Showing papers on "Pixel published in 2002"

Example-based super-resolution

Abstract: We call methods for achieving high-resolution enlargements of pixel-based images super-resolution algorithms. Many applications in graphics or image processing could benefit from such resolution independence, including image-based rendering (IBR), texture mapping, enlarging consumer photographs, and converting NTSC video content to high-definition television. We built on another training-based super-resolution algorithm and developed a faster and simpler algorithm for one-pass super-resolution. Our algorithm requires only a nearest-neighbor search in the training set for a vector derived from each patch of local image data. This one-pass super-resolution algorithm is a step toward achieving resolution independence in image-based representations. We don't expect perfect resolution independence-even the polygon representation doesn't have that-but increasing the resolution independence of pixel-based representations is an important task for IBR.

Fast bilateral filtering for the display of high-dynamic-range images

Abstract: We present a new technique for the display of high-dynamic-range images, which reduces the contrast while preserving detail. It is based on a two-scale decomposition of the image into a base layer, encoding large-scale variations, and a detail layer. Only the base layer has its contrast reduced, thereby preserving detail. The base layer is obtained using an edge-preserving filter called the bilateral filter. This is a non-linear filter, where the weight of each pixel is computed using a Gaussian in the spatial domain multiplied by an influence function in the intensity domain that decreases the weight of pixels with large intensity differences. We express bilateral filtering in the framework of robust statistics and show how it relates to anisotropic diffusion. We then accelerate bilateral filtering by using a piecewise-linear approximation in the intensity domain and appropriate subsampling. This results in a speed-up of two orders of magnitude. The method is fast and requires no parameter setting.

Limits on super-resolution and how to break them

Abstract: Nearly all super-resolution algorithms are based on the fundamental constraints that the super-resolution image should generate low resolution input images when appropriately warped and down-sampled to model the image formation process. (These reconstruction constraints are normally combined with some form of smoothness prior to regularize their solution.) We derive a sequence of analytical results which show that the reconstruction constraints provide less and less useful information as the magnification factor increases. We also validate these results empirically and show that, for large enough magnification factors, any smoothness prior leads to overly smooth results with very little high-frequency content. Next, we propose a super-resolution algorithm that uses a different kind of constraint in addition to the reconstruction constraints. The algorithm attempts to recognize local features in the low-resolution images and then enhances their resolution in an appropriate manner. We call such a super-resolution algorithm a hallucination or reconstruction algorithm. We tried our hallucination algorithm on two different data sets, frontal images of faces and printed Roman text. We obtained significantly better results than existing reconstruction-based algorithms, both qualitatively and in terms of RMS pixel error.

Detection algorithms for hyperspectral imaging applications

Abstract: We introduce key concepts and issues including the effects of atmospheric propagation upon the data, spectral variability, mixed pixels, and the distinction between classification and detection algorithms. Detection algorithms for full pixel targets are developed using the likelihood ratio approach. Subpixel target detection, which is more challenging due to background interference, is pursued using both statistical and subspace models for the description of spectral variability. Finally, we provide some results which illustrate the performance of some detection algorithms using real hyperspectral imaging (HSI) data. Furthermore, we illustrate the potential deviation of HSI data from normality and point to some distributions that may serve in the development of algorithms with better or more robust performance. We therefore focus on detection algorithms that assume multivariate normal distribution models for HSI data.

Hyperspectral image data analysis

Abstract: The fundamental basis for space-based remote sensing is that information is potentially available from the electromagnetic energy field arising from the Earth's surface and, in particular, from the spatial, spectral, and temporal variations in that field. Rather than focusing on the spatial variations, which imagery perhaps best conveys, why not move on to look at how the spectral variations might be used. The idea was to enlarge the size of a pixel until it includes an area that is characteristic from a spectral response standpoint for the surface cover to be discriminated. The article includes an example of an image space representation, using three bands to simulate a color IR photograph of an airborne hyperspectral data set over the Washington, DC, mall.

Drizzle: A Method for the Linear Reconstruction of Undersampled Images

Abstract: We have developed a method for the linear reconstruction of an image from undersampled, dithered data. The algorithm, known as Variable-Pixel Linear Reconstruction, or informally as “Drizzle”, preserves photometry and resolution, can weight input images according to the statistical significance of each pixel, and removes the effects of geometric distortion both on image shape and photometry. This paper presents the method and its implementation. The photometric and astrometric accuracy and image fidelity of the algorithm as well as the noise characteristics of output images are discussed. In addition, we describe the use of drizzling to combine dithered images in the presence of cosmic rays.

Learning gender with support faces

Abstract: Nonlinear support vector machines (SVMs) are investigated for appearance-based gender classification with low-resolution "thumbnail" faces processed from 1,755 images from the FERET (FacE REcognition Technology) face database. The performance of SVMs (3.4% error) is shown to be superior to traditional pattern classifiers (linear, quadratic, Fisher linear discriminant, nearest-neighbor) as well as more modern techniques, such as radial basis function (RBF) classifiers and large ensemble-RBF networks. Furthermore, the difference in classification performance with low-resolution "thumbnails" (21/spl times/12 pixels) and the corresponding higher-resolution images (84/spl times/48 pixels) was found to be only 1%, thus demonstrating robustness and stability with respect to scale and the degree of facial detail.

A Wavelet-Based Algorithm for the Spatial Analysis of Poisson Data

Abstract: Wavelets are scalable, oscillatory functions that deviate from zero only within a limited spatial regime and have average value zero, and thus may be used to simultaneously characterize the shape, location, and strength of astronomical sources. But in addition to their use as source characterizers, wavelet functions are rapidly gaining currency within the source detection field. Wavelet-based source detection involves the correlation of scaled wavelet functions with binned, two-dimensional image data. If the chosen wavelet function exhibits the property of vanishing moments, significantly nonzero correlation coefficients will be observed only where there are high-order variations in the data; e.g., they will be observed in the vicinity of sources. Source pixels are identified by comparing each correlation coefficient with its probability sampling distribution, which is a function of the (estimated or a priori known) background amplitude. In this paper, we describe the mission-independent, wavelet-based source detection algorithm "WAVDETECT," part of the freely available Chandra Interactive Analysis of Observations (CIAO) software package. Our algorithm uses the Marr, or "Mexican Hat" wavelet function, but may be adapted for use with other wavelet functions. Aspects of our algorithm include: (1) the computation of local, exposure-corrected normalized (i.e., flat-fielded) background maps; (2) the correction for exposure variations within the field of view (due to, e.g., telescope support ribs or the edge of the field); (3) its applicability within the low-counts regime, as it does not require a minimum number of background counts per pixel for the accurate computation of source detection thresholds; (4) the generation of a source list in a manner that does not depend upon a detailed knowledge of the point spread function (PSF) shape; and (5) error analysis. These features make our algorithm considerably more general than previous methods developed for the analysis of X-ray image data, especially in the low count regime. We demonstrate the robustness of WAVDETECT by applying it to an image from an idealized detector with a spatially invariant Gaussian PSF and an exposure map similar to that of the Einstein IPC; to Pleiades Cluster data collected by the ROSAT PSPC; and to simulated Chandra ACIS-I image of the Lockman Hole region.

Systematic methods for the computation of the directional fields and singular points of fingerprints

Abstract: The first subject of the paper is the estimation of a high resolution directional field of fingerprints. Traditional methods are discussed and a method, based on principal component analysis, is proposed. The method not only computes the direction in any pixel location, but its coherence as well. It is proven that this method provides exactly the same results as the "averaged square-gradient method" that is known from literature. Undoubtedly, the existence of a completely different equivalent solution increases the insight into the problem's nature. The second subject of the paper is singular point detection. A very efficient algorithm is proposed that extracts singular points from the high-resolution directional field. The algorithm is based on the Poincare index and provides a consistent binary decision that is not based on postprocessing steps like applying a threshold on a continuous resemblance measure for singular points. Furthermore, a method is presented to estimate the orientation of the extracted singular points. The accuracy of the methods is illustrated by experiments on a live-scanned fingerprint database.

Rapid shape acquisition using color structured light and multi-pass dynamic programming

Abstract: This paper presents a color structured light technique for recovering object shape from one or more images. The technique works by projecting a pattern of stripes of alternating colors and matching the projected color transitions with observed edges in the image. The correspondence problem is solved using a novel, multi-pass dynamic programming algorithm that eliminates global smoothness assumptions and strict ordering constraints present in previous formulations. The resulting approach is suitable for generating both high-speed scans of moving objects when projecting a single stripe pattern and high-resolution scans of static scenes using a short sequence of time-shifted stripe patterns. In the latter case, space-time analysis is used at each sensor pixel to obtain inter-frame depth localization. Results are demonstrated for a variety of complex scenes.

Spatial/spectral endmember extraction by multidimensional morphological operations

Abstract: Spectral mixture analysis provides an efficient mechanism for the interpretation and classification of remotely sensed multidimensional imagery. It aims to identify a set of reference signatures (also known as endmembers) that can be used to model the reflectance spectrum at each pixel of the original image. Thus, the modeling is carried out as a linear combination of a finite number of ground components. Although spectral mixture models have proved to be appropriate for the purpose of large hyperspectral dataset subpixel analysis, few methods are available in the literature for the extraction of appropriate endmembers in spectral unmixing. Most approaches have been designed from a spectroscopic viewpoint and, thus, tend to neglect the existing spatial correlation between pixels. This paper presents a new automated method that performs unsupervised pixel purity determination and endmember extraction from multidimensional datasets; this is achieved by using both spatial and spectral information in a combined manner. The method is based on mathematical morphology, a classic image processing technique that can be applied to the spectral domain while being able to keep its spatial characteristics. The proposed methodology is evaluated through a specifically designed framework that uses both simulated and real hyperspectral data.

Segmentation and histogram generation using the HSV color space for image retrieval

Abstract: We have analyzed the properties of the HSV (hue, saturation and value) color space with emphasis on the visual perception of the variation in hue, saturation and intensity values of an image pixel. We extract pixel features by either choosing the hue or the intensity as the dominant property based on the saturation value of a pixel. The feature extraction method has been applied for both image segmentation as well as histogram generation applications - two distinct approaches to content based image retrieval (CBIR). Segmentation using this method shows better identification of objects in an image. The histogram retains a uniform color transition that enables us to do a window-based smoothing during retrieval. The results have been compared with those generated using the RGB color space.

A hierarchical approach to robust background subtraction using color and gradient information

Abstract: We present a background subtraction method that uses multiple cues to detect objects robustly in adverse conditions. The algorithm consists of three distinct levels, i.e., pixel level, region level and frame level. At the pixel level, statistical models of gradients and color are separately used to classify each pixel as belonging to background or foreground. In the region level, foreground pixels obtained from the color based subtraction are grouped into regions and gradient based subtraction is then used to make inferences about the validity of these regions. Pixel based models are updated based on decisions made at the region level. Finally, frame level analysis is performed to detect global illumination changes. Our method provides the solution to some of the common problems that are not addressed by most background subtraction algorithms, such as fast illumination changes, repositioning of static background objects, and initialization of background model with moving objects present in the scene.

High dynamic range display devices

Abstract: A display has a first array of pixels connected to emit individually-controlled amounts of light onto a second array of pixels. The second array of pixels is connected to pass individually-controllable portions of the light. The display has a controller connected to control the pixels of the first and second arrays of pixels. The controller is configured to control the pixels of the first array of pixels to provide on the second array of pixels an approximation of an image specified by image data; determine a difference between the image and the approximation of the image at individual pixels of the second array of pixels; and control the pixels of the second array of pixels to modulate the approximation of the image according to the corresponding determined differences.

VAT: a tool for visual assessment of (cluster) tendency

Abstract: A method is given for visually assessing the cluster tendency of a set of Objects O = {o/sub 1/, . . . ,o/sub n/} when they are represented either as object vectors or by numerical pairwise dissimilarity values. The objects are reordered and the reordered matrix of pair wise object dissimilarities is displayed as an intensity image. Clusters are indicated by dark blocks of pixels along the diagonal.

Semiconductor device and manufacturing method thereof

Abstract: A semiconductor device having high operating performance and reliability, and a manufacturing method thereof are provided. An LDD region 207 provided in an n-channel TFT 302 forming a driving circuit enhances the tolerance for hot carrier injection. LDD regions 217-220 provided in an n-channel TFT (pixel TFT) 304 forming a pixel portion greatly contribute to the decrease in the OFF current value. Here, the LDD region of the n-channel TFT of the driving circuit is formed such that the concentration of the n-type impurity element becomes higher as the distance from an adjoining drain region decreases.

Integral three-dimensional imaging with digital reconstruction

Abstract: A computer-based three-dimensional image reconstruction method and system are presented. An elemental image array of a three-dimensional object is formed by a micro-lens array, and recorded by a CCD camera. Three-dimensional images are reconstructed by extracting pixels periodically from the elemental image array using a computer. Images viewed from an arbitrary angle can be retrieved by shifting which pixels are to be extracted. Image processing methods can be used to enhance the reconstructed image. Further, the digitally reconstructed images can be sent via a network. A system for imaging a three-dimensional object includes a micro-lens array that generates an elemental image array. The elemental image array is detected by a CCD camera to generate digitized image information. A computer processes the digitized image information to reconstruct an image of the three-dimensional object. A two-dimensional display device may be connected directly or indirectly to the computer to display the image of the three-dimensional object. The computer may also be used to generate virtual image information of a virtual three-dimensional object. This can then be combined with the digitized image information to provide combined image information. The two-dimensional display device may be used to display a virtual image or a combined image.

Methods, systems, and programming for computer display of images, text, and/or digital content

Abstract: A bitmap of a shape, such as a font, can be subpixel optimized by producing for each of a display's subpixels a coverage value representing the percent of its area covered by the shape being represented and by distributing, to prevent color imbalance, an amount of a given subpixel's coverage value to nearby subpixels of different colors as a function of the percent of the given subpixel's coverage value that causes color imbalance. Web pages can be displayed with scaled-down and subpixel optimized images. A given layout of a Web page can be displayed at each of at least two different selected scale factors, with the font bitmaps used to represent characters in the display at each scale factor having their shape and pixel alignment selected to improve readability for the particular pixel size at which they are displayed at each such scale factor.

A compact and efficient image retrieval approach based on border/interior pixel classification

Abstract: This paper presents \bic (Border/Interior pixel Classification), a compact and efficient CBIR approach suitable for broad image domains It has three main components: (1) a simple and powerful image analysis algorithm that classifies image pixels as either border or interior, (2) a new logarithmic distance (dLog) for comparing histograms, and (3) a compact representation for the visual features extracted from images Experimental results show that the BIC approach is consistently more compact, more efficient and more effective than state-of-the-art CBIR approaches based on sophisticated image analysis algorithms and complex distance functions It was also observed that the dLog distance function has two main advantages over vectorial distances (eg, L1): (1) it is able to increase substantially the effectiveness of (several) histogram-based CBIR approaches and, at the same time, (2) it reduces by 50% the space requirement to represent a histogram

Interacting at a distance: measuring the performance of laser pointers and other devices

Abstract: It is difficult to interact with computer displays that are across the room. A popular approach is to use laser pointers tracked by a camera, but interaction techniques using laser pointers tend to be imprecise, error-prone, and slow. Although many previous papers discuss laser pointer interaction techniques, none seem to have performed user studies to help inform the design. This paper reports on two studies of laser pointer interactions that answer some of the questions related to interacting with objects using a laser pointer. The first experiment evaluates various parameters of laser pointers. For example, the time to acquire a target is about 1 second, and the jitter due to hand unsteadiness is about ±8 pixels, which can be reduced to about ±2 to ±4 pixels by filtering. We compared 7 different ways to hold various kinds of laser pointers, and found that a laser pointer built into a PalmOS device was the most stable. The second experiment compared 4 different ways to select objects on a large projected display. We found that tapping directly on a wall-size SmartBoard was the fastest and most accurate method, followed by a new interaction technique that copies the area of interest from the big screen to a handheld. Third in speed was the conventional mouse, and the laser pointer came in last, with a time almost twice as long as tapping on the SmartBoard

A feature-based, robust, hierarchical algorithm for registering pairs of images of the curved human retina

Abstract: This paper describes a robust hierarchical algorithm for fully-automatic registration of a pair of images of the curved human retina photographed by a fundus microscope. Accurate registration is essential for mosaic synthesis, change detection, and design of computer-aided instrumentation. Central to the algorithm is a 12-parameter interimage transformation derived by modeling the retina as a rigid quadratic surface with unknown parameters. The parameters are estimated by matching vascular landmarks by recursively tracing the blood vessel structure. The parameter estimation technique, which could be generalized to other applications, is a hierarchy of models and methods, making the algorithm robust to unmatchable image features and mismatches between features caused by large interframe motions. Experiments involving 3,000 image pairs from 16 different healthy eyes were performed. Final registration errors less than a pixel are routinely achieved. The speed, accuracy, and ability to handle small overlaps compare favorably with retinal image registration techniques published in the literature.

Image processing system to control vehicle headlamps or other vehicle equipment

Abstract: An imaging system (fig.1) includes an image array sensor (101) including a plurality of pixels (102). Each pixel (102) generates a signal indicative of the amount of light received on the pixel. The imaging system further includes an analog to digital converter (103) for converting the signals from the pixels into a digital value. The system further includes a memory (106) including a plurality of allocated storage locations for storing the digital values from the converter. The number of allocated storage locations in the memory is less than the number of pixels in the image array sensor.

Image Distortion Correction in EPI: Comparison of Field Mapping With Point Spread Function Mapping

Abstract: Echo-planar imaging (EPI) can provide rapid imaging by acquiring a complete k-space data set in a single acquisition. However, this approach suffers from distortion effects in geometry and intensity, resulting in poor image quality. The distortions, caused primarily by field inhomogeneities, lead to intensity loss and voxel shifts, the latter of which are particularly severe in the phase-encode direction. Two promising approaches to correct the distortion in EPI are field mapping and point spread function (PSF) mapping. The field mapping method measures the field distortions and translates these into voxel shifts, which can be used to assign image intensities to the correct voxel locations. The PSF approach uses acquisitions with additional phase-encoding gradients applied in the x, y, and/or z directions to map the 1D, 2D, or 3D PSF of each voxel. These PSFs encode the spatial information about the distortion and the overall distribution of intensities from a single voxel. The measured image is the convolution of the undistorted density and the PSF. Measuring the PSF allows the distortion in geometry and intensity to be corrected. This work compares the efficacy of these methods with equal time allowed for field mapping and PSF mapping.

Optical reader having partial frame operating mode

Abstract: The invention is an optical reader having a 2D image sensor that is configured to operate in a partial frame capture mode. In a partial frame operating mode, the reader clocks out and captures at least one partial frame of image data having image data corresponding to less than all of the pixels of an image sensor pixel array. In one embodiment, the reader operating in a partial frame operating mode captures image data corresponding to a linear pattern of pixels of the image sensor, reads the image data, attempts to decode for a decodable 1D symbol which may be represented in the image data, and captures a full frame of image data if the image data reading reveals a 2D symbol is likely to be present in a full field of view of the 2D image sensor.

Multispectral color image capture using a liquid crystal tunable filter

Abstract: We describe the experimental setup of a multispectral color image acquisition system consisting of a professional monochrome CCD camera and a tunable filter in which the spectral transmittance can be controlled electronically. We perform a spectral characterization of the acquisition system taking into account the acquisition noise. To convert the camera output signals to device-independent color data, two main approaches are proposed and evaluated. One consists in applying regression methods to convert from the K camera outputs to a device-independent color space such as CIEXYZ or CIELAB. Another method is based on a spectral model of the acquisition system. By inverting the model using a principal eigenvector approach, we estimate the spectral reflectance of each pixel of the imaged surface.

Color texture moments for content-based image retrieval

Abstract: We adopt a local Fourier transform as a texture representation scheme and derive eight characteristic maps for describing different aspects of cooccurrence relations of image pixels in each channel of the (SVcosH, SVsinH, V) color space. Then we calculate the first and second moments of these maps as a representation of the natural color image pixel distribution, resulting in a 48-dimensional feature vector. The novel low-level feature is named color texture moments (CTM), which can also be regarded as a certain extension to color moments in eight aspects through eight orthogonal templates. Experiments show that this new feature can achieve good retrieval performance for CBIR.