Palmprint feature extraction using 2-D Gabor filters

Biometric identification is an emerging technology that can solve security problems in our networked society. A few years ago, a new branch of biometric technology, palmprint authentication, was proposed (Pattern Recognition 32(4) (1999) 691) whereby lines and points are extracted from palms for personal identification. In this paper, we consider the palmprint as a piece of texture and apply texture-based feature extraction techniques to palmprint authentication. A 2-D Gabor filter is used to obtain texture information and two palmprint images are compared in terms of their hamming distance. The experimental results illustrate the effectiveness of our method.     

基于多小波基的联合图像去噪方法

小波广泛应用于图像去噪,多个小波基联合运用可以把小波的光滑性、紧支性、正交性、对称性等结合起来。首先用多个小波基分别对含噪图像进行分解、阈值处理和重构,得到多幅重构图像,然后对这些图像进行算术平均,得到最终去噪图像。实验结果表明,该方法去噪效果比单一的小波基方法有明显的改善。     

Image denoising with complex ridgelets

In this paper, we propose a novel image denoising method by incorporating the dual-tree complex wavelets into the ordinary ridgelet transform. The approximate shift invariant property of the dual-tree complex wavelet and the high directional sensitivity of the ridgelet transform make the new method a very good choice for image denoising. We apply the digital complex ridgelet transform to denoise some standard images corrupted with additive white noise. Experimental results show that the new method outperforms VisuShrink, the ordinary ridgelet image denoising, and wiener2 filter both in terms of peak signal-to-noise ratio and in visual quality. In particular, our method preserves sharp edges better while removing white noise. Complex ridgelets could be applied to curvelet image denoising as well.     

Image denoising feedback framework using split Bregman approach

In this paper, an image denoising feedback framework is proposed for both color and range images. The proposed method works on an error minimization principle using split Bregman method. At first image is denoised by computing means in the local neighborhood. The pixels that have big differences from the center of the local neighborhood compared to the noise variance are then extracted from the denoised image. There is a low correlation between the extracted pixels and their local neighborhood. This information is fed to the feedback function and denoising is performed again, iteratively, to minimize the error. In most cases, the proposed framework yields best results both qualitatively and quantitatively. It shows better denoising results than the bilateral filtering when the edge information in the input images is affected by intense noise. Moreover, during the denoising process feedback function ensures that the edges are not over smoothed. The proposed framework is applied to denoise both color and range images, which shows it works effectively on a wide variety of images unlike the evaluated state-of-the-art denoising methods.     

Improved image denoising via RAISR with fewer filters

In recent years,accurate Gaussian noise removal has attracted considerable attention for mobile applications,as in smart phones.Accurate conventional denoising ...     

Online self-repair of FIR filters

Chip-level failure detection has been a target of research for some time, but today's very deep-submicron technology is forcing such research to move beyond detection. Repair, especially self-repair, has become very important for containing the susceptibility of today's chips. This article introduces a self-repair-solution for the digital FIR filter, one of the key blocks used in DSPs.     

Image denoising with norm weighted fusion estimators

In recent era, the weighted matrix rank minimization is used to reduce image noise, promisingly. However, low-rank weighted conditions may cause oversmoothing or oversharpening of the denoised image. This demands a clever engineering algorithm. Particularly, to remove heavy noise in image is always a challenging task, specially, when there is need to preserve the fine edge structures. To attain a reliable estimate of heavy noise image, a norm weighted fusion estimators method is proposed in wavelet domain. This holds the significant geometric structure of the given noisy image during the denoising process. Proposed method is applied on standard benchmark images, and simulation results outperform the most popular rivals of noise reduction approaches, such as BM3D, EPLL, LSSC, NCSR, SAIST, and WNNM in terms of the quality measurement metric PSNR (dB) and structural analysis SSIM indices.     

Image denoising using statistical model based on quaternion wavelet domain

Image denoising is the basic problem of image processing.Quaternion wavelet transform is a new kind of multiresolution analysis tools.Image via quaternion wavelet transform,wavelet coefficients both in intrascale and in interscale have certain correlations.First,according to the correlation of quaternion wavelet coefficients in interscale,non-Gaussian distribution model is used to model its correlations,and the coefficients are divided into important and unimportance coefficients.Then we use the non-Gaussian distribution model to model the important coefficients and its adjacent coefficients,and utilize the MAP method estimate original image wavelet coefficients from noisy coefficients,so as to achieve the purpose of denoising.Experimental results show that our algorithm outperforms the other classical algorithms in peak signal-to-noise ratio and visual quality.     

非下采样剪切波域的临近支持向量机去噪方法

非下采样剪切波(nonsubsampled Shearlet)是一种优秀的多尺度几何分析工具,其不仅可以检测到所有奇异点,而且能够自适应跟踪奇异曲线方向。基于非下采样剪切波,提出了一种使用带有一致性的临近支持向量机(a Proximal Classifier with Consistency,PCC)的图像去噪算法。首先,应用非下采样剪切波把含噪图像分解成不同尺度不同方向的子带;其次,非下采样剪切波系数通过PPC训练被分成两类(无噪系数和噪声系数);最后应用自适应阈值对含噪系数进行去噪。仿真实验结果表明,本文算法不仅拥有较强的抑制噪声能力,而且具有较好的边缘保护能力。     

A Neural Network Approach for Designing 2-D FIR Filters with Arbitrary Magnitude Responses

In this paper, a Neural Network Approach （NNA） is presented for the design of two-dimensional （2-D） Finite-Impulse Response （FIR） filters with arbitrary magnitude responses. Its main goal is to minimize the frequency-domain error function. To illustrate the feasibility of the NNA, a Back.Propagation Neural Network （BPNN） model is chosen, and the convergence and stability of the BPNN are studied. The method avoids matrix inversion operation, and makes a very fast calculation of the filter＇s coefficients possible when the learning rate of the BPNN is selected suitably. The novel feature of the proposed approximation approach is that it can be used to design any kind of 2-D linear-phase FIR filters without any symmetry constraint on their magnitude-frequency responses. Several optimal design examples are given to illustrate the effectiveness of the proposed approach.     

Image denoising in undecimated dual-tree complex wavelet domain using multivariate t-distribution

Multimedia Tools and Applications - Denoising of natural images is a basic problem in image processing. The present paper proposes a new algorithm for image denoising based on the maximum...     

二维经验模态分解域的新型HMT模型图像去噪

二维经验模态分解(Bidimensional Empirical Mode Decompositio,BEMD)是一种优秀的多尺度几何分析工具,特别适用于非线性、非平稳信号的分析处理。以BEMD与新型隐马尔可夫树(Hidden Markov Tree,HMT)模型理论为基础,提出了一种基于BEMD的新型HMT模型的图像去噪算法。该算法的基本思想是,首先对含噪图像进行BEMD变换,然后采用新型HMT模型对BEMD系数进行建模,并通过期望最大(EM)算法对图像BEMD的HMT模型参数进行估计,最后对训练后的BEMD系数进行逆变换,以获得去噪图像。仿真实验结果表明,该算法不仅拥有较强的抑制噪声能力,而且具有较好的边缘保护能力,其整体性能优于现有HMT图像去噪方案。     

FIR filters for online trajectory planning with time- and frequency-domain specifications

In this paper, the use of FIR (Finite Impulse Response) filters for planning minimum-time trajectories for robots or automatic machines under constraints of velocity, acceleration, etc. is presented and discussed. In particular, the relationship between multi-segment polynomial trajectories, i.e. trajectories composed of several polynomial segments, each one possibly characterized by constraints on one or more specific derivatives (i.e. velocity, acceleration, jerk, etc.), and FIR filters disposed in a cascade configuration is demonstrated and exploited in order to design a digital filter for online trajectory planning. The connection between analytic functions and dynamic filters allows a generalization of these trajectories, usually obtained by second- or third-order polynomial functions (e.g. trapezoidal velocity and double S velocity trajectories), to a generic order with only a modest increase of the complexity. As a matter of fact, the computation of trajectories with higher degree of continuity simply requires additional FIR filters in the chain. Moreover, the modular structure of the planner provides a direct frequency characterization of the motion law. In this way, it is possible to define the trajectories by considering constraints expressed in the frequency-domain besides the classical time-domain specifications, such as bounds on velocity, acceleration, and so on. Two examples illustrate the main features of the proposed trajectory planner, in particular with respect to the problems of multi-point trajectories generation and residual vibrations suppression.     

Image authentication based on perceptual hash using Gabor filters

Image authentication is an important topic in image forensics, which tells whether an image is tampered or not or even tells the tampered regions. To implement image authentication, image hash techniques have been reported recently. In this paper, we investigate existing image hash algorithms, and design an novel image hash based on human being's visual system. In this algorithm, we capture the perceptual characters of the image using Gabor filter which can sense the directions in the image just like human’s primary visual cortex. For a given image, we compute the reference scale, direction and block to make sure the final hash can resist against rotation, scale, and translation attacks while maintain the sensitivity to local malicious manipulations. In addition, it has another promising ability to locate the tampered image blocks, and approximately determining the type of tampering methods (delete, add, modify) and the original direction of each block. This ability is very useful in forensics. The experimental results show that the strategy of the reference metrics works quite well and our method is much more effective than the other state of art image hash methods. Moreover, our method can still locate the content-altering changes even undergo some content-preserving manipulations.     

二维条形码的识别及应用

Datamatrix是矩阵式二维条形码之一,由于其尺寸小、保密性好、纠错能力强而广泛使用在工业、数据安全及其他行业.在实际应用中,由扫描设备得到的二维条码图像通常有噪声、污损及背景干扰,并具有平移、旋转.为提高实际应用中的正确译码率,通过预处理和两次图像定位,从实际图像中分割出Datamatrix图像,再从中读取数据,进行正确译码,完成对Datamatrix的识别.     

Computing 2-D min, median, and max filters

Fast algorithms for computing min, median, max, or any other order statistic filter transforms are described. The algorithms take constant time per pixel to compute min or max filters and polylog time per pixel, in the size of the filter, to compute the median filter. A logarithmic time per pixel lower bound for the computation of the median filter is shown     

Parameterization of 3-channel non-separable 2-D wavelets and filters

We propose a complete parameterization presentation of the 3-channelbivariate non-separable orthogonal FIR filter, and describe the sufficient condition ofgenerating continuous wavelet bases. Given the results above, a non-separable,compactly supported, orthogonal, continuous parameterized bivariate wavelet bank is setup here.