Image denoising algorithm based on even step-length generalized cross validation model

Generalized cross validation (GCV) is a significant mean square error (MSE) estimator. It is widely used for image denoising because it can provide an optimal denoising threshold for these wavelet coefficients of noise image. However, the computational complexity of GCV is higher than that of the universal threshold denoising algorithm. In this study, an efficient and fast image denoising algorithm is proposed based on even step-length (ESL) GCV model. In ESL-GCV model, only the thresholds on even points are calculated from four to the maximum wavelet coefficient. In addition, the ESL-GCV model is optimized using the integer wavelet transform (IWT). These experimental results show that the IWT-based ESL-GCV model can provide lower computational complexity and the better peak signal-to-noise ratio (PSNR) than those of the traditional GCV. The proposed algorithm has important theoretical and practical value for image denoising in the future.     

Parametric Blur Estimation Using the Generalized Cross-Validation Criterion and a Smoothness Constraint on the Image

Parametric blur estimation is proposed using the Generalized Cross-Validation Criterion (GCV) and a smoothness constraint on the image. This approach bypasses the need to identify the image AR parameters. The blur synthesis domains that may be used in the GCV criterion are compared experimentally. The criterion is then applied, in a myopic scheme, on blurred and unblurred images. Results with synthetic and real blurs are discussed.     

Improved image tamper localisation using chaotic maps and self-recovery

In this paper an image tamper localisation scheme is proposed in which authentication bits of a 2 × 2 image block are generated using the chaotic maps. Further the scheme is improved by including a self-recovery method to recover the tampered regions. To improve the quality of the recovered image, two different sets of restoration bits of a block are generated and each one is embedded into randomly selected distinct blocks. The proposed tamper detection scheme performs better than some of the recent schemes proposed by the researchers. The experimental results demonstrate the accuracy and fragility of the tamper detection scheme, and the efficacy of the recovery method.     

A neutrosophic filter for high-density Salt and Pepper noise based on pixel-wise adaptive smoothing parameter

Image indeterminacy has been neglected in most traditional filtering algorithms. This paper proposes a pixel-wise adaptive neutrosophic filter based on neutrosophic indeterminacy feature to remove high-level Salt-and-Pepper noise. In the proposed algorithm, the indeterminacy of a pixel is quantified by a Neutrosophic Set and innovatively exploited as an efficient characteristic of measuring the similarity of pixels. In order to adjust the smoothing parameter of the weight function pixel-wise adaptively, the uncertainty of a pixel is utilized as an indicator of image contents. Extensive experiments on numerous images demonstrate that with a 3 × 3 window, our method outperforms many existing denoising methods in terms of noise suppression and detail preservation.     

Ripplet: A new transform for image processing

Efficient representation of images usually leads to improvements in storage efficiency, computational complexity and performance of image processing algorithms. Efficient representation of images can be achieved by transforms. However, conventional transforms such as Fourier transform and wavelet transform suffer from discontinuities such as edges in images. To address this problem, we propose a new transform called ripplet transform. The ripplet transform is a higher dimensional generalization of the curvelet transform, designed to represent images or two-dimensional signals at different scales and different directions. Specifically, the ripplet transform allows arbitrary support c and degree d while the curvelet transform is just a special case of the ripplet transform (Type I) with c = 1 and d = 2. Our experimental results demonstrate that the ripplet transform can provide efficient representation of edges in images. The ripplet transform holds great potential for image processing such as image restoration, image denoising and image compression.     

Performance evaluation of the fractional wavelet filter: A low-memory image wavelet transform for multimedia sensor networks

Existing image wavelet transform techniques exceed the computational and memory resources of low-complexity wireless sensor nodes. In order to enable multimedia wireless sensors to use image wavelet transforms techniques to pre-process collected image sensor data, we introduce the fractional wavelet filter. The fractional wavelet filter computes the wavelet transform of a 256 × 256 grayscale image using only 16-bit fixed-point arithmetic on a micro-controller with less than 1.5 kbyte of RAM. We comprehensively evaluate the resource requirements (RAM, computational complexity, computing time) as well as image quality of the fractional wavelet filter. We find that the fractional wavelet transform computed with fixed-point arithmetic gives typically negligible degradations in image quality. We also find that combining the fractional wavelet filter with a customized wavelet-based image coding system achieves image compression competitive to the JPEG2000 standard.     

一种基于小波域HMT模型的图像去噪方法研究

小波图像去噪已经成为图像去噪的主要方法之一。利用小波变换在去除噪声时,可提取并保存对视觉起主要作用的边缘信息,但现有的去噪声方法忽略了小波系数之间的相关性。针对这一不足,在小波域隐Markov树模型(HMT) 的基础上提出了一种图像去噪新方法。实验结果表明,与普通的小波去噪方法相比,该方法不但可以保留图像的边缘信息,而且能提高去噪后图像的峰值信噪比。     

一种新的小波图像去噪方法

小波图像去噪已经成为目前图像去噪的主要方法之一，目前的研究主要集中于如何选取阈值使去噪达到较好的效果。边缘信息是图像最为有用的高频信息，在图像去噪的同时，应尽量保留图像的边缘信息，基于这一思想，提出一种新的小波图像去噪方法。用数学形态学算子对图像小波变换后的小波系数进行处理，以去除具有较小支持域的噪声，保留具有连续支持域的边缘。实验结果表明，与普通的小波阈值去噪方法相比，该方法不但可以保留图像的边缘信息，而且能提高去噪后图像的峰值信噪比2～5dB，提高信噪比6～10dB。     

磁共振图像阀值去噪方法研究

磁共振成像已成为脑功能病理和解剖研究的主要手段，是医学影像学领域中最活跃的技术。由于在成像过程中复杂的电磁场环境容易受到人体热噪声干扰，使得磁共振图像去噪成为很重要的研究热点。小波分析具有多尺度分辨和去相关性等特点，在去除被白噪声污染的磁共振图像方面得到了广泛应用。但磁共振图像经传统的小波分析去噪后，细节信息部分丢失，图像的边缘变得模糊．针时这些问题，时经典的小波阀值去噪方法进行了改进，将关键参数取值与预估计联系起来，将阀值的选定与图像的局部特征结合起来，提出一种灵活的、自适应的去噪新方法。与经典方法相比，采用本方法处理的噪声图像去噪后图像的细节更丰富，边缘信息完善，视觉效果更好。     

基于GCV理论的小波自适应多阈值图像去噪方法

小波图像去噪是小波应用较成功的一个领域,而其中最重要的一个环节就是最优阈值的确定问题。提出了一种基于小波变换的多阈值图像去噪的改进方法。这种方法是在各分解尺度上的各子带选择不同的最佳阈值,而最佳阈值的选取是基于GCV理论。利用此理论不仅能获得最小均方意义上的渐进最优解,而且不需要知道噪声的先验知识。通过实验证明,这种方法不仅能获得很好的去噪效果,而且由于不需要对噪声进行估计从而减小了计算量。     

基于平移不变和改进的广义交叉验证准则去噪研究

基于软阈值的广义交叉验证准则(GCV)已运用于图像去噪中．但此方法对图像峰值信噪比(PSNR)的改善有限,不能有效保持细节。针对此问题,改进了广义交叉验证准则．且将平移不变应用于去噪。即先将原图像分别进行水平和垂直方向上的平移,然后将平移后的每幅图像变换到小波域．使用分块广义交叉验证准则对小波系数去噪后实行小波逆变换和反平移。最后将反平移后的各图像平均,即得恢复的图像。实验结果表明,该方法能有效恢复图像细节,图像的PSNR值和主观效果都有较大改善。     

Compressive image broadcasting in MIMO systems with receiver antenna heterogeneity

In this work we consider image delivery in MIMO broadcast networks with diverse channel quality and varying numbers of antennas across receivers. In such systems, performance is normally constrained by the weakest users with either a low channel SNR or only a single receive antenna. To address both dimensions of heterogeneity, we propose a new analog image delivery system that adapts seamlessly along both dimensions simultaneously. Our sender scales the DWT coefficients according to a power allocation strategy, and generates linear combinations of the coefficients using compressive sensing (CS), before transmitting them with amplitude modulation. On the receiving side, the received physical layer symbols are passed directly to the source decoder without conventional MIMO decoding, and the DWT coefficients are recovered using a CS decoder.There are two main contributions of our system. First, integrating CS into MIMO transmission ensures that the reconstructed image quality at the receivers is commensurate with both the channel SNR and the MIMO channel dimension. Second, we introduce a power allocation strategy to achieve a performance tradeoff between receivers with different antenna numbers. Experimental results show that the proposed system outperforms both the analog reference SoftCast and the conventional digital system known as HM-STBC. The average gain is 2.92 dB over SoftCast for single-antenna users and 1.53 dB over HM-STBC for two-antenna users.     

Depth and depth–color coding using shape-adaptive wavelets

We present a novel depth and depth–color codec aimed at free-viewpoint 3D-TV. The proposed codec uses a shape-adaptive wavelet transform and an explicit encoding of the locations of major depth edges. Unlike the standard wavelet transform, the shape-adaptive transform generates small wavelet coefficients along depth edges, which greatly reduces the bits required to represent the data. The wavelet transform is implemented by shape-adaptive lifting, which enables fast computations and perfect reconstruction. We derive a simple extension of typical boundary extrapolation methods for lifting schemes to obtain as many vanishing moments near boundaries as away from them. We also develop a novel rate-constrained edge detection algorithm, which integrates the idea of significance bitplanes into the Canny edge detector. Together with a simple chain code, it provides an efficient way to extract and encode edges. Experimental results on synthetic and real data confirm the effectiveness of the proposed codec, with PSNR gains of more than 5 dB for depth images and significantly better visual quality for synthesized novel view images.     

Genetic algorithm based on discrete wavelet transformation for fractal image compression

In this paper, a genetic algorithm (GA) based on discrete wavelet transformation (DWT) is proposed to overcome the drawback of the time-consuming for the fractal encoder. First, for each range block, two wavelet coefficients are used to find the fittest Dihedral block of the domain block. The similar match is done only with the fittest block to save seven eighths redundant MSE computations. Second, embedding the DWT into the GA, a GA based on DWT is built to fast evolutionary speed further and maintain good retrieved quality. Experiments show that, under the same number of MSE computations, the PSNR of the proposed GA method is reduced 0.29 to 0.47 dB in comparison with the SGA method. Moreover, at the encoding time, the proposed GA method is 100 times faster than the full search method, while the penalty of retrieved image quality is relatively acceptable.     

A novel high-performance reversible data hiding scheme using SMVQ and improved locally adaptive coding method

Reversible data hiding is a method that not only embeds secret data but also reconstructs the original cover image without distortion after the confidential data are extracted. In this paper, we propose novel reversible data hiding scheme that can embed high capacity of secret bits and recover image after data extraction. Our proposed scheme depends on the locally adaptive coding scheme (LAC) as Chang&Nguyen’s scheme and SMVQ scheme. Experimental results show that the compression rate of our proposed scheme is 0.33 bpp on average. To embed secret bits we propose the normal-hiding scheme and the over-hiding scheme which have an average embedding rate of 2.01 bpi and 3.01 bpi, more than that of Chang&Nguyen’s scheme (1.36 bpi). The normal-hiding scheme and the over-hiding scheme also has high embedding efficiency of 0.28 and 0.36 on average, which are better than that of Chang&Kieu’s scheme (0.12), Chang&Nguyen’s scheme (0.18) and Chang&Nguyen’s scheme (0.16).     

Blind self-authentication of images for robust watermarking using integer wavelet transform

In this letter a new technique for robust watermarking of images is proposed based on second-generation wavelets (lifting-based integer wavelet transform). The scheme along with its robustness has got the capability of blind self-authentication of the watermarked images. The watermarked images show no perpetual degrading and gives peak signal to noise ratio (PSNR) in excess of 40 dB due to the use of integer-to-integer transform. Simulation results show the superior performance of the proposed technique as compared to a similar existing scheme under different attacks such as filtering, compression and rotation. Further, due to two independent process of watermark sequence extraction proposed in this letter, any alteration in the image results into dissimilar sequences, thereby detecting tampering in the image.     

Highly enhanced light extraction from organic light emitting diodes with little image blurring and good color stability

We report a highly enhanced light extraction from a top emission organic light emitting diode with little image blurring and color variation with viewing angle. Direct integration of a high refractive index micro lens array on the top of the transparent indium zinc oxide top electrode of a green phosphorescent OLED showed a significant enhancement of light extraction to get EQE of 44.7% from 27.6%, the power efficiency of 134.7 lm/w from 85.9 lm/W and the current efficiency of 217.2 cd/A from 120.7 cd/A without image blurring. In addition, the device showed excellent color stability on viewing angle with Commission Internationale de l’Eclairage (CIE) coordinate of Δx = 0.01, Δy = 0.01 as the viewing angle varied from 0° to 60°.     

An augmented Lagrangian approach to general dictionary learning for image denoising

This paper presents an augmented Lagrangian (AL) based method for designing of overcomplete dictionaries for sparse representation with general l_q-data fidelity term (q ? 2). In the proposed method, the dictionary is updated via a simple gradient descent method after each inner minimization step of the AL scheme. Besides, a modified Iterated Shrinkage/Thresholding Algorithm is employed to accelerate the sparse coding stage of the algorithm. We reveal that the dictionary update strategy of the proposed method is different from most of existing methods because the learned dictionaries become more and more complex regularly. An advantage of the iterated refinement methodology is that it makes the method less dependent on the initial dictionary. Experimental results on real image for Gaussian noise removal (q = 2) and impulse noise removal (q = 1) consistently demonstrate that the proposed approach can efficiently remove the noise while maintaining high image quality.     

A low power dissipation high-speed CMOS image sensor with column-parallel sigma–delta ADCs

A 60frames/s CMOS image sensor with column-parallel inverter-based sigma–delta (ΣΔ) ADCs is proposed in this paper. In order to improve the robustness of the inverter, instead of constant power supply, two buffers are designed to provide power supply for inverters. Instead of using of an operational amplifier, an inverter-based switch-capacitor (SC) circuit is adopted to low-voltage low-power ΣΔ modulator. Detailed analysis and design optimization are provided. Due to the use of the inverter-based ΣΔ ADCs, the conversion speed is improved while reducing the area and power consumption. The proposed CMOS image sensor has been fabricated with 0.18 μm CMOS process. The measurement results show that the random noise (RN) is 7e_rms⁻, the pixel conversion gain is 100 μV/e⁻. Since the measured full well capacity of the pixel is 25000e⁻, the CMOS image sensor achieves a 71 dB dynamic range (DR). The total power consumption at 60frame/s is 58.2 mW.     

基于整数小波变换与快速递推GCV的静止图像去噪

基于整数小波提出一种用于自然景物图像去噪的快速递推广义交叉验证(FR-GCV)算法,首先对图像做整数小波变换(IWT),计算小波图像中小波系数的分布概率,然后利用整数递推,降低了GCV函数计算复杂度。最后通过小波系数下采样和阈值上界限定进一步降低了算法复杂度。实验结果表明,FR-GCV算法对自然景物图像去噪耗时比GCV算法降低了90%以上。FR-GCV算法能够快速求取去噪最优阈值,在静止图像去噪领域具有较强的实际意义。