Edmund Y. Lam

A mathematical analysis of the DCT coefficient distributions for images

Over the past two decades, there have been various studies on the distributions of the DCT coefficients for images. However, they have concentrated only on fitting the empirical data from some standard pictures with a variety of well-known statistical distributions, and then comparing their goodness of fit. The Laplacian distribution is the dominant choice balancing simplicity of the model and fidelity to the empirical data. Yet, to the best of our knowledge, there has been no mathematical justification as to what gives rise to this distribution. We offer a rigorous mathematical analysis using a doubly stochastic model of the images, which not only provides the theoretical explanations necessary, but also leads to insights about various other observations from the literature. This model also allows us to investigate how certain changes in the image statistics could affect the DCT coefficient distributions.

Standard cell layout with regular contact placement

The practicability and methodology of applying regularly placed contacts on layout design of standard cells are studied. The regular placement enables more effective use of resolution enhancement technologies, which in turn allows for a reduction of critical dimensions. Although placing contacts on a grid adds restrictions during cell layout, overall circuit area can be made smaller by a careful selection of the grid pitch, allowing slight contact offset, applying double exposure, and shrinking the minimum size and pitch. The contact level of 250 nm standard cells was shrunk by 10%, resulting in an area change ranging from -20% to +25% with an average decrease of 5% for the 84 cells studied. The areas of two circuits, a finite-impulse-response (FIR) filter and an add-compare-select (ACS) unit in the Viterbi decoder, decrease by 4% and 2%, respectively.

Performance optimization for gridded-layout standard cells

The grid placement of contacts and gates enables more effective use of resolution enhancement techniques, which in turn allow a reduction of critical dimensions. Although the regular placement adds restrictions during cell layout, the overall circuit area can be made smaller and the extra manufacturing cost can be kept to the lowest by a careful selection of the grid pitch, using template-trim lithography method, allowing random contact placement in the vertical direction, and using rectangular rather than square contacts. The purpose of this work is to optimize the gridded-layout-based process. The trade-off between the layout area and manufacturing cost, and the determination of the minimum grid pitch are discussed in this paper. We demonstrate that it is a 1-D scaling instead of the conventional 2-D scaling for standard cells and the narrow MOSFETs inside after the application of the gridded layout on the contact and gate levels. The corresponding effects on circuit performances, including the leakage current, are also explored.

A Total Variation Regularization Based Super-Resolution Reconstruction Algorithm for Digital Video

Super-resolution (SR) reconstruction technique is capable of producing a high-resolution image from a sequence of low-resolution images. In this paper, we study an efficient SR algorithm for digital video. To effectively deal with the intractable problems in SR video reconstruction, such as inevitable motion estimation errors, noise, blurring, missing regions, and compression artifacts, the total variation (TV) regularization is employed in the reconstruction model. We use the fixed-point iteration method and preconditioning techniques to efficiently solve the associated nonlinear Euler-Lagrange equations of the corresponding variational problem in SR. The proposed algorithm has been tested in several cases of motion and degradation. It is also compared with the Laplacian regularization-based SR algorithm and other TV-based SR algorithms. Experimental results are presented to illustrate the effectiveness of the proposed algorithm.

Mobile-Phone Antenna Design

This paper is a survey of internal antennas in mobile phones from 1997 to 2010. It covers almost 60 GSM and 3G handsets, ranging from the first GSM handset with an internal antenna to the current Nokia, Sony-Ericsson, Motorola, and Apple handsets. The paper discusses different types of mobile-phone antennas, feeding structures, active antennas, isolation, and antenna loading techniques. This paper examines different design techniques for mobile-phone antennas, and the limitations of antenna design due to manufacturing technologies and the effect of handset materials. Antenna performance parameters, including S parameters, radiation efficiency, SAR, and TRP/TIS are reported for the surveyed handsets. The effective antenna volume for every antenna is calculated, in order to determine the average volume/space required for each antenna type and the corresponding performance. Some of the handsets are further simulated using commercial electromagnetic simulators to illustrate the electromagnetic-field distributions. This paper summarizes the antenna design parameters as a function of handset performance, and presents a short summary of design procedure.

Extending the Depth of Field in a Compound-Eye Imaging System with Super-Resolution Reconstruction

Optical device miniaturization is highly desirable in many applications. Direct down-scaling of traditional imaging system is one approach, but the extent to which it can be minimized is limited by the effect of diffraction. Compound-eye imaging system, which utilizes multiple microlenses in image capture is a promising alternative. In this paper, we explore the possibility of an incorporation of phase masks in such a system to extend the depth of field. Simulation experiments are conducted to verify the feasibility of the system

Investigation of Computational Compound-Eye Imaging System With Super-Resolution Reconstruction

Despite the emerging architectural designs of compound-eye imaging systems, the post-processing algorithms for the reconstruction of the final image from the multiple sub-images is still not fully developed to maturity, resulting in poor quality or low resolution of the reconstructed images. In this paper, we describe and investigate a practical computational compound-eye imaging system with super-resolution reconstruction. This methodology can enhance the image quality by increasing the resolution of the reconstructed image. A virtual compound-eye camera is built to demonstrate the feasibility of the system. Simulation results which investigate the tolerance of the system to lens diversity, for instance focal length and aberrations, are also presented

Combining gray world and retinex theory for automatic white balance in digital photography

White balancing is an important step in the digital camera processing pipeline to adjust the color of the pixels under different illuminations. Efficient automatic white balance is usually a required component of a consumer digital camera because many users would not prefer to handle this task manually. Gray world assumption and retinex theory are two common methods used, but their aims vary and their applicability depends on the nature of the images. In this paper, we present an effective technique that combines the two together, while preserving the strength of the two methods. Experimental results confirm that our approach is a viable alternative to the two existing methods.

Automatic source camera identification using the intrinsic lens radial distortion

Source camera identification refers to the task of matching digital images with the cameras that are responsible for producing these images. This is an important task in image forensics, which in turn is a critical procedure in law enforcement. Unfortunately, few digital cameras are equipped with the capability of producing watermarks for this purpose. In this paper, we demonstrate that it is possible to achieve a high rate of accuracy in the identification by noting the intrinsic lens radial distortion of each camera. To reduce manufacturing cost, the majority of digital cameras are equipped with lenses having rather spherical surfaces, whose inherent radial distortions serve as unique fingerprints in the images. We extract, for each image, parameters from aberration measurements, which are then used to train and test a support vector machine classifier. We conduct extensive experiments to evaluate the success rate of a source camera identification with five cameras. The results show that this is a viable approach with high accuracy. Additionally, we also present results on how the error rates may change with images captured using various optical zoom levels, as zooming is commonly available in digital cameras.

Three-dimensional microscopy and sectional image reconstruction using optical scanning holography

Fast acquisition and high axial resolution are two primary requirements for three-dimensional microscopy. However, they are sometimes conflicting: imaging modalities such as confocal imaging can deliver superior resolution at the expense of sequential acquisition at different axial planes, which is a time-consuming process. Optical scanning holography (OSH) promises to deliver a good trade-off between these two goals. With just a single scan, we can capture the entire three-dimensional volume in a digital hologram; the data can then be processed to obtain the individual sections. An accurate modeling of the imaging system is key to devising an appropriate image reconstruction algorithm, especially for real data where random noise and other imaging imperfections must be taken into account. In this paper we demonstrate sectional image reconstruction by applying an inverse imaging sectioning technique to experimental OSH data of biological specimens and visualizing the sections using the OSA Interactive Science Publishing software.