Search

The color appearance of a food image is very important for making it look delicious. Many food images acquired in low illumination environments do not give delicious impressions because of the deterioration of their color appearance. We propose an image enhancement method to improve the color appearance of a food image acquired in a low illumination environment. In the proposed method, the lightness of each pixel is first enhanced within a color gamut on an equal hue plane in the CIELAB color space. The boundary of the color gamut is given by a lookup table calculated in advance. Then, when the hue of the pixel is attributed to foods, the chroma is enhanced within the color gamut. In this case, we consider Helmholtz–Kohlrausch (H-K) effect, which indicates the relation of the hue, chroma, and lightness in human visual perception. According to H-K effect, the perceived lightness increases as the chroma is enhanced. Therefore, the perceived lightness is preserved after the chroma enhancement, so that the food image is not too enhanced. To verify the effectiveness of the proposed method, experiments are conducted using several food images.

Generally, in color image contrast enhancement, the intensity or saturation component of an image is processed without the change of hue component of the image. In some color space used for contrast enhancement, processed pixel values are located outside of the gamut of the RGB color space. In that case, a clipping process is required to make the pixel values located inside of the gamut. However, the clipping process changes the combination of R, G, and B components and causes unexpected hue changes. In the present paper, we propose a hue-preserving contrast enhancement method for color images without gamut problem. In the proposed method, the contrast enhancement of a color image is realized by spreading out the distribution of pixel values on a constant hue plane in the RGB color space. The constant hue plane is represented as the triangle whose vertices correspond to white, black, and the maximally saturated colors. By spreading out the distribution of the pixel values on the constant hue plane, a remedy for the gamut problem and hue-preserving contrast enhancement are realized. Through experiments, the effectiveness of the proposed method is verified.

White balance (WB) is the correction process to remove unrealistic color casts caused by an illuminant color. For high-quality image correction, it is important to accurately estimate the illuminant color. In this study, we propose an accurate illuminant color estimation method for WB based on pseudo-detection of specular reflection. In the proposed method, pixels which are brighter than their neighborhood and their neighboring darker pixels are found at first. Then, the averages of their pixel values are calculated respectively, and then the specular reflection is detected approximately as the difference of the average values. Finally, the illuminant color is estimated as the specular reflection itself. The accurate estimation capability of the proposed method is verified through a series of comparative experiments with typical WB methods.

As the filter which can effectively remove the small amplitude noises on digital images, the e - filter has been proposed. In order to effectively use this filter, a smoothing parameter e - filter should be appropriately estimated before applying it. To address this problem, the authors proposed the parameter estimation method based on Hellinger distance (HD). In the method, HD between a residual signal and assumed noise distribution was evaluated, and a parameter e of the e - filter was estimated by finding the value giving minimum distance. However, the enough discussion on use of HD has not been made. In this paper, it is attempted to utilize not only the HD, but also various distribution distances in the parameter estimation, and their performances and characteristics are compared and analyzed experimentally. Furthermore, the parameter estimation method is extended to be applicable for the vector e - filter for the color images. Consequently, through the experiments, it is shown that L1-norm or maximum norm is appropriate as the distribution distance used in the parameter estimation methods from the view points of the simplicity of the calculation and MSE performance in the filtering.

This paper describes a novel image filtering method for removal of random-valued impulse noise superimposed on color images. In color image filtering, generally, it is preferable to deal with the R, G, and B components of each pixel of a color image as elements of a vectorized signal as in the vector median filter rather than component-wise signals in order to prevent color shift after filtering. Furthermore, in impulse noise removal, it is essential to use switching type filtering manner as in the switching median filter to preserve the detailed part of an original image with good quality. By taking those fundamentals in consideration, in this study, we propose a switching-type vector median filter with non-local processing mainly consists of a noise detection part and a noise removal one. Concretely, we propose a noise detector which proactively finds out noise-corrupted pixels by focusing attention on isolation tendencies of pixels of interest in difference images between RGB components. Furthermore, as a noise removal filter, we propose an extended version of Non-local median filter (NL-MF), which we proposed previously for grayscale images, named Non-local vector median filter (NL-VMF) which deals with color images. The proposed method realizes superior balance between the detail-preservation based on the proactive noise detection and the noise removal by the non-local switching vector median filtering. The effectiveness and the validity of the proposed method are verified by some experiments using natural color images.

A new recoloring method to improve visibility of indiscriminable colors for protanopes or deuteranopes is proposed. In the proposed method, yellow-blue components of a color image perceived by protanopes/deuteranopes are adequately modified. Moreover, the gamut mapping is considered to obtain proper output color values in this method.

In this letter, we propose a new tuning method of e value, which is a parameter in the e-filter, using a metric between signal distributions, i.e., Hellinger distance. The difference between the input and output signals is evaluated using Hellinger distance and used for the parameter tuning in the proposed method.

In this paper we propose a radial basis function network (RBFN) based nonlinear filter with a basic framework of a linear Wiener filter. In addition, in order to improve the filtering performance, we further propose a novel nonlinear filter, which is synthesized by a hybridization of an RBFN filter and a linear Wiener filter. The proposed filters are realized with a least mean square error scheme using higher order statistics of a target signal and an observation noise. The validity and the effectiveness of the proposed filters have been verified by applying them to the actual filtering problems of the noisy images.

The estimation of the point-spread function (PSF) is one of very important and indispensable tasks for the practical image restoration. Especially, for the motion blur, various PSF estimation algorithms have been developed so far. However, a majority of them becomes useless in the low blurred signal-to-noise ratio (BSNR) environment. This paper describes a new robust PSF estimation algorithm based on Hough transform concerning gradient vectors, which can accurately and robustly estimate the motion blur PSF even in low BSNR case. The effectiveness and validity of the proposed algorithm are verified by applying it to the PSF estimation and the image restoration for noisy and motion blurred images.

The independent component analysis (ICA) in the frequency domain is a method to deal with a blind signal separation problem in which propagation time delays are included in the mixing process of signals. We propose an extended method of the frequency-domain ICA accompanying the estimation of the relative propagation time delays and the propagation coefficient ratios. The effectiveness of the proposed method has been confirmed by simulation experiments. In addition, the sound localization by the proposed method is further discussed.

A generalized Hough transform is an effective method for an arbitrary shape detection in a contour image. However, the conventional generalized Hough transform is not suitable for a noisy and blurred image. This paper describes a generalized fuzzy Hough transform which is derived by fuzzifying the vote process in the Hough transform. The present generalized fuzzy Hough transform enables a detection of an arbitrary shape in a very noisy, blurred, and even distorted image. The effectiveness of the present method has been confirmed by some preliminary experiments for artificially produced images and for actual digital images taken by an ordinary digital camera

A new natural gradient-based algorithm for a blind deconvolution of blurred images is proposed. In the proposed algorithm, phase spectral constraint condition is newly introduced, and a natural gradient descent method on an amplitude spectrum is applied. The effectiveness of the proposed algorithm is confirmed by comparing with the conventional method.

In this letter, a novel color removal method considering differences of colors in an input color image and achromatic color preservation is proposed. The achromatic color preservation is assigning lightness values to gray-levels concerning achromatic pixels for natural impression. The effectiveness and validity of the proposed method are verified by experiments.

In this paper, we propose a novel nonlinear filter, with a basic framework of the Wiener filter, based on a radial basis function (RBF) network. The proposed filter is synthesized with a least mean square scheme using higher order statistics of a target signal and an observation noise. The validity and effectiveness of the proposed filter are verified by applying it to the actual filtering problem of the noisy images.

A switching median filter (SMF) is effective for impulse noise removal while still preserving edges in an input image. This filter firstly detects pixels corrupted by impulse noise and then filters only the noise-corrupted pixels. However the noise detection process does not always work perfectly. Particularly, pixels constituting thin lines in an input image tend to be incorrectly detected as noise-corrupted pixels, and such pixels are filtered despite the needlessness of the filtering. As the result of the filtering, the image might be over-smoothed and be deteriorated throughout the entire image. To cope with this problem, we propose a new impulse noise removal method based on a one-dimensional SMF and a space-filling curve which reflects structural contexts of an input image. The effectiveness of the proposed method is verified by some experiments.

The bilateral filter can remove Gaussian noise while preserving edges of the objects in an image. However the bilateral filter can not remove impulsive noise due to its edge preservation property. Therefore some robust bilateral filters have been proposed in order to also deal with impulsive noise. To make the bilateral filter more robust, this paper proposes a trilateral filter which incorporates a third weighting function based on the rank order information of pixel values into the bilateral filter. The effectiveness of the proposed method is verified in comparison with other conventional methods in experiments using the natural digital images corrupted by mixed Gaussian and impulsive noise.

In this paper, we propose a method for coronary plaque boundary calculation in Intravascular Ultrasound (IVUS) image by using a modified Perona Malik Diffusion (PMD) filter and the Takagi-Sugeno (T-S) fuzzy model. The modified PMD filter is designed based on the coronary plaque boundary direction in IVUS image to reduce the speckle noise and to enhance the coronary plaque boundary. Searching areas for the plaque boundaries are automatically set by using the weighted image separability and some heuristic rules. The coronary plaque boundaries are interpolated by the polynomials inferred by the fuzzy rules. It has been confirmed that the accuracy of the proposed method is better than that of the method using the normal PMD filter.

In this paper, we present a method for parameter tuning of membership functions in Takagi-Sugeno (T-S) fuzzy model using Particle Swarm Optimization (PSO). This is applied to plaque boundary extraction in Intravascular Ultrasound (IVUS) image. Searching areas for coronary plaque boundaries are automatically set by using weighted image separability and some heuristic rules. The coronary plaque boundaries are interpolated by polynomials inferred by fuzzy rules. PSO tunes the parameters of the membership functions in the antecedent parts of the fuzzy rules. The accuracy of the proposed method is better than that of our previous method.

In this paper, in order to evaluate quantitatively the stochastic characteristics of the fluctuating signal, a generalized n-th order fluctuation rate index is first proposed. Then, the high order equivalence relationship between two time constants of the ideal and the exponential type mean square operations is derived based on the proposed n-th order fluctuation rate index. More concretely, this equivalence relationship is newly discussed especially concerning the third order fluctuation rate index which occupies the first fundamental position in case of hierarchically evaluating the non-Gaussian property, and the well-known equivalence relationship T≅2γ(T: time constant of an ideal type mean squaring operation, γ: time constant of an exponential type mean squaring operation) is again found in a special case with n=2 from the different viewpoint in the time domain under the above same fluctuation rate index. The effectiveness of the proposed theory has been confirmed well experimentally by the applications to the actual sound data.

This paper describes a fuzzy rule-based simple interpolation algorithm for discrete data. A simple interpolation algorithm between two noise-free points in the Euclidean space is proposed by taking into account the fuzzy effects of the surrounding points. The advanced algorithm for the noisy points is proposed, where interpolation is done by classifying the neighboring noisy points into some groups. The validity and the effectiveness of these methods have been verified by computer simulations and by applications to the actual time series noisy data. The experimental results have provided substantial proofs for practical use.

This paper describes a new method of state estimation for a stochastic sound system when the observation system is non-linear and quantized. The method is based on a generalized statistical linearization technique, with use of the two-sided Laplace transform. The non-linear quantized observation system is linearized statistically by using the generalized statistical linearization technique, with the introduction of a new type of error criterion. The resulting linearized system can be obtained analytically by using the two-sided Laplace transform. As a result, linear filtering theory can then be applied to the resulting linearized system. The validity and effectiveness of this method have been confirmed by using a computer simulation and by applying it to the problem of estimating a reverberation time from room acoustics quantized observation data.

To generate a monochrome image from a color one, usually, lightness components of the color image are extracted and used. This transformation is generally not invertible and it is impossible to reconstruct the original color information from a yielded monochrome image. In this paper, we propose an invertible color-to-monochrome transform method obtaining the monochrome image which contains the color information and enables us to reconstruct a color image. In the proposed method, colors in an input image are clustered by k-means-based algorithm with a lightness constraint, and the output monochrome image is obtained using the clustering result.

The back projection for lost pixel (BPLP) is a method which generates the eigenspace reflecting characteristics of an input image and restores missing pixels of the image using it. In this paper, a new image enlargement method is proposed by applying the BPLP to the predefined codebook-based method. In the proposed method, the eigenspace is generated using the predefined codebook consists of a set of low-high frequency component pairs. Then, the image enlargement, which estimates high frequency components fast and preserve the sharpness, is accomplished by the BPLP. Through experiments, the effectiveness of the proposed method is illustrated.

In this paper, a new noise removal method for color images is described. In the proposed method, first, a tentative output image in which noise is almost perfectly removed is obtained using the iterative robust switching vector median-based vector e-filter. Then, the residual components between the input and tentative output images are calculated, and image components constitute edges, corner, and other image details are extracted from the residual components using the correlation characteristic in RGB components. Consequently, a final output is obtained by adding the extracted image components into the tentative output image. The effectiveness of the proposed method is illustrated by some experiments.

In the conventional boundary extraction methods, the medical doctors have to put the several seed points on the IVUS image. However, those methods need some iterative processes and take big computing time. To meet with the practical demand in clinic, a quick boundary extraction by a small number of seed points is definitely necessary. In this paper, to cope with those problems, we propose a new fuzzy rule-based boundary extraction method, which achieves a fast, precise and robust extraction of plaque. In the proposed method, a boundary to be extracted is piecewise approximated by polynomial series inferred by fuzzy rules. The proposed method employs not only the information of the seed points itself but also the separability information of the IVUS image. The validity and the effectiveness of the proposed method have been verified by applying it to the practical problem of plaque boundary extraction in clinic.

In order to convert a color image into a monochrome one, luminance components of the pixels have been used as gray-levels for the representation of the monochrome image in HDTV standard. However, saliencies of the image embedded only in the chrominance components are disappeared in the monochrome image converted by using luminance components. To cope with this problem, A.A. Gooch et al. have proposed the salience-preserving color removal method called ldquoColor2Gray.ldquo In this paper, at first, Color2Gray is analyzed and its mathematical property is revealed. Then, fast Color2Gray algorithm is proposed by use of the mathematical property. Finally, some experiments prove the effectiveness of the proposed algorithm.

A switching median filter is effective for the impulse noise elimination while preserving edges and details of an image. In the switching median filter, an impulse noise detector is employed before filtering, and the detection results are used to control whether the pixels should be filtered or not. However, the conventional impulse detector tends to misjudge noise-free pixels constructing line structures to be the noises. In this paper, a new impulse noise detector based on the minimum spanning tree is proposed, and it is applied to the switching median filtering in order to eliminate the impulse noise effectively even for the image including line structures. Through the experiments, the effectiveness of the proposed impulse nosie detector is illustrated.

In the typical photo re-touching software, in order to convert a color image into a monochrome one, only luminance components of the color image are extracted, and are used to represent the monochrome image. This luminance extraction is useful and enough for obtaining a monochrome image in most cases. However, saliencies of the image included only in the chrominance components are not preserved at all. In this study, we propose a new color removal method that takes difference of the chrominance into account in addition to luminance of an image. The effectiveness of the proposed method is verified by applying it to some images.

The estimation of the point-spread function (PSF) is one of very important and indispensable tasks for the practical image restoration. Especially, for the motion blur, various PSF estimation algorithms have been developed so far. However, a majority of them becomes useless in the highly noisy environment. This paper describes a new robust PSF estimation algorithm based on Hough transform concerning gradient vectors, which can accurately and robustly estimate the motion blur PSF even in highly noisy case. The effectiveness and validity of the proposed algorithm are verified by applying it to the PSF estimation for noisy and motion blurred images.

A novel plaque boundary extraction method for an intravascular ultrasound (IVUS) image employing image separability and Takagi-Sugeno's fuzzy inference model is presented. The membership functions are allocated heuristically with special consideration of the features of the IVUS image. The method has not only improved the accuracy of plaque boundary extraction but has also reduced the workload of medical doctors.

A new image enlargement method for the estimation of high frequency image components lost in the sampling process is proposed. The method uses a codebook consisting of the low- and high-frequency components of an original image itself that is to be enlarged. The effectiveness of the proposed method is verified by comparing with conventional methods.

A new filter based on the statistical independence between signals is presented. The original signal is estimated by the weighted sum of each harmonic of the observation corrupted by noise. The filter is synthesised by adjusting the weighting parameters by learning with the use of higher order moments of observation noise without any information about the dynamics of a target system.

This paper describes a new method for parameter identification for stochastic systems based on a new evaluation measure of the statistical independence between an input signal and extraneous noise. The error criterion used here enables an assessment to be made of the extremities of the distribution function of the random process as well as of the neighborhood of the mean value. Specifically, the joint probability density function (PDF) for an input signal and an extraneous noise of arbitrary distribution type is expanded into an infinite series, with the product of each marginal PDF included in the first expansion term. The expansion coefficients defined by the linear and nonlinear mutual correlations then display with precision and hierarchically the deviation from statistical independence. Several types of estimation methods are derived for evaluating and then decreasing this deviation. The validity and the effectiveness of the proposed methods are experimentally confirmed by applying them to identifying sound insulation parameters in room acoustics.

This paper describes a new attempt at predicting the fluctuations of an environmental noise system by use of adaptive functions. Principally, some kinds of correlations (in some cases, in the form of simple inertia) can be actually found between the recorded past data and the future values. In this paper, the basic correlative relationship of this kind is expressed universally with the use of newly introduced adaptive functions of both the linear, and nonlinear types. In prediction algorithms, discrete or continuous adaptive functions of orthogonal and/or nonorthogonal types are employed to meet the actual observation mechanism and to get the effective data processing. Finally, the effectiveness of the proposed method has been confirmed by computer simulation and application to actually obtained road traffic noise data.