A comparison of computational color constancy algorithms. I: Methodology and experiments with synthesized data

We introduce a context for testing computational color constancy, specify our approach to the implementation of a number of the leading algorithms, and report the results of three experiments using synthesized data. Experiments using synthesized data are important because the ground truth is known, possible confounds due to camera characterization and pre-processing are absent, and various factors affecting color constancy can be efficiently investigated because they can be manipulated individually and precisely. The algorithms chosen for close study include two gray world methods, a limiting case of a version of the Retinex method, a number of variants of Forsyth's (1990) gamut-mapping method, Cardei et al.'s (2000) neural net method, and Finlayson et al.'s color by correlation method (Finlayson et al. 1997, 2001; Hubel and Finlayson 2000) . We investigate the ability of these algorithms to make estimates of three different color constancy quantities: the chromaticity of the scene illuminant, the overall magnitude of that illuminant, and a corrected, illumination invariant, image. We consider algorithm performance as a function of the number of surfaces in scenes generated from reflectance spectra, the relative effect on the algorithms of added specularities, and the effect of subsequent clipping of the data. All data is available on-line at http://www.cs.sfu.ca//spl sim/color/data, and implementations for most of the algorithms are also available (http://www.cs.sfu.ca//spl sim/color/code).     

Further results on local stability of REM algorithm with time-varying delays

This letter presents some further results on the local stability in equilibrium for Internet congestion control algorithm proposed by Low et al., (IEEE/ACM Transactions on Networking, 1999). The propagation delay d(t) is assumed to be time-varying and have maximum and minimum delay bounds (i.e., d/sub m//spl les/d(t)/spl les/d/sub M/), which is more general than the assumption (0相似文献   

CSD-a new unified threshold metric of evaluating LCD viewing angle by color saturation degradation

Lower luminance contrast ratio and chromatic changes affect the visual performance (i.e. color shift) of a thin-film transistor liquid-crystal device (TFT-LCD) at large viewing angles. The de facto method of defining viewing angle, contrast ratio of luminance, fails to represent the substantial visual performance viewed at a larger angle. We found the degradation of color saturation, dS*/d/spl theta/, to be an appropriate metric to aid the conventional viewing angle definition (CR /spl ges/ 10). We empirically determined the threshold for defining the color viewing angles of TFT-LCDs, CVA = {/spl theta//spl par/dS*/d/spl theta/| /spl les/ 0.03}, which reflects the variation not only in chromaticity but also in luminance. The proposed metric was evaluated by psychophysical experiments, whose results validate the efficacy of the proposed metric.     

High-accuracy FDTD solution of the absorbing wave equation, and conducting Maxwell's equations based on a nonstandard finite-difference model

We previously introduced high-accuracy finite-difference time-domain (FDTD) algorithms based on nonstandard finite differences (NSFD) to solve the nonabsorbing wave equation and the nonconducting Maxwell equations. We now extend our methodology to the absorbing wave equation and the conducting Maxwell equations. We first derive an exact NSFD model of the one-dimensional wave equation, and extend it to construct high-accuracy FDTD algorithms to solve the absorbing wave equation, and the conducting Maxwell's Equations in two and three dimensions. For grid spacing h, and wavelength /spl lambda/, the NSFD solution error is /spl epsiv//spl sim/(h//spl lambda/)/sup 6/ compared with (h//spl lambda/)/sup 2/ for ordinary FDTD algorithms using second-order central finite-differences. This high accuracy is achieved not by using higher-order finite differences but by exploiting the analytical properties of the decaying-harmonic solution basis functions. Besides higher accuracy, in the NSFD algorithms the maximum time step can be somewhat longer than for the ordinary second-order FDTD algorithms.     

Improving gamut mapping color constancy

The color constancy problem, that is, estimating the color of the scene illuminant from a set of image data recorded under an unknown light, is an important problem in computer vision and digital photography. The gamut mapping approach to color constancy is, to date, one of the most successful solutions to this problem. In this algorithm the set of mappings taking the image colors recorded under an unknown illuminant to the gamut of all colors observed under a standard illuminant is characterized. Then, at a second stage, a single mapping is selected from this feasible set. In the first version of this algorithm Forsyth (1990) mapped sensor values recorded under one illuminant to those recorded under a second, using a three-dimensional (3-D) diagonal matrix. However because the intensity of the scene illuminant cannot be recovered Finlayson (see IEEE Trans. Pattern Anal. Machine Intell. vol.18, no.10, p.1034-38, 1996) modified Forsyth's algorithm to work in a two-dimensional (2-D) chromaticity space and set out to recover only 2-D chromaticity mappings. While the chromaticity mapping overcomes the intensity problem it is not clear that something has not been lost in the process. The first result of this paper is to show that only intensity information is lost. Formally, we prove that the feasible set calculated by Forsyth's original algorithm, projected into 2-D, is the same as the feasible set calculated by the 2-D algorithm. Thus, there is no advantage in using the 3-D algorithm and we can use the simpler, 2-D version of the algorithm to characterize the set of feasible illuminants. Another problem with the chromaticity mapping is that it is perspective in nature and so chromaticities and chromaticity maps are perspectively distorted. Previous work demonstrated that the effects of perspective distortion were serious for the 2-D algorithm. Indeed, in order to select a sensible single mapping from the feasible set this set must first be mapped back up to 3-D. We extend this work to the case where a constraint on the possible color of the illuminant is factored into the gamut mapping algorithm. We show here that the illumination constraint can be enforced during selection without explicitly intersecting the two constraint sets. In the final part of this paper we reappraise the selection task. Gamut mapping returns the set of feasible illuminant maps. Our new algorithm is tested using real and synthetic images. The results of these tests show that the algorithm presented delivers excellent color constancy.     

Numerical aperture of single-mode photonic crystal fibers

We consider the problem of radiation into free space from the end-facet of a single-mode photonic crystal fiber (PCF). We calculate the numerical aperture NA = sin /spl theta/ from the half-divergence angle /spl theta/ /spl sim/ tan/sup -1/ (/spl lambda///spl pi//spl omega/) with /spl pi//spl omega//sup 2/ being the effective area of the mode in the PCF. For the fiber first presented by Knight et al. (1996), we find a NA /spl sim/ 0.07 which compares to standard fiber technology. We also study the effect of different hole sizes and demonstrate that the PCF technology provides a large freedom for NA engineering. Comparing to experiments we find good agreement.     

TFT-LCD显色恒定性与独立性研究

显色恒定性及独立性是对TFT-LCI）高品质彩色图像显示的重要要求之一本文根据色度学和显示器像素空间混色的理论，提出了TFT-LCD）显色恒定性及独立性的理论依据．并通过详细的测试，分析了TFT-LCD实际的显色恒定性及独立性．讨论了可能引起TFT-LCD不能满足显色恒定性及独立性要求的原因。     

Edge-based color constancy.

Color constancy is the ability to measure colors of objects independent of the color of the light source. A well-known color constancy method is based on the gray-world assumption which assumes that the average reflectance of surfaces in the world is achromatic. In this paper, we propose a new hypothesis for color constancy namely the gray-edge hypothesis, which assumes that the average edge difference in a scene is achromatic. Based on this hypothesis, we propose an algorithm for color constancy. Contrary to existing color constancy algorithms, which are computed from the zero-order structure of images, our method is based on the derivative structure of images. Furthermore, we propose a framework which unifies a variety of known (gray-world, max-RGB, Minkowski norm) and the newly proposed gray-edge and higher order gray-edge algorithms. The quality of the various instantiations of the framework is tested and compared to the state-of-the-art color constancy methods on two large data sets of images recording objects under a large number of different light sources. The experiments show that the proposed color constancy algorithms obtain comparable results as the state-of-the-art color constancy methods with the merit of being computationally more efficient.     

Radiometric and spectral performance and calibration of the GHz bands of EOS MLS

This paper describes radiometric performance and prelaunch radiometric and spectral calibrations of the GHz component of the Microwave Limb Sounder (MLS) experiment on NASA's Aura spacecraft. Estimated systematic scaling uncertainties (3/spl sigma/) on limb port radiances are /spl sim/0.5% from radiometric calibration and /spl sim/0.5% to /spl sim/1% from spectral calibrations. Operational noise performance is consistent with prelaunch expectations, and in-orbit measurements to date indicate no changes in noise characteristics, and no observable calibration drifts. Spectral baseline has remained stable to /spl sim/20 mK since launch. Refinements to calibrations based on in-flight data are discussed, and radiometric calibration algorithms are described.     

Dual Image Reversible Data Hiding by Modifying Least-Significant-Bit Matching Revisited Rules

The concept of dual image reversible data hiding (DIRDH) is the technique that can produce two camouflage images after embedding secret data into one original image. Moreover, not only can the secret data be extracted from two camouflage images but also the original image can be recovered. To achieve high image quality, Lu et al.'s method applied least-significant-bit (LSB) matching revisited to DIRDH. In order to further improve the image quality, the proposed method modifies LSB matching revisited rules and applies them to DIRDH. According to the experimental results, the image quality of the proposed method is better than that of Lu et al.'s method.     

）istortion-Free Data Embedding Scheme for Hig Dynamic Range Images

Distortion-free data embedding is a technique which can assure that not only the secret data is correctly extracted but also the cover media is recovered without any distortion after secret data is extracted completely. Because of these advantages, this technique attracts the attention of many researchers. In this paper, a new distortion-free data embedding scheme for high dynamic range (HDR) images is proposed. By depending on Cartesian product, this scheme can obtain higher embedding capacity while maintaining the exactly identical cover image and stego image when using the tone mapping algorithms. In experimental results, the proposed scheme is superior to Yu et al.'s scheme in regard to the embedding rate-an average embedding rate of 0.1355 bpp compared with Yu et al.'s scheme (0.1270 bpp).     

A new class of chromatic filters for color image processing. theory and applications

This paper advances a new framework for chromatic filtering of color images. The chromatic content of a color image is encoded in the CIE u'v' chromaticity coordinates whereas the achromatic content is encoded as CIE Y tristimulus value. Within the u'v' chromaticity diagram, colors are added according to the well-known center of gravity law of additive color mixtures, which is generalized here into a nonlinear filtering scheme for processing the two chromatic signals u' and v'. The achromatic channel Y can be processed with traditional filtering schemes, either linear or nonlinear, depending on the specific task at hand. The most interesting characteristics of the new filtering scheme are: 1) the elimination of color smearing effects along edges between bright and dark areas; 2) the possibility of processing chromatic components in a noniterative fashion through linear convolution operations; and 3) the consequent amenability to computationally efficient implementations with fast Fourier transform. The paper includes several examples with both synthetic and real images where the performance of the new filtering method is compared with that of other color image processing algorithms.     

640 /spl times/ 512 pixel long-wavelength infrared narrowband, multiband, and broadband QWIP focal plane arrays

A 640 /spl times/ 512 pixel, long-wavelength cutoff, narrowband (/spl Delta//spl lambda///spl lambda//spl sim/10%) quantum-well infrared photodetector (QWIP) focal plane array (FPA), a four-band QWIP FPA in the 4-15 /spl mu/m spectral region, and a broadband (/spl Delta//spl lambda///spl lambda/ /spl sim/ 42%) QWIP FPA having a 15.4 /spl mu/m cutoff have been demonstrated. In this paper, we discuss the electrical and optical characterization of these FPAs, and their performance. In addition, we discuss the development of a very sensitive (NEDT /spl sim/ 10.6 mK) 640 /spl times/ 512 pixel thermal imaging camera having a 9 /spl mu/m cutoff.     

Low-voltage high performance compact all cascode CMOS current mirror

A high performance and compact current mirror with extremely low input and high output resistances (R/sub in//spl sim/0.01/spl Omega/, R/sub out//spl sim/10 G/spl Omega/), high copying accuracy, very low input and output voltage requirements (V/sub in/, V/sub out//spl ges/V/sub DSsat/), high bandwidth (200 MHz using a 0.5 /spl mu/m CMOS technology) and low settling time (25 ns) is proposed. Simulations and experimental results are shown that validate the circuit.     

Electron beam direct-write tunable polymeric waveguide grating filter

We report a tunable electron beam direct-write polymeric waveguide Bragg grating filter based on a negative tone epoxy, The waveguide filter, with a 5-mm-long first-order grating, exhibits a transmission peak of -27 dB and a 3-dB bandwidth of /spl sim/0.8 nm, and there is an excellent agreement between experimental data and simulation results. The temperature response of the filter is also characterized. The rate of change of refractive index dn/dT is /spl sim/ -1.8 /spl times/ 10/sup -4///spl deg/1C at 1550-nm wavelength for both transverse electric and transverse magnetic polarizations, and the rate of change of peak wavelength d/spl lambda//dT is /spl sim/ -0.14 nm//spl deg/C. The tuning performance is comparable to other grating devices fabricated using multiple processing steps.     

A sensitivity and linearity improvement of a 100-dB dynamic range CMOS image sensor using a lateral overflow integration capacitor

In a CMOS image sensor featuring a lateral overflow integration capacitor in a pixel, which integrates the overflowed charges from a fully depleted photodiode during the same exposure, the sensitivity in nonsaturated signal and the linearity in saturated overflow signal have been improved by introducing a new pixel circuit and its operation. The floating diffusion capacitance of the CMOS image sensor is as small as that of a four transistors type CMOS image sensor because the lateral overflow integration capacitor is located next to the reset switch. A 1/3-inch VGA format (640/sup H//spl times/480/sup V/ pixels), 7.5/spl times/7.5 /spl mu/m/sup 2/ pixel color CMOS image sensor fabricated through 0.35-/spl mu/m two-poly three-metal CMOS process results in a 100 dB dynamic range characteristic, with improved sensitivity and linearity.     

The impact of uniform strain applied via bonding onto plastic substrate on MOSFET performance

For the first time, this letter presents a novel post-backend strain applying technique and the study of its impact on MOSFET device performance. By bonding the Si wafer after transistor fabrication onto a plastic substrate (a conventional packaging material FR-4), a biaxial-tensile strain (/spl sim/0.026%) was achieved globally and uniformly across the wafer due to the shrinkage of the bonded adhesive. A drain-current improvement (average /spl Delta/I/sub d//I/sub d//spl sim/10%) for n-MOSFETs uniformly across the 8-in wafer is observed, independent of the gate dimensions (L/sub g//spl sim/55 nm -0.530 /spl mu/m/W /spl sim/2-20 /spl mu/m). The p-MOSFETs also exhibited I/sub d/-improvement by /spl sim/7% under the same biaxial-tensile strain. The strain impact on overall device characteristics was also studied, including increased gate-induced drain leakage and short-channel effects.     

Reduced-state representations for trellis codes using constellation symmetry

This paper presents a symmetry-based technique for trellis-code state-diagram reduction that has more general applicability than the quasi-regularity technique of Rouanne et al. and Zehavi et al. for trellis codes using standard constellations and labelings. For a 2/sup /spl nu/x/-state trellis code, the new technique reduces the 2/sup 2/spl nu/x/ state diagram to 2/sup /spl nu/x+/spl nu/q/-state diagram where 0/spl les//spl nu//sub q//spl les//spl nu//sub x/. The particular value of /spl nu//sub q/ depends on the constellation labeling and the convolutional encoder. For standard rate-k/(k+1) set-partitioned trellis codes, /spl nu//sub q/=0, and the overall number of states is the same with the new technique as with quasi-regularity. For codes that are not quasi-regular (and thus not amenable to the quasi-regularity technique), the new technique often provides some improvement (when /spl nu//sub q/相似文献   

Representations of finite fields and the Naini-Seberry partitioning problem

Safavi-Naini and Seberry suggested a subliminal channel based on partitioning a set /spl Gamma/ of the generator matrices of error-correcting codes. We give a systematic procedure to partition /spl Gamma/ into E/sub i/'s with |E/sub i/|=2/sup m/ for any m, which is more efficient than the method in Yang, C-N et al., (1997).     

基于分数阶Retinex的低照度图像增强方法

低照度彩色图像增强在生活中起着重要作用,传统的低照度彩色图像增强算法往往会引起图像的不同程度失真。为了增强低照度彩色图像而又不引起图像失真,本文提出了一种新的低照度图像自适应对比度增强算法。将分数阶微积分、传统Retinex变分法与分段对数变换饱和度增强法相结合,构造一种新的分数阶Retinex图像增强算法。实验结果表明,该方法具有增强图像对比度的同时又能保持边缘和纹理细节的能力。与传统低照度图像增强算法相比,能突出图像的细节纹理信息,同时图像色度和亮度也有明显改善。