Search

Most far-field optical imaging systems rely on lenses and spatially resolved detection to probe distinct locations on the object. We describe and demonstrate a high-speed wide-field approach to imaging that instead measures the complex spatial Fourier transform of the object by detecting its spatially integrated response to dynamic acousto-optically synthesized structured illumination. Tomographic filtered back-projection is applied to reconstruct the object in two or three dimensions. This technique decouples depth of field and working distance from resolution, in contrast to conventional imaging, and can be used to image biological and synthetic structures in fluoresced or scattered light employing coherent or broadband illumination. We discuss the electronically programmable transfer function of the optical system and its implications for imaging dynamic processes. We also explore wide-field fluorescence imaging in scattering media by coherence gating. Finally, we present two-dimensional high-resolution tomographic image reconstructions in both scattered and fluoresced light demonstrating a thousandfold improvement in the depth of field compared to conventional lens-based microscopy. [c] 2010 Optical Society of America OCIS codes: 110.0180, 110.6960, 110.4850, 180.6900, 180.1655, 180.2520.

A single scanning nano-slit is used to study aerial image characteristics. Finite-difference time-domain simulations reveal that, in the far field of such a slit, the detected image contrast is very high over a large spatial frequency range regardless of the polarization direction. In the near field, the TM polarization shows a decrease in contrast at larger spatial frequencies. Experiments verify this characteristic using a 125 nm wide slit on an aluminum mask at a wavelength of 658 nm. Unlike the light transmission characteristics of a nano-slit, which are greatly influenced by slit width and metal mask thickness, it is shown that image contrast measurement is almost insensitive to small changes in these parameters. It is found that defects on the metal mask play an important role in accurate analysis of the system. [c] 2010 Optical Society of America OCIS codes: 110.1220, 110.2960, 260.3910, 260.5430, 310.6628.

A step-zoom and reimaging structure were utilized to construct a dual field-of-view optical design for high-magnification switching in the 3-5 [micro]m spectral band. The design has a flexible optomechanical layout, which means it can be utilized for multipurpose applications. The effects of the surrounding environmental temperature and axial gradient temperature are analyzed using the concept of thermal resistance, and the thermal compensation is discussed. A description of the zooming mechanism and optomechanical control is offered. [c] 2010 Optical Society of America OCIS codes: 110.3080, 220.3620.

Recent developments in micro-electro-mechanical systems and biotechnology have presented an emerging need for observation of dynamic targets in three-dimensional space. Unfortunately, conventional microscopes with fixed optical parameters have difficulty supplying sufficient vision information because of occlusion, small field of view, and low depth resolution. This paper introduces the design of a variable view imaging system that can supply a flexible view with a relatively large zenith angle and simple kinematics. Because of several performance factors, a multiobjective optimization process is applied to achieve an appropriate design. A prototype system is developed and used to verify the proposed design. OCIS codes: 110.1080, 220.1080, 150.3040, 170.0110, 170.0180, 220.4830.

We present a configuration for a real-time spatial image processor that is based upon an imaging setup in which a grating with Fourier coefficients with tunable phase is attached to the object plane. The illumination that is used for the proposed concept is spatially incoherent. By proper adjusting of the magnification of the imaging system to the spatial period of the grating and the sampling grid of the camera, the aliasing effect due to the digital sampling realizes a nonuniform and tunable spectral distribution (a filter) that is applied over the spectrum of the object. Preliminary numerical and experimental demonstration of the operation principle is provided with a spatial LiNb[O.sub.3] hexagonal grating. [c] 2009 Optical Society of America OCIS codes: 070.6110, 100.2980.

The article presents an improved method for predicting discrete cosine transform (DCT) coefficients of an image block by using 1-D DCT of the rightmost column of pixels of the block on the left side and bottom row pixels of the block on the top side. Experimental results show about one-third AC energy packed, just in the first two coefficients.

A new method is developed to obtain multilevel annotations of large-scale images automatically. A novel hierarchical boosting algorithm is used to resolve the problem of huge intraconcept visual diversity for the image concepts at the higher levels of the concept ontology.

The 2-D walking ant histogram (WAH) descriptor, which enables shape and texture retrievals over general-purpose multimedia databases, is presented. The design of the descriptor is based on multiscale subsegment analysis over edge field and hence can deal directly with arbitrary images without any unreliable segmentation or object extraction preprocessing stage.

A robust multiple model data association (MMDA) tracker based on bank of nonlinear filters is used to address object tracking in ultrasound images. The algorithm helps to determine, which model is active at each instant of time and updates its state by propagating the probability distribution using robust estimation techniques.

The performance of the segmentation scheme that couples contrast criteria in flooding with texture information is demonstrated. The modeling of the scheme is designed via partial differential equation (PDEs) and efficient incorporation of the available contrast and texture information.

The use of an image-scale pyramid and directional filtering in the spatial domain for fingerprint image enhancement is demonstrated to improve the matching performance as well as the computational efficiency.

Two novel algorithms for the simultaneous estimation of the image and the unknown hyperparameters in total variation (TV)-based image restoration problems are presented. The algorithms provide approximation to the posterior distribution of the latent variables using variational methods.

A specially parameterized scheme for approximate joint diagonalization of estimated spectrum matrices is developed to determine the succinct set of mixture parameters, the static coefficients and the shift values. The performance of algorithm for both synthetic mixtures and with real-life reflection images is demonstrated.

A method for accurate and efficient localization of local image features, which makes a significant contribution to the problem of feature based object detection and recognition is presented. The design of the method is based on multiresolution Gabor features and statistical ranking using multiple hypothesis testing.

A method for identification of binary objects from a given set of radon projections is described. The properties of the radon transform could be used to determine the relation between the radon transforms of f and g and also extract the parameters involved in the sequence transformation.

The theoretical results from the invertibility of filter banks under continuous spherical convolution are used to develop a general framework for the design of invertible filter banks on the sphere. The application of self-invertible spherical steerable pyramid in denoising experiment and computational complexity of the filtering framework is also demonstrated.

A different approach to use numerical optimization techniques to improve the orthogonality which would facilitate accurate image reconstruction is presented. The simulation results suggest that the optimized digital orthogonal rotation-invariant moments (ORIMs) allow accurate reconstruction of images and could also represent subtle image details.

A new family of multivariate gamma based distributions for multisensor synthetic aperture radar (SAR) images, referred to as multisensor multivariate gamma distributions (MuMGDs) is studied. The findings suggest that the MuMGDs can be used to detect changes in SAR images acquired by different sensors having different numbers of looks.

A multilinear expansion of the probability density function of the source statistics is used to refine the classical independent component analysis (ICA) decomposition. The new enhanced multilinear ICA (MICA) model can be used to analyze nonstationarity natural images using natural scene statistics models.

Due to the limited resolution of both cameras and displays, acuity of artificial vision systems is currently well below the human eye. Visual acuity, in cameras as well as in animal eyes, can be increased by making smaller receptors or bigger eyes. In some applications, the size of the camera is constrained, so alternative solutions must be sought. This paper presents a robotic dual-camera vision system whose design is inspired by the visual system of jumping spiders (Salticidae family). The system is composed of a telephoto camera whose field of view (FOV) can be moved within the larger FOV of a wide-angle camera and allows to form a high-resolution image, i.e., an image with the FOV of the wide-angle camera, yet having the same resolution as the telephoto camera. We describe the design of the robotic system, the direct and inverse kinematics, and the image processing algorithms that allow to build the high-resolution image. Images from experiments are presented, together with a discussion on sources of errors and possible solutions. The system is particularly useful for fixed-camera monitoring or teleoperation applications, such as remote surveillance and minimally invasive surgery. The system achieves seven times higher resolution than typical commercial endoscopes. Index Terms--Augmented reality, panoramic stitching, robot vision, scanning mirror.

A novel method for examining image reconstruction from randomly scattered sensors in a multidimensional space is presented. The efficiency of the iterative estimation method in image reconstruction and field estimation is demonstrated.

A noniterative Bayes least squares-Gaussian scale mixtures (BLS-GSM) model that uses space-variant GSM is proposed for image restoration. Results suggest that the method enhances performance and robustness for denoising and deconvolution.

A two stage method for obtaining phase and object estimates from phase-diversity time series data is presented. Findings reveal the effectiveness of the method as it allows for the utilization of larger data sets.

A novel algorithm based on the motor map neural network is presented for solving the palette re-ordering problem. Results suggest the effectiveness of the algorithm in terms of compression ratio and zero-order entropy of local differences.

Eight optimization methods for nonrigid registration based on the maximization of mutual information in combination with a deformation field parameterized by cubic B-splines are compared. Results suggest that the Robbins-Monro method is most preferred as the computation time can be drastically reduced.

A handheld near-infrared imager was developed for real-time monitoring of tissue physiologic changes in response to dynamic compression stimuli. Both 2D and 3D imaging schemas were developed for reconstruction of tissue heterogeneities based on optical measurements. The handheld imager and the dynamic imaging schema were validated on both benchtop phantoms and in vivo human tissues. The benchtop tests demonstrated that the imager was able to reconstruct absorption properties of the embedded heterogeneity with accuracy and repeatability. The tests on in vivo human tissues demonstrated that the imager was able to generate various dynamic loading profiles with reproducibility and to detect tissue optical, mechanical, and physiologic changes under the dynamic loading condition. OCIS codes: 170.2655, 170.3010, 170.5280, 170.6510, 170.6935.

The new emphasis on Anti-Terrorism and Force Protection (AT/FP), for both shore and sea platform protection, has resulted in a need for infrared imager design and evaluation tools that demonstrate field performance against U.S. Navy AT/FP requirements. In the design of infrared imaging systems for target acquisition, a discrimination criterion is required for successful sensor realization. It characterizes the difficulty of the task being performed by the observer and varies for different target sets. This criterion is used in both assessment of existing infrared sensor and in the design of new conceptual sensors. We collected 12 small craft signatures (military and civilian) in the visible band during the day and the long-wave and midwave infrared spectra in both the day and the night environments. These signatures were processed to determine the targets' characteristic dimension and contrast. They were also processed to band limit the signature's spatial information content (simulating longer range), and a perception experiment was performed to determine the task difficulty ([N.sub.50] and [V.sub.50]). The results are presented and can be used for Navy and Coast Guard imaging infrared sensor design and evaluation. OCIS codes: 110.6820, 110.3080, 110.4100.

Several architectures of wavefront sensors have been developed since the rise of adaptive optics. In all cases, optical elements are placed in front of image sensors. This makes the sensor quite bulky, expensive, and sensitive to optical misalignment. I propose two novel architectures fully embedded in the image sensor that require no additional optical element. The sensor can be placed directly in the beam to analyze, leading to small, easy to use, and cost-efficient systems. The two architectures are described before testing by simulation of their ability to sense the wavefront distortion and their sensitivity to signal-to-noise ratio. OCIS codes: 010.1080, 040.0040.

Herein, we present a novel concept for fully integrated dual-modality in vivo tomographic imaging yielding simultaneous detection of positron and optically labeled probes in small animals. The imager consists of an allocation of optical detector modules and, in radial extension, the allocation of positron emission detector modules. Laser scanning and large-field light sources are integrated to facilitate fluorescence imaging in addition to bioluminescence imaging. Each optical detector unit consists of a large-area photon sensor for light detection, a microlens array for field-of-view definition, a septum mask for cross-talk suppression, and a transferable filter for wavelength selection. To prove the working principle of the dual-modality detector system a pair of optical detectors along with a large-field excitation source was placed inside the bore of a Siemens EXACT HR+ scanner, performing simultaneous imaging. The imaging characteristics of the optical detector were evaluated experimentally using a prototypical setup with geometrical phantoms. The sensitivity of the optical detector prototype was found less than that of a reference CCD camera. We propose several ways of increasing optical detector sensitivity. Index Terms--Dual-modality imaging, optical imaging, positron emission tomography, small animal imaging.

In this paper, a biologically inspired, CNN-based, multi-channel, texture boundary detection technique is presented. The proposed approach is similar to human vision system. The algorithm is simple and straightforward such that it can be implemented on the cellular neural networks (CNNs). CNN contains several important advantages, such as efficient real-time processing capability and feasible very large-scale integration (VLSI) implementation. The proposed algorithm also had been widely tested on synthetic texture images. Those texture images are randomly selected from the Brodatz textures database [1]. According to our simulation results, the boundaries of uniform textures can be detected quite successfully. For the nonuniform or nonregular textures, the results also indicate meaningful properties, and the properties also are consistent to the human visual sensation. The proposed algorithm also has been implemented on the CNN Universal Machine (CNN-UM), and yields similar results as the simulation on the PC. Based on the efficient performance of CNN-UM, the algorithm becomes very fast. Index Terms--Cellular neural networks (CNNs), early vision system, Gabor filter, retinex model, texture segregation.

Future optical microsystems, such as microelectromechanical system (MEMS) scanners and micromirrors, will extend the resolution and sensitivity offered by their predecessors. These systems face the challenge of achieving nanometer precision subjected to various disturbances. Predicting the performance of such systems early in the design process can significantly impact the design cost and also improve the quality of the design. Our approach aims to predict the performance of such systems under various disturbance sources and develop a generalized design approach for MEMS structures. In this study, we used ANSYS for modeling and dynamic analysis of a torsional MEMS scanner mirror. ANSYS modal analysis results, which are eigenvalues (natural frequencies) and eigenvectors (mode shapes), are used to obtain the state-space representation of the mirror. The state-space model of the scanner mirror was reduced using various reduction techniques to eliminate the states that are insignificant for the transfer functions of interest. The results of these techniques were compared to obtain the best approach to obtain a lower order model that still contains all the relevant dynamics of the original model. After the model size is reduced significantly, a disturbance analysis is performed using Lyapunov approach to obtain root-mean-square values of the mirror rotation angle under the effect of a disturbance torque. The magnitude levels of the disturbance torque are obtained using an experimental procedure. The disturbance analysis framework is combined with the sensitivity analysis to determine the critical design parameters for optimizing the system performance. [DOI: 10.1115/1.2756087]

We describe a method in which holograms can be produced by calculation from images captured by integral photography (IP). We present a basic algorithm obtained by simulating IP reconstruction, in which conditions are set so as not to cause aliasing in the holograms after the calculations. To reduce the calculation load, we also propose a way to limit the range of calculation considering the distribution of light and a way to shift the optical field on the exit plane of microlenses in a lens array. Finally, by optical experiments, we confirm that three-dimensional images can be reconstructed from holograms calculated from an integral photograph of a real object captured with an IP camera. OCIS codes: 070.6110, 090.1760, 090.2870, 100.2650.

We present in this paper a post-processor for the reduction of color bleeding in 4:1:1 digital video decompressed sequences. It is an extension to the specific case of 4:1:1 video format of our previous work on color image enhancement. The proposed algorithm is nonlinear and adaptive depending on the local characteristics of the video material. First, the different parts of the algorithm are described in detail: this include the detection, then the removal of color bleeding in coded video. Finally experimental results are shown, that exhibit the improvement of color fidelity of the post-processed images. Index Terms--Color bleeding, image coding, image enhancement.