Matrix-addressed liquid crystal devices, such as liquid crystal television spatial light modulators (LCTV-SLM5) have a suitability for displaying holograms electrically in real time because of their ability to easily display image patterns by electric signals. In our electroholographic system, a CCD camera reads a holographic interference fringe pattern, i.e., a hologram, which is then transmitted
to and subsequently displayed on an LCTV-SLM by video signals. The LCTV-SLM is illuminated with collimated laser light and the image of the electrohologram is reconstructed in real time. Because the carrier frequency of the hologram is very low, a limitation imposed by the relatively low resolution of the CCD camera and the LCTV-SLM, a spatial-filtering technique is applied to remove the zeroth-order diffracted light from the reconstructed image.
CCD signal processing schemes attempt to reduce the
effect of (KTC), 1/f, and broadband noise on the output signal. A number of schemes have been reported over the years. These schemes employ time delay and subtraction to eliminate KTC noise and attenuate 1/f noise. They also include a low-pass function to
reduce the effect of broadband noise. Signal processing schemes include dual-slope integration, correlated-double sampling, a variation of correlated-double sampling referred to as switchedexponential
filtering, and transversal filters. Signal processing that does not use delay and subtraction to eliminate KTC noise is also discussed. A consistent technique is used to analyze the various processing schemes. Transfer functions for signal and noise are presented
for each. Performance comparisons are given with emphasis on their applicability to relatively high speed CCD readout applications (readout rates of 1 Mpixel/s and faster).
A high-speed method for circular object location is proposed. The Hough transform is a robust method for detecting parametrically described curves. However, when more than two parameters exist, evaluation of the possible parameters is very expensive. Through the use ofthe geometricproperty that states that the middle point of the hypotenuse of a right triangle is the center of the circumscribed circle, the proposed method can significantly reduce the time needed for evaluating all possible parameters. Experimental
results are also given to show the effectiveness of the proposed method.
We study the use of mathematical morphology for handprinted character recognition. Our approach uses the morphological skeleton transform as the shape descriptor. An efficient skeletonmatching algorithm, which renders the similarity between two skeletons
as a distance measure, is employed. Based on this distance measure, a character is classified by a minimum distance classifier. The morphological skeleton transform contains complete shape information
and is shown as a powerful descriptor for this class of shapes. We also study the pattern spectrum as a shape descriptor for hand-printed characters. However, the pattern spectrum conveys only information about the shape/size distribution of a given object, which turns out to be not very efficient for hand-printed characters. Experimental results demonstrate the efficiency of the skeletonbased approach and the inadequacy of the pattern-spectrum-based approach.
There are hundreds of different image file specifications in existence. A recent informal suivey recorded almost 100 formats in use by USENET readers alone. Thus, an imaging practitioner is faced with a large and sometimes bewildering range of image file
standards to choose from, which, when coupled with the sparsity of studies in the area, makes acquiring a general overview of the field a difficult task. We seek to address this problem by reviewing the overall topic of image formats, describing the most notable standards, proposing a set of related metrics, and providing a source of further information.
A hyperacuity laser imager utilizes gray source data and overscanned (enhanced sampling reconstruction) gray exposure to achieve fully saturated black-and-white image rendering on the photoreceptor at 4800/in. addressability in both the fast and the slow
scan directions, while maintaining a 400 lines/in. frequency response. Conventional laser printing techniques render bit-mapped images at resolutions finer than required by the human visual system (for example, 600 to 1200 bit/in.) in an attempt to meet a related requirement for edge placement, precision of text, line art, and halftones. Unfortunately, this excessive bit-mapped resolution is inefficient and cumbersome to maintain in a printing system and is stiil not nearly enough to satisfy the human visual system's need for edge precision. On the other hand, a hyperacuity imager has the edge precision necessary to satisfy the human visual system's needs, and trades off excessive resolution to do so. We examine the overscanned, gray exposure characteristics of the hyperacuity laser imager and discuss how it can be viewed as a natural evolution of the current trends in laser printing technology. In addition, an order
of magnitude relationship is established between addressabiity and frequency response for a hyperacuity imager, which is shown to significantly
match the needs of the human visual system.
We investigate a method of ordering pixels (the elements of a rectangular matrix) based on an arithmetic progression with wrap-around (modular arithmetic). For appropriate choices of the progression's parameters, based on a generalization of Fibonacci numbers and the golden mean, we find equidistributed collections ofpixels formed by subintervals of the pixelprogression or "shuffle."
We illustrate this equidistributivity with a novel approach to progressive rendering of a synthetic image, and we suggest several opportunities for its application to other areas of image processing.