrieval (speed vs accuracy), and new advancement of visual classification from species to cultivar.

(14). Prof. David Suter, Edith Cowan University，Australia

Talk Title: Computer Vision from Geometric/Model-based to Deep Learning

Abstract:This talk will give an overview of some aspects of research conducted by the speaker over the last 10 years or so. A large portion of that work concerns robust fitting for model fitting (where the precise parameters of the model are of primary interest) and for segmentation (where the model fit is mainly as a way of breaking the data into meaningful parts). There are a wide range of applications, some of which will be illustrated. Lately, such model-based approaches have been challenged and in some cases supplanted by deep learning approaches - we have some recent work where one tries to leverage “the best of both approaches”. The latter part of the talk will illustrate current projects that span medical imaging and signal processing (e.g., Aortic Calcification assessment, Cardiac Arrhythmia classification) and semantic aspects of computer vision.

(15).Prof. Zhang Yang, Professor at the University of Sheffield, UK

Talk Title: Further investigation of hydrogen flame colour

(16). Prof. Gill Dobbie, University of Auckland, New Zealand

Talk Title: Novel Frameworks for Effective Classification in Data Streams

Abstract: For many applications, it is important to ensure accurate classification in real time. Traditionally, a drift detector would be used to detect a concept drift in a data stream, and a new classifier would be built based on the new data after a drift has been detected. This technique is time consuming, so we looked for smarter ways to build the new classifier.In our recent work, we have described a novel framework that stores classifiers it has built, and considers using one of those rather than building a new classifier overtime a drift occurs. The challenges of this work are describing a measure of similarity of classifiers and providing guarantees around the memory it uses. In the talk, I will describe the framework, how we address the challenges, and further improvements we have made to the framework.

(17). Prof. Ren Kuanfang, First class professor in Rouen University, France

Talk Title: Characterization of large non-spherical objects using Vectorial Complex Ray Model

Abstract: In the Vectorial Complex Ray Model (VCRM), the curvature of the wavefront is introduced as a new property of a ray in the geometrical optics (or ray model in general sense). This property improves considerably the precision of the ray model. The VCRM permits therefore to predict precisely and in fine the scattering diagrams of large objects of any shape with smooth surface. It can be applied to the direct problems such as the calculation of the signal scattered by a objet, or to the inverse problems as in the metrology of particles or other objects.
The presentation will focus on the application of the VCRM in the characterization of large objects. Three typical shapes of object (sphere, spheroid and pendant drop) will be dealt with in details to show the principle and the power of the model. In addition, two free software will also be presented: ABSphere calculates all properties in the interaction of a homogenous or stratified sphere with a shaped beam and VCRMEll2D is fundamental in development of VCRM.

(18). Prof. Koichi Shimizu, Waseda University, Japan

Talk Title: Transillumination imaging of animal body with NIR light  – For medical application of propagation analysis of scattered light

Abstract: Near-infrared light (700-1200 nm wavelength) has relatively high transmission through animal body, and has been used extensively in biomedical applications. Based on the theory of light scattering in random media, we have analyzed the light propagation in an animal body. Here, we report the techniques we developed using the results of the analysis. They are the extraction of near-axis scattered light, the improvement of transcutaneous fluorescent images, and the scattering suppression in transillumination imaging. Their applications are transillumination imaging of veins and arteries, functional transillumination imaging and 3 dimensional transillumination imaging. In addition, new attempts are shown such as the use of the time-reversal and deep-learning principles.

(19). A/Prof Zhou Jun, Griffith University, Australia

Talk Title: Material Based Object Tracking in Hyperspectral Videos

Abstract:Traditional color images only depict color intensities in red, green and blue channels, often making object trackers fail in challenging scenarios, e.g., background clutter and rapid changes of target appearance. Alternatively, material information of targets contained in a large amount of bands of hyperspectral images (HSI) is more robust to these challenging conditions. This talk introduces a comprehensive study on how material information can be utilized to boost object tracking from three aspects: benchmark dataset, material feature representation and material based tracking. In terms of benchmark, we construct a datasetof fully-annotated videos which contain both hyperspectral and color sequences of the same scene. Material information is represented by spectral-spatial histogram of multidimensional gradient, which describes the 3D local spectral-spatial structure in an HSI, and abundances which encode the underlying material distribution. These two types of features are embedded into correlation filters which tracks the movement of objects. Our study shows that object tracking in hyperspectral videos has great potential and advantages.

(20). A/Prof. Abd-Krim Seghouane, University of Melbourne, Australia

Talk Title: Parse Principal Component Analysis with Preserved Sparsity Pattern

Abstract: Principal component analysis (PCA) is widely used for feature extraction and dimension reduction in pattern recognition and data analysis. Despite its popularity, the reduced dimension obtained from PCA is difficult to interpret due to the dense structure of principal loading vectors. To address this issue, several methods have been proposed for sparse PCA, all of which estimate loading vectors with few non-zero elements. However, when more than one principal component is estimated, the associated loading vectors do not possess the same sparsity pattern. Therefore, it becomes difficult to determine a small subset of variables from the original feature space that have the highest contribution in the principal components. To address this issue, an adaptive block sparse PCA method is proposed. The proposed method is guaranteed to obtain the same sparsity pattern across all principal components. Experiments show that applying the proposed sparse PCA method can help improve the performance of feature selection for image processing applications. We further demonstrate that the proposed sparse PCA method can be used to improve the performance of blind source separation for functional magnetic resonance imaging (fMRI) data.

(21). A/Prof. Yuan Zhen，University of Macau, Macau, China

Talk Title: Quantitative Multi-contrast Photoacoustic Tomography

Abstract: Photoacoustic tomography (PAT) is a non-invasive, non-ionizing, and inexpensive monitoring and imaging technique that uses near-infrared light and acoustic measurements to probe tissue optical properties. Regional variations in oxy- and deoxy-hemoglobin concentration as well as blood flow can be imaged by monitoring spatial-temporal variations in the light absorption and acoustic pressure, giving PAT the special ability to directly measure the quantitative hemodynamic, metabolic, optical and mechanical parameters and neuronal responses to cells (neurons), tissues and organs activation with high spatial and temporal resolution as well as good penetration depth. These capabilities make PAT a unique stand-alone imaging tool and useful complement to fMRI, PET and EEG/MEG in studies of normal physiology and pathology. In this talk, Dr. Yuan will mainly talk about his recent research work on the development of novel and quantitative photoacoustic tomography (PAT) imaging techniques, which involves the optical absorption, scattering, physiology, mechanical and molecular contrasts.

(22). Dr. Gu Lin, National Institute of Informatics, Japan

Talk Title: Interpretable Medical Imaging Analysis

Abstract: Though deep learning has shown successful performance in classifying the label and severity stage of certain disease, most of them give few evidence on how to make prediction. Here, we propose to exploit the interpretability of deep learning application in medical diagnosis. Inspired by Koch’s Postulates, a well-known strategy in medical research to identify the property of pathogen, we define a pathological descriptor that can be extracted from the activated neurons of a diabetic retinopathy detector. To visualize the symptom and feature encoded in this descriptor, we propose a GAN based method to synthesize pathological retinal image given the descriptor and a binary vessel segmentation. Besides, with this descriptor, we can arbitrarily manipulate the position and quantity of lesions. As verified by a panel of 5 licensed ophthalmologists, our synthesized images carry the symptoms that are directly related to diabetic retinopathy diagnosis. The panel survey also shows that our generated images is both qualitatively and quantitatively superior to existing methods.

(23). Dr. Patrick Stegmann, Joint Center for Satellite Data Assimilation, USA

Talk Title: An Overview of Light Scattering in the Community Radiative Transfer Model

Abstract: The Community Radiative Transfer Model (CRTM) is a fast, one-dimensional, and scalar radiative transfer (RT) model for MW, IR, and optical sensors in continuous development at the Joint Center for Satellite Data Assimilation (JCSDA) and is a core component of the NOAA/NCEP satellite data analysis system and the novel Joint Effort for Satellite Data Assimilation Integration (JEDI) project of the JSCDA. The CRTM framework is in essence a modular Fortran 90 library that provides different components to represent different aspects of atmospheric radiative transfer, such as gas absorption, particle scattering, and surface optics. An advanced feature of the CRTM is not only the possibility to use it as a Forward Operator to synthesize satellite data, but also its ability to compute Tangent-Linear (TL), Adjoint (AD), and Jacobian (K-Matrix) results for a given atmospheric state. This capability is crucial for satellite data assimilation in modern numerical weather prediction models. With this capability the CRTM represents a strategic asset to all partner agencies of the JCSDA and one of the few other radiative transfer models with similar operational speed and applicability is the Radiative Transfer for TOVS (RTTOV) model curated by the EUMETSAT Numerical Weather Prediction Satellite Application Facility (NWP FAS). Given this background, it is essential to maintain the advantage that CRTM represents in the face of new satellite sensors, new developments in computer infrastructure, such as heterogeneous architectures, and new software frameworks such as JEDI. Light scattering on the radiative transfer side of the CRTM is computed by using the Advanced Adding-Doubling method (AAD) as the default solver. Different kinds of atmospheric particles can be considered in terms of one-dimensional layers, such as precipitating and non-precipitating hydrometeors, and various aerosol classes.This presentation will give an overview of recent advances in the light scattering properties included in the CRTM, specifically updates in the snow and ice cloud properties, as well as future plans to include polarized solvers in the CRTM. This will require providing the full Müller matrices of scattering particles as a database, instead of just the scalar phase function.
 

(24). Dr. Qin Peiyuan, University of Technology Sydney, Australia

Talk Title: Beam Switching Conformal Transmitarray

Abstract: Transmitarray antennas have attracted considerable attention due to their merits of high gain, low profile and flexible radiation performance. For many wireless communication platforms, such as satellites, aircrafts, and unmanned aerial vehicles (UAV), conformal high-gain antennas are needed in order to meet the aerodynamic requirements.  For these applications, conformal transmitarray antennas, which are designed to follow the shapes of various mounting platforms, are highly desired as part of the platform surface, front in particular, can accommodate the transmitarray with a feed placed behind. Unfortunately, most of the current reported array elements are not suitable for conformal transmitarray antennas. In this presentation, a fixed-beam conformal transmitarray using an ultra-thin element is presented first. Then, the capability to mechanically switching its beam is investigated. It is found that by rotating the feed horn to different positions, the main beam of the array can be switched to ±15°, ±10°, ±5° and 0°. More details will be presented in the forum.

5. Registration Fee

Totally Free