Jin Wei

Bio: Jin Wei is an academic researcher from Ningbo University. The author has contributed to research in topics: Non-local means & Curvelet. The author has an hindex of 2, co-authored 3 publications receiving 8 citations.

Noise removal approach using Curvelet transform

Abstract: Edges in image feature one kind of linear singularity.The aim of Multiscale Geometric Analysis(MGA) which includes Curvelet transform and Ridgelet Transform is to find a kind of optimal representation of such type of image in the sense of nonlinear approximation.Based on Curvelet transform,one image denoising approach is proposed in the paper.Due to the inherent relativity between Curvelet coefficients,this method adopts Curvelet coefficients adaptively with WindowShrink scheme via window/neighborhood processing.Experiments show that this method not only keeps the edge of image but also yields de-noised images with higher PSNR value(PSNR = 29.93 with noise variance σ=25) and better visual quality.

A Robust Adaptive Zero-watermarking Algorithm in DWT-DCT Domain

Abstract: In order to overcome the limitation of the watermarked image quality degradation due to the some changes of the original image data when embedding watermark in the traditional digital watermarking schemes, a robust adaptive zero-watermarking algorithm based on the combined transforms of the Discrete Wavelet Transform (DWT) and Discrete Cosine Transform (DCT) is proposed First, the original digital image is decomposed into the appropriate levels of DWT, and the DCT is applied to the obtained wavelet approximation sub-image, and then the DCT AC coefficients are selected adaptively in orders according to the size of the embedded digital watermark and the coefficient difference value Finally, after the combination of the original binary character watermark and the binary key image which is obtained from comparing the selected DCT AC coefficients, the embedding of zero-watermark is realized by registering to the center of authentication The results demonstrate that the proposed zero-watermarking algorithm not only ensures the watermarked image quality without distortion, but also has very good robustness against the conventional treatments (for example: noise addition, filtering, JPEG compression, rotation) , and is better than the some reported zero-watermarking algorithms

Image Denoising in Multi-wavelet Domain Based on Particle Swarm Optimization

Abstract: An approach of image denoising in multi-wavelet domain based on particle swarm optimization was proposed. Firstly, particle swarm optimization was used to construct the adaptive pre-filters of CL multi-wavelet transform. Then noised image was decomposed by multi-wavelet transform and the coefficients were processed using threshold scheme according to the energy distribution of coefficients. Finally, denoised image could be obtained by inverse multi-wavelet transform. Experiments show, the proposed approach outperforms traditional wavelet denoising methods in terms of PSNR and visual effects. Moreover, the approach is beyond median filter and Wiener filter obviously.

Image denoising method based on curvelet transform

Abstract: We propose a new method for image denoising based on the curvelet transform using cycle spinning. The curvelet transform, a new multiscale transform with the character of anisotropy, was developed from the wavelet transform. It has overcome some inherent limitations of wavelet transform in analyzing signals with dimension higher than 1-D. Experimental results show that the proposed image denoising method based on curvelet transform yields denoised images with higher PSNR and exhibits better perceptual quality than the ones denoised by wavelet transform.

Curvelet Transform and its Application in Seismic Data Denoising

Abstract: Curvelet transform is a new multi-scale transform developed upon wavelet transform. Beside scale and position, its constructive factors still include directions. All these make curvelet transform have a better directional characteristic. Based on these properties, we transform seismic data into curvelet domain, apply a window-shrinking algorithm to attenuate the random noises and improve the quality of seismic data finally. Both model and real data all obtain good results. It is available and necessary to set shrinking window size as 3*3 or 5*5 and the value of σ as 6-7% of maximum amplitude in seismic data denoising.

Self-adaptive step firefly algorithm based robust watermarking method in DWT-SVD domain

Abstract: Digital image watermarking, a data hiding method has become widespread to prohibit illegal use of personal files lately. In this study, a robust Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) based technique is proposed for copyright protection. The cover image is firstly decomposed into sub-bands by DWT. Low frequency sub-band is then divided into non-overlapping blocks. Blocks where watermark will be embedded are selected depending on their standard deviation values. Selected blocks are transformed to U, S, and V matrices by SVD. Watermark is embedded into the second row value in the first column of U component obtained by SVD. Embedding scaling factor is determined by Self-Adaptive Step Firefly Algorithm (SASFA) to balance robustness and transparency that are contradictory to each other. Firefly Algorithm (FA) is a simple, easy to implement, and flexible algorithm but it can pass over the global optimum or get trapped at local optimum. Therefore, SASFA, which constitutes the next step of each firefly based on its previous and present situations is used for global exploration of the solution space. Fibonacci-Lucas Transform (FLT) is applied to binary watermark to provide the security of watermarking scheme. Achieved scrambled watermark bits are used in embedding process. Performance of the proposed scheme is measured by using Bit Error Rate (BER), Normalized Correlation (NC), and Peak Signal-to-Noise Ratio (PSNR). Experimental results prove that the method has high robustness level against attacks and a good visual quality.

A Zero-watermark Method Based on Texture Characteristic of Image Blocks for Stereo Images

Abstract: A zero-watermark method based on the texture characteristic of image block for stereo images is proposed. Different blocks have own textures, and the images can be classified into two types according to the block energies, including smooth and non-smooth. The block energy is calculated by summing of alternative current coefficients. The block type relationships between blocks are employed for constituting the zero-watermark information. The zero-watermark is embedded into images without any modification, and solves the contradiction between transparency and robustness of the watermarking. Experimental results show the transparency and robustness against attacks such as noise, JPEG compression, Gaussian filtering, median and average filtering, and crops.

A new denoising method of SAR images in curvelet domain

Abstract: In this paper a new method of speckle reduction of SAR images in curvelet domain is proposed. In the method, curvelet transform is integrated with wavelet filtering. The new method consists of five parts: preprocessing, curvelet transform (CT), curvelet coefficients processing and two inverse transforms. In the preprocessing step, homomorphic transform is applied to convert multiplicative noise in SAR images to an additive noise which is suitable to be dealt with curvelets. After curvelet transform, curvelet coefficients are thresholded by using soft and hard thresholding functions with improved rules. In hard thresholding rule, noise variations are obtained by using noise parameter estimation. In soft thresholding rule, a classic soft thresholding function and thresholding rule used in wavelet domain is combined with curvelets. Finally, inverse CT and exponential transform are employed to reconstruct denoising image. Comparisons of speckle removing results by using different thresholding methods are also given in this paper. It can be seen that the method presented in the paper is an effective one.