This PDF file contains the front matter associated with SPIE Proceedings Volume 9501, including the Title Page, Copyright information, Table of Contents, Introduction (if any), Authors, and Conference Committee listing.

Current satellite on-board processing units usually follow a conservative scheme where simple operations are performed on the acquired pixels. Frequently, only image compression is done on-board, and in most cases, this compression is applied at a constant global bit-rate in order to simplify the storage and the acquisition and download schedules, as highly predictable data volumes are produced.

This constant or fixed rate paradigm represents in fact a huge constraint for image compressors. Firstly, it can be hard to obtain with classical entropy coders, because their variable-length codes naturally produce variable bit-rates. Secondly and more importantly, the same compression ratio must be applied to every image, without being able to take into account its content, its degree of interest or even its entropy.

Moreover, remote sensing imagery has become a crucial instrument in a large number of civil and military applications and then, image-quality requirements are more and more difficult to satisfy because every final user has specific needs. Thus, as with fixed rate compression some image areas are better compressed than others, image-quality assessments must be established based on worst-case analysis, which provides very low compression ratios, even for state-of-the-art compressors.

CNES has been working for the last years in the characterization of image-quality requirements imposed by final users, in order to establish a relationship between the local image characteristics and the associated image quality requirements, or in other words, the tolerated compression losses.

As a result, the new functionalities included in the next generation of CNES image compressors will permit to accurately and locally adjust the compression ratio: the target quality level will be adapted for every area in the image taking into account not only its entropy but also its degree of interest.

This new trend has required the adoption of variable rate compression, which has had a significant impact in other associated elements such as mission scheduling and storage. Other interesting on-board processing techniques have also been introduced in order to fully exploit the capacities of this new kind of compression.

A number of compression methods have been used to compress hyperspectral images. However, these methods may fail to retain all the discriminant characteristics of hyperspectral images since some discriminant features may not be high in signal energy. Also, it has been reported that compression may improve classification performance in some cases. In this paper, we investigate these problems, and analyze potential discriminant information loss and classification performance improvement. We perform some experiments using various compression methods. We examine this phenomenon and its implication.

A hyperspectral image compression method is proposed using an online dictionary learning approach. The online learning mechanism is aimed at utilizing least number of dictionary elements for each hyperspectral image under consideration. In order to meet this “sparsity constraint”, basis pursuit algorithm is used. Hyperspectral imagery from AVIRIS datasets are used for testing purposes. Effects of non-zero dictionary elements on the compression performance are analyzed. Results indicate that, the proposed online dictionary learning algorithm may be utilized for higher data rates, as it performs better in terms of PSNR values, as compared with the state-of-the-art predictive lossy compression schemes.

On the basis of the theory of the biorthogonal invariant set multiwavelets (BISM) which is established by Micchelli and Xu, a biorthogonal invariant set multi-wavelets (BISM) filter is designed and the algorithms of decomposition and reconstruction of this filter are given in this paper, and it has many characteristics, such as symmetry, compact support, orthogonality and low complexity. In this filter, the self-affine triangle domain is as support interval, and constant function is as scaling function. Advantages such as low algorithm complexity, the energy and entropy in high concentration after transformation, no blocking effect to facilitate parallel computing are analyzed when the biorthogonal invariant sets multiwavelet (BISM) filters are used image compression. Finally, the validity of image compression algorithm based on biorthogonal invariant set multiwavelet is verified by the approximate JPEG2000 framework.

In this study, a hardware efficient field-programmable-gate-array (FPGA) implementation of the JPEG-LS encoder for lossless image compression is introduced. Encoder architecture comprises both regular mode and run mode with run interruption sample encoding procedures for full compliance with the ISO/ITU standard. Differently from former reported implementations, prediction error computation is optimized with pipeline data forwarding technique for optimum delay and minimum complexity. Besides, procedures of the run-length encoding are realized with normalization scheme using look-up tables without update latency. Synthesis results showed that proposed optimizations improved the processing speed of the encoder noticeably while FPGA hardware footprint is significantly reduced.

Using maximal simplex volume as an optimal criterion for finding endmembers is a common approach and has been widely studied in the literature. Interestingly, very little work has been reported on how simplex volume is calculated. It turns out that the issue of calculating simplex volume is much more complicated and involved than what we may think. This paper investigates this issue from two different aspects, geometric structure and eigen-analysis. The geometric structure is derived from its simplex structure whose volume can be calculated by multiplying its base with its height. On the other hand, eigen-analysis takes advantage of the Cayley-Menger determinant to calculate the simplex volume. The major issue of this approach is that when the matrix is ill-rank where determinant is desired. To deal with this problem two methods are generally considered. One is to perform data dimensionality reduction to make the matrix to be of full rank. The drawback of this method is that the original volume has been shrunk and the found volume of a dimensionality-reduced simplex is not the real original simplex volume. Another is to use singular value decomposition (SVD) to find singular values for calculating simplex volume. The dilemma of this method is its instability in numerical calculations. This paper explores all of these three methods in simplex volume calculation. Experimental results show that geometric structure-based method yields the most reliable simplex volume.

A new effective image super resolution (SR) algorithm which is a hybrid of multiple frame Variational Bayesian (VB) reconstruction and single frame Dictionary Learning (DL) reconstruction method is developed to reconstruct a high resolution (HR) satellite image in this article. Firstly, by employing a variational Bayesian analysis, the unknown high resolution image, the acquisition process, the motion parameters and the unknown model parameters are built together in a single mathematical model with a Bayesian formula, and then the distributions of all unknowns are jointly estimated. Without any parameter adjustment, an HR image is adaptively reconstructed from multiple frame low resolution (LR) images. Secondly, by taking the above HR image as input, a higher resolution image can be rebuilt utilizing the statistical correlation between the HR and LR images which is obtained via the DL method. The VB method effectively uses non-redundant information between LR images to recover HR satellite images. Benefit from the dictionary training of magnanimity image, the DL algorithm is able to provide more high-frequency image details, which means this hybrid of VB and DL method combines the above advantages. The experiments show that this proposed algorithm can effectively increase the image resolution of remote sensing images by 0.5times at least comparing with low resolution image.

Extreme learning machine (ELM) and kernel ELM (KELM) can offer comparable performance as the standard powerful classifier―support vector machine (SVM), but with much lower computational cost due to extremely simple training step. However, their performance may be sensitive to several parameters, such as the number of hidden neurons. An empirical linear relationship between the number of training samples and the number of hidden neurons is proposed. Such a relationship can be easily estimated with two small training sets and extended to large training sets so as to greatly reduce computational cost. Other parameters, such as the steepness parameter in the sigmodal activation function and regularization parameter in the KELM, are also investigated. The experimental results show that classification performance is sensitive to these parameters; fortunately, simple selections will result in suboptimal performance.

Collecting reliable and accurate MR data on time plays a vital role in the mobile communication network optimization. However, with the increment of the number of mobile users, network bandwidth cannot meet with mass transfer of MR. A high performance and high compression ratio GSM-MR compression algorithm is proposed to gain better transfer time. This algorithm utilizes two step sorting in order to reduce the distance between similar content, based on the analytic result about similarities of GSM-MR data sorting by different fields. Experimental results reveal that the algorithm does not only decrease compression consuming time, but also ascends compression ratio with the increment of the size of compression data.

Pesticide residue detection in agriculture crops is a challenging issue and is even more difficult to quantify pesticide residue resident in agriculture produces and fruits. This paper conducts a series of base-line experiments which are particularly designed for three specific pesticides commonly used in Taiwan. The materials used for experiments are single leaves of vegetable produces which are being contaminated by various amount of concentration of pesticides. Two sensors are used to collected data. One is Fourier Transform Infrared (FTIR) spectroscopy. The other is a hyperspectral sensor, called Geophysical and Environmental Research (GER) 2600 spectroradiometer which is a batteryoperated field portable spectroradiometer with full real-time data acquisition from 350 nm to 2500 nm. In order to quantify data with different levels of pesticide residue concentration, several measures for spectral discrimination are developed. Mores specifically, new measures for calculating relative power between two sensors are particularly designed to be able to evaluate effectiveness of each of sensors in quantifying the used pesticide residues. The experimental results show that the GER is a better sensor than FTIR in the sense of pesticide residue quantification.

For over two decades, Los Alamos National Laboratory programs have included an active research effort utilizing satellite observations of terrestrial lightning to learn more about the Earth’s RF background. The FORTE satellite provided a rich satellite lightning database, which has been previously used for some event classification, and remains relevant for advancing lightning research. Lightning impulses are dispersed as they travel through the ionosphere, appearing as nonlinear chirps at the receiver on orbit. The data processing challenge arises from the combined complexity of the lightning source model, the propagation medium nonlinearities, and the sensor artifacts. We continue to develop modern event classification capability on the FORTE database using adaptive signal processing combined with compressive sensing techniques. The focus of our work is improved feature extraction using sparse representations in overcomplete analytical dictionaries. We explore two possible techniques for detecting lightning events, and showcase the algorithms on few representative data examples. We present preliminary results of our work and discuss future development.

In this paper, an automatic fractional coefficient setting method of fractional-order Darwinian particle swarm optimization (FODPSO) is proposed for hyperspectral image segmentation. The spectrum has been already taken into consideration by integrating various types of band selection algorithms, firstly. We provide a short overview of the hyperspectral image to select an appropriate set of bands by combining supervised, semi-supervised and unsupervised band selection algorithms. Some approaches are not limited in regards to their spectral dimension, but are limited with respect to their spatial dimension owing to low spatial resolution. The addition of spatial information will be focused on improving the performance of hyperspectral image segmentation for later fusion or classification. Many researchers have advocated that a large fractional coefficient should be in the exploration state while a small fractional coefficient should be in the exploitation, which does not mean the coefficient purely decrease with time. Due to such reasons, we propose an adaptive FODPSO by setting the fractional coefficient adaptively for the application of final hyperspectral image segmentation. In fact, the paper introduces an evolutionary factor to automatically control the fractional coefficient by using a sigmoid function. Therefore, fractional coefficient with large value will benefit the global search in the exploration state. Conversely, when the fractional coefficient has a small value, the exploitation state is detected. Hence, it can avoid optimization process get trapped into the local optima. Ultimately, the experimental segmentation results prove the validity and efficiency of our proposed automatic fractional coefficient setting method of FODPSO compared with traditional PSO, DPSO and FODPSO.

Intel Many Integrated Core (MIC) ushers in a new era of supercomputing speed, performance, and compatibility. It allows the developers to run code at trillions of calculations per second using the familiar programming model. In this paper, we present our results of optimizing the updated Goddard shortwave radiation Weather Research and Forecasting (WRF) scheme on Intel Many Integrated Core Architecture (MIC) hardware. The Intel Xeon Phi coprocessor is the first product based on Intel MIC architecture, and it consists of up to 61 cores connected by a high performance on-die bidirectional interconnect. The co-processor supports all important Intel development tools. Thus, the development environment is familiar one to a vast number of CPU developers. Although, getting a maximum performance out of Xeon Phi will require using some novel optimization techniques. Those optimization techniques are discusses in this paper. The results show that the optimizations improved performance of the original code on Xeon Phi 7120P by a factor of 1.3x.

Progressive band processing (PBP) processes data band by band according to the Band SeQuential (BSQ) format acquired by a hyperspectral imaging sensor. It can be implemented in real time in the sense that data processing can be performed whenever bands are available without waiting for data completely collected. This is particularly important for satellite communication when data download is limited by bandwidth and transmission. This paper presents a new concept of processing a well-known technique, Orthogonal Subspace Projection (OSP) band by band, to be called PBPOSP. Several benefits can be gained by PBP-OSP. One is band processing capability which allows different receiving ends to process data whenever bands are available. Second, it enables users to identify significant bands during data processing. Third, unlike band selection which requires knowing the number of bands needed to be selected or band prioritization PBP-OSP can process arbitrary bands in real time with no need of such prior knowledge. Most importantly, PBP can locate and identify which bands are significant for data processing in a progressive manner. Such progressive profile resulting from PBP-OSP is the best advantage that PBP-OSP can offer and cannot be accomplished by any other OSP-like operators.

OSP has been used widely in detection and abundance estimation for about twenty years. But it can’t apply nonnegative and sum-to-one constraints when being used as an abundance estimator. Fully constrained least square algorithm does this well, but its time cost increases greatly as the number of endmembers grows. There are some tries for unmixing spectral under fully constraints from different aspects recently. Here in this paper, a new fully constrained unmixing algorithm is prompted based on orthogonal projection process, where a nearest projected point is defined onto the simplex constructed by endmembers. It is much easier, and it is faster than FCLS with the mostly same unmixing results. It is also compared with other two constrained unmixing algorithms, which shows its effectiveness too.

In the study, we develop a multiple constrained signal subspace projection (SSP) approach to target detection. Instead of using single constraint on target detection, we design an optimal filter with multiple constraints on desired targets by using SSP. The proposed SSP approach fully exploits the orthogonal property of two orthogonal subspaces: one denoted signal subspace containing desired and undesired/background targets; the other denoted noise subspace, which is orthogonal to signal subspace. By projecting the weights of the detection filter on the signal subspace, the proposed SSP can reduces some estimation errors in target signatures and alleviate the performance degradation caused by uncertainty of target signature. The SSP approach can detect desired targets, suppress undesired targets and minimize the interference effects. In experiments, we provide three methods in selecting multiple constraints of the desired target: Kmeans, principal eigenvectors and endmenber extracting techniques. Simulation results show that the proposed SSP with multiple constraints selected by K-means has better detection performance. Furthermore, the proposed SSP with multiple constraints is a robust detection approach which could overcome the uncertainty of desired target signature in real image data.

A novel approach to mammographic image segmentation, termed as PCNN-based level set algorithm, is presented in this paper. Just as its name implies, a method based on pulse coupled neural network (PCNN) in conjunction with the variational level set method for medical image segmentation. To date, little work has been done on detecting the initial zero level set contours based on PCNN algorithm for latterly level set evolution. When all the pixels of the input image are fired by PCNN, the small pixel value will be a much more refined segmentation. In mammographic image, the breast tumor presents big pixel value. Additionally, the mammographic image with predominantly dark region, so that we firstly obtain the negative of mammographic image with predominantly dark region except the breast tumor before all the pixels of an input image are fired by PCNN. Therefore, in here, PCNN algorithm is employed to achieve mammary-specific, initial mass contour detection. After that, the initial contours are all extracted. We define the extracted contours as the initial zero level set contours for automatic mass segmentation by variational level set in mammographic image analysis. What’s more, a new proposed algorithm improves external energy of variational level set method in terms of mammographic images in low contrast. In accordance with the gray scale of mass region in mammographic image is higher than the region surrounded, so the Laplace operator is used to modify external energy, which could make the bright spot becoming much brighter than the surrounded pixels in the image. A preliminary evaluation of the proposed method performs on a known public database namely MIAS, rather than synthetic images. The experimental results demonstrate that our proposed approach can potentially obtain better masses detection results in terms of sensitivity and specificity. Ultimately, this algorithm could lead to increase both sensitivity and specificity of the physicians’ interpretation of mammograms in clinical practice.

The Weather Research and Forecast (WRF) Double Moment 5-class (WDM5) mixed ice microphysics scheme predicts the mixing ratio of hydrometeors and their number concentrations for warm rain species including clouds and rain. WDM5 can be computed in parallel in the horizontal domain using multi-core GPUs. In order to obtain a better GPU performance, we manually rewrite the original WDM5 Fortran module into a highly parallel CUDA C program. We explore the usage of coalesced memory access and asynchronous data transfer. Our GPU-based WDM5 module is scalable to run on multiple GPUs. By employing one NVIDIA Tesla K40 GPU, our GPU optimization effort on this scheme achieves a speedup of 252x with respect to its CPU counterpart Fortran code running on one CPU core of Intel Xeon E5-2603, whereas the speedup for one CPU socket (4 cores) with respect to one CPU core is only 4.2x. We can even boost the speedup of this scheme to 468x with respect to one CPU core when two NVIDIA Tesla K40 GPUs are applied.

This paper puts emphasis on how to model and correct image blur that arises from a camera’s ego motion while observing a distant star scene. Concerning the significance of accurate estimation of point spread function (PSF), a new method is employed to obtain blur kernel by thinning star motion path. In particular, how the blurred star image can be corrected to reconstruct the clear scene with a thinning motion blur model which describes the camera’s path is presented. This thinning motion path to build blur kernel model is more effective at modeling the spatially motion blur introduced by camera’s ego motion than conventional blind estimation of kernel-based PSF parameterization. To gain the reconstructed image, firstly, an improved thinning algorithm is used to obtain the star point trajectory, so as to extract the blur kernel of the motion-blurred star image. Then how motion blur model can be incorporated into the Richardson-Lucy (RL) deblurring algorithm, which reveals its overall effectiveness, is detailed. In addition, compared with the conventional estimated blur kernel, experimental results show that the proposed method of using thinning algorithm to get the motion blur kernel is of less complexity, higher efficiency and better accuracy, which contributes to better restoration of the motion-blurred star images.

Endmember finding has received considerable interest in hyperspectral imaging. In reality an endmember finding algorithm (EFA) suffers from endmember variability which causes inaccuracy, inconsistency and instability. In this case a real endmember may not exist but rather appears as its variant, referred to as virtual signature (VS). This paper presents a new approach to finding VSs by taking endmember variability into account. It first determines a required number of endmember classes by virtual dimensionality (VD), then designs an unsupervised method to find endmember classes and finally develops an iterative algorithm to find VSs. Comprehensive experiments including synthetic and real image scenes are conducted to demonstrate effectiveness of the proposed approach.

Human action recognition and analysis is an active research topic in computer vision for many years. This paper presents a method to represent human actions based on trajectories consisting of 3D joint positions. This method first decompose action into a sequence of meaningful atomic actions (actionlets), and then label actionlets with English alphabets according to the Davies-Bouldin index value. Therefore, an action can be represented using a sequence of actionlet symbols, which will preserve the temporal order of occurrence of each of the actionlets. Finally, we employ sequence comparison to classify multiple actions through using string matching algorithms (Needleman-Wunsch). The effectiveness of the proposed method is evaluated on datasets captured by commodity depth cameras. Experiments of the proposed method on three challenging 3D action datasets show promising results.

With the improvement of spatial resolution of hyperspectral imagery, it is more reasonable to include spatial information in classification. The resulting spectral-spatial classification outperforms the traditional hyperspectral image classification with spectral information only. Among many spectral-spatial classifiers, support vector machine with composite kernel (SVM-CK) can provide superior performance, with one kernel for spectral information and the other for spatial information. In the original SVM-CK, the spatial information is retrieved by spatial averaging of pixels in a local neighborhood, and used in classifying the central pixel. Obviously, not all the pixels in such a local neighborhood may belong to the same class. Thus, we investigate the performance of Gaussian lowpass filter and an adaptive filter with weights being assigned based on the similarity to the central pixel. The adaptive filter can significantly improve classification accuracy while the Gaussian lowpass filter is less time-consuming and less sensitive to the window size.

Virtual dimensionality (VD) has been widely used to estimate number of endmembers in the past. Unfortunately, the original idea of VD was developed to specify the number of spectrally distinct signatures in hyperspectral data where there is no provided specific definition of what “spectrally distinct signatures” are. As a result, many techniques developed to estimate VD have produced various values for VD. This paper addresses this issue by develops a target specified VD (TSVD) theory where the value of VD is completely determined by targets of interest. In particular, the VD techniques can be categorized according to targets characterized by eigenvalues/eigenvectors and real target signal sources which are used for a binary composite hypothesis testing problem. For the latter case the Automatic Target Generation Process (ATGP) is particularly used to generate real target signal sources to replace eigenvalues/eigenvectors as signal sources to be used for the binary hypothesis testing problem. In order to find probability distributions under each hypothesis the extreme theory used by Maximum Orthogonal Complement Algorithm (MOCA) is used for their derivations. As a result, VD can be estimated by two types of signals sources, eigenvalues/eigenvectors along with two types of detectors, maximum likelihood detector and Neyman-Pearson detector.

A novel approach for microcalcification clusters detection is proposed. At the first time, we make a short analysis of mammographic images with microcalcification lesions to confirm these lesions have much greater gray values than normal regions. After summarizing the specific feature of microcalcification clusters in mammographic screening, we make more focus on preprocessing step including eliminating the background, image enhancement and eliminating the pectoral muscle. In detail, Chan-Vese Model is used for eliminating background. Then, we do the application of combining morphology method and edge detection method. After the AND operation and Sobel filter, we use Hough Transform, it can be seen that the result have outperformed for eliminating the pectoral muscle which is approximately the gray of microcalcification. Additionally, the enhancement step is achieved by morphology. We make effort on mammographic image preprocessing to achieve lower computational complexity. As well known, it is difficult to robustly achieve mammograms analysis due to low contrast between normal and lesion tissues, there are also much noise in such images. After a serious preprocessing algorithm, a method based on blob detection is performed to microcalcification clusters according their specific features. The proposed algorithm has employed Laplace operator to improve Difference of Gaussians (DoG) function in terms of low contrast images. A preliminary evaluation of the proposed method performs on a known public database namely MIAS, rather than synthetic images. The comparison experiments and Cohen’s kappa coefficients all demonstrate that our proposed approach can potentially obtain better microcalcification clusters detection results in terms of accuracy, sensitivity and specificity.

Pixel Purity Index (PPI) is a very popular endmember finding algorithm due to its availability in ENVI software. According to the band sequential (BSQ) format of data acquisition this paper introduces a new concept of executing PPI band-by-band in a progressive manner. It is called progressive band processing of PPI (PBP-PPI) which allows users to process PPI band by band without waiting for full bands of data information acquired. To accomplish this goal PPI must be capable of calculating and updating PPI counts of data samples band by band. Furthermore, progressive-band-processing progressive PPI (PBP-P-PPI) and progressive-band-processing causal PPI (PBP-C-PPI) are proposed to address the issues that the number of skewers is undefined and only partial pixels are available correspondingly. Many benefits can be gained from PBP-PPI, for example, providing progressive profiles of PPI counts of data samples as more bands are included for data processing, finding crucial bands according to progressive changes in PPI counts.

Collaborative representation classifier (CRC) has been applied to hyperspectral image classification, which intends to use all the atoms in a dictionary to represent a testing pixel for label assignment. However, some atoms that are very dissimilar to the testing pixel should not participate in the representation, or their contribution should be very little. The regularized version of CRC imposes strong penalty to prevent dissimilar atoms with having large representation coefficients. To utilize spatial information, the weighted sum of local spatial neighbors is considered as a joint spatial-spectral feature, which is actually for regularized CRC-based classification. This paper proposes its kernel version to further improve classification accuracy, which can be higher than those from the traditional support vector machine with composite kernel and the kernel version of sparse representation classifier.

The relative motion between remote sensing satellite sensor and objects is one of the most common reasons for remote sensing image degradation. It seriously weakens image data interpretation and information extraction. In practice, point spread function (PSF) should be estimated firstly for image restoration. Identifying motion blur direction and length accurately is very crucial for PSF and restoring image with precision. In general, the regular light-and-dark stripes in the spectrum can be employed to obtain the parameters by using Radon transform. However, serious noise existing in actual remote sensing images often causes the stripes unobvious. The parameters would be difficult to calculate and the error of the result relatively big. In this paper, an improved motion blur parameter identification method to noisy remote sensing image is proposed to solve this problem. The spectrum characteristic of noisy remote sensing image is analyzed firstly. An interactive image segmentation method based on graph theory called GrabCut is adopted to effectively extract the edge of the light center in the spectrum. Motion blur direction is estimated by applying Radon transform on the segmentation result. In order to reduce random error, a method based on whole column statistics is used during calculating blur length. Finally, Lucy-Richardson algorithm is applied to restore the remote sensing images of the moon after estimating blur parameters. The experimental results verify the effectiveness and robustness of our algorithm.

The most widely used community weather forecast and research model in the world is the Weather Research and Forecast (WRF) model. Two distinct varieties of WRF exist. The one we are interested is the Advanced Research WRF (ARW) is an experimental, advanced research version featuring very high resolution. The WRF Nonhydrostatic Mesoscale Model (WRF-NMM) has been designed for forecasting operations. WRF consists of dynamics code and several physics modules. The WRF-ARW core is based on an Eulerian solver for the fully compressible nonhydrostatic equations. In the paper, we optimize a meridional (north-south direction) advection subroutine for Intel Xeon Phi coprocessor. Advection is of the most time consuming routines in the ARW dynamics core. It advances the explicit perturbation horizontal momentum equations by adding in the large-timestep tendency along with the small timestep pressure gradient tendency. We will describe the challenges we met during the development of a high-speed dynamics code subroutine for MIC architecture. Furthermore, lessons learned from the code optimization process will be discussed. The results show that the optimizations improved performance of the original code on Xeon Phi 7120P by a factor of 1.2x.

This paper develops a completely new technology,) from a hyperspectral imaging perspective, called Hyperspectral Vital Sign Signal Analysis (HyVSSA. A hyperspectral image is generally acquired by hundreds of contiguous spectral bands, each of which is an optical sensor specified by a particular wavelength. In medical application, we can consider a patient with different vital sign signals as a pixel vector in hyperspectral image and each vital sign signal as a particular band. In light of this interpretation, a revolutionary concept is developed, which translates medical data to hyperspectral data in such a way that hyperspectral technology can be readily applied to medical data analysis. One of most useful techniques in hyperspectral data processing is, Anomaly Detection (AD) which in this medical application is used to predict outcomes such as transfusion, length of stay (LOS) and mortality using various vital signs. This study compared transfusion prediction performance of Anomaly Detection (AD) and Logistic Regression (LR).

An approach for haze removal utilizing polarimetric imaging and multi-scale analysis has been developed to solve one problem that haze weather weakens the interpretation of remote sensing because of the poor visibility and short detection distance of haze images. On the one hand, the polarization effects of the airlight and the object radiance in the imaging procedure has been considered. On the other hand, one fact that objects and haze possess different frequency distribution properties has been emphasized. So multi-scale analysis through wavelet transform has been employed to make it possible for low frequency components that haze presents and high frequency coefficients that image details or edges occupy are processed separately. According to the measure of the polarization feather by Stokes parameters, three linear polarized images (0°, 45°, and 90°) have been taken on haze weather, then the best polarized image min I and the worst one max I can be synthesized. Afterwards, those two polarized images contaminated by haze have been decomposed into different spatial layers with wavelet analysis, and the low frequency images have been processed via a polarization dehazing algorithm while high frequency components manipulated with a nonlinear transform. Then the ultimate haze-free image can be reconstructed by inverse wavelet reconstruction. Experimental results verify that the dehazing method proposed in this study can strongly promote image visibility and increase detection distance through haze for imaging warning and remote sensing systems.

In this paper, a parallel FPGA architecture implementing BCH error correction for NAND flash memories in storage modules of Earth observation satellites (EOS) is presented. In order to satisfy the error correction requirements of NAND flash memories the designed module is capable of correcting 4 erroneous bits in code words (CW) of 4148 bits using shortened BCH (4148, 4096) algorithm. Besides the standard BCH encoding and decoding procedure a code selection module is added to the architecture to guess whether or not the CW has more than 4 errors. If more than 4 errors occurs in the CW the architecture understands that the CW is uncorrectable and does not try to change any bits which will most probably inject more errors to the CW. The design is implemented on a XC7VX485T FPGA from Virtex 7 series of Xilinx and with a 300 MHz system clock, it is capable of encoding and decoding information at 4,78GBits/sec.

Aiming to realize super resolution(SR) to single image and video reconstruction, a super resolution camera model is proposed for the problem that the resolution of the images obtained by traditional cameras behave comparatively low. To achieve this function we put a certain driving device such as piezoelectric ceramics in the camera. By controlling the driving device, a set of continuous low resolution(LR) images can be obtained and stored instantaneity, which reflect the randomness of the displacements and the real-time performance of the storage very well. The low resolution image sequences have different redundant information and some particular priori information, thus it is possible to restore super resolution image factually and effectively. The sample method is used to derive the reconstruction principle of super resolution, which analyzes the possible improvement degree of the resolution in theory. The super resolution algorithm based on learning is used to reconstruct single image and the variational Bayesian algorithm is simulated to reconstruct the low resolution images with random displacements, which models the unknown high resolution image, motion parameters and unknown model parameters in one hierarchical Bayesian framework. Utilizing sub-pixel registration method, a super resolution image of the scene can be reconstructed. The results of 16 images reconstruction show that this camera model can increase the image resolution to 2 times, obtaining images with higher resolution in currently available hardware levels.

A simulation method for analyzing polarization states for infrared scenes is proposed in order to study the polarization features of infrared spontaneous emission deeply, since current infrared polarization devices can’t show the polarization signature of infrared spontaneous emission for a target or an object well. A preliminary analysis on polarization characteristics of infrared spontaneous emission in the ideal case is carried out and also a corresponding ideal model is established through Kirchhoff’s law and the Fresnel theorem. Based on the newly built ideal model, a three-dimensional (3D) scene modeling and simulation based on the OpenSceneGraph (OSG) rendering engine is utilized to obtain the polarization scene of infrared emission under ideal conditions. Through the corresponding software, different infrared scenes can be generated by adjusting the input parameters. By interacting with the scene, the infrared polarization images can be acquired readily, also a fact can be obviously confirmed that the degree of linear polarization (DoLP) for an object in the 3D scene varies with the many factors such as emission angle and complex refractive index. Moreover, large difference between two kinds of material such as metal and nonmetal in the polarization characteristics of infrared spontaneous emission at the same temperature can be easily discerned in the 3D scene. The 3D scene simulation and modeling in the ideal case provides a direct understanding on infrared polarization property, which is of great significance for the further study of infrared polarization characteristics in the situation of real scenes.

This study focuses on the key theory of lightning detection, exposure and the experiments. Firstly, the algorithm based on differential operation between two adjacent frames is selected to remove the lightning background information and extract lighting signal, and the threshold detection algorithm is applied to achieve the purpose of precise detection of lightning. Secondly, an algorithm is proposed to obtain scene exposure value, which can automatically detect external illumination status. Subsequently, a look-up table could be built on the basis of the relationships between the exposure value and average image brightness to achieve rapid automatic exposure. Finally, based on a USB 3.0 industrial camera including a CMOS imaging sensor, a set of hardware test platform is established and experiments are carried out on this platform to verify the performances of the proposed algorithms. The algorithms can effectively and fast capture clear lightning pictures such as special nighttime scenes, which will provide beneficial supporting to the smartphone industry, since the current exposure methods in smartphones often lost capture or induce overexposed or underexposed pictures.

Remote Sensing Image can be degraded by a variety of causes during acquisition, transmission, compression, storage and reconstruction. Noise is one of the most important degradation factors. Quantifying its impact on the image may be useful for applications such as improving the acquisition system and thus the quality of the produced images. Objective Image Quality Measure (IQA) methods can be classified by whether a reference image, representing the original signal exists. In the case of remote sensing, the ideal un-degraded image is not available. No-reference (NR) method is required to blindly assess the image quality. In this paper, a new no-reference algorithm is proposed to quantify noise based on local phase coherence (LPC). This algorithm assumes that the input image is contaminated by additive zero mean Gaussian noise. Firstly, a LPC map of degraded image is constructed and the image edge is extracted by modifying the noise threshold. Secondly, the edge is removed from the LPC map. Then, the noise level can be quantified by the remaining noise information and little “residual” information of the LPC map. Experiment results show that the proposed algorithm correlates well with subjective quality evaluations and has high estimation accuracy especially for Gaussian noise-infected images.

The need for a portable image acquiring system has become as strong as the extension of digital imaging technology, for this, a new mono-centric wide-field optical system is proposed. Recently, some high-resolution and wide-field imaging systems have been raised already, with which fairly clear and wide field of view (FOV) images could be easily obtained, however, their sizes are comparatively too large to be conveniently carried . With ZEMAX, a new optical design is emulated by scaling the structure of current wide-field optical systems and introducing the proposed lens-let arrays, the size of the whole system is comparatively smaller with the structure consisting of a two-glass mono-centric lens, lens-let array (the lenses in the array can be different), and a specific detector. Lens-let array is used to make the image plane from curve to almost flat. This hardware is small enough to apply to helmets and computers and the FOV of which is wide. Verified by a series of merit function, this optical design is found to have an acceptable imaging resolution and the computational imaging method is applied to this system to acquire a higher imaging resolution. From each lens-let a series of low resolution images are obtained and in this system a high-resolution image can be retrieved from multiple low-resolution images with super-resolution reconstruction method. Compared from the size and the imaging resolution, this new optical design is much smaller and has a higher imaging resolution.