Denoising of Document Images using Discrete Curvelet Transform for OCR Applications
نویسندگان
چکیده
In this paper, a denoising and binarization scheme of document images corrupted by white Gaussian noise and Impulse noise is presented using Curvelet Transform. The ability of sparse representation and edge preservation of Curvelet transform is utilized. Impulse noise gets added during document scanning or after binarization of scanned document images. White Gaussian noise corrupts the document images during transmission. The presence of either type of noise or a combination of them can severely degrade the performance of any OCR system. In the proposed denoising scheme, the curvelet transform is used with level dependent threshold calculated by modified sqtwolog method (universal threshold) at each scale with estimation of noise standard deviation. The noisy curvelet coefficients are thresholded by Hard Thresholding method. After curvelet based denoising, the image is binarized using global Otsu method and post processed to smoothen the text boundaries and remove isolated pixels for better OCR performance. The curvelet based scheme is compared with a wavelet transform based scheme and a modified wavelet based scheme with edge preservation. The results show that curvelet based scheme performs better in case of images containing Gaussian, Impulse and a combination of both the noises.
منابع مشابه
An Efficient Curvelet Framework for Denoising Images
Wiener filter suppresses noise efficiently. However, it makes the out image blurred. Curvelet preserves the edges of natural images perfectly, but, it produces visual distortion artifacts and fuzzy edges to the restored image, especially in homogeneous regions of images. In this paper, a new image denoising framework based on Curvelet transform and wiener filter is proposed, which can stop nois...
متن کاملComparison of Real and Complex-valued Versions of Wavelet Transform, Curvelet Transform and Ridgelet Transform for Medical Image Denoising
In this study; medical images were denoising with multiresolution analyses using real-valued wavelet transform (RVWT), complex-valued wavelet transform (CVWT), ridgelet transform (RT), real-valued first-generation curvelet transform (RVFG CT), real-valued second-generation curvelet transform (RVSG CT), complex-valued second-generation curvelet transform (CVSG CT) and results are compared. First...
متن کاملNovel Automated Method for Minirhizotron Image Analysis: Root Detection using Curvelet Transform
In this article a new method is introduced for distinguishing roots and background based on their digital curvelet transform in minirhizotron images. In the proposed method, the nonlinear mapping is applied on sub-band curvelet components followed by boundary detection using energy optimization concept. The curvelet transform has the excellent capability in detecting roots with different orient...
متن کاملHidden Markov Tree Modeling of the Uniform Discrete Curvelet Transform for Image Denoising
The uniform discrete curvelet transform (UDCT) is a modified version of the curvelet transform based on the filter bank implementation in the frequency domain. In this project, the hidden Markov tree model, which captures the parent-children dependency, is used for the uniform discrete curvelet transform. Also, the image denoising results are shown as an application of the modeling.
متن کاملMRI Denoising Using Sparse Based Curvelet Transform with Variance Stabilizing Transformation Framework
We develop an efficient MRI denoising algorithm based on sparse representation and curvelet transform with variance stabilizing transformation framework. By using sparse representation, a MR image is decomposed into a sparsest coefficients matrix with more no of zeros. Curvelet transform is directional in nature and it preserves the important edge and texture details of MR images. In order to g...
متن کامل