A Thresholding-Based Salt and Pepper Noise Removal Using B-Spline Interpolation in MRI Images
Figlu Mohanty, Suvendu Rup, Bodhisattva Dash · 2015
This paper deals with a cubic B-Spline interpolation technique to shed the impulse noise and preserve the edges in MRI images. MRI images play a vital role in diagnosis and treatment in many specialties such as neuro-imaging, cardiovascular, musculoskeletal, liver & gastrointestinal MRI etc. To remove the impulse noise, a thresholding based two-step process is adopted. The proposed method operates as follows: in the first step, the uncorrupted pixels are obtained by checking each pixel value with the threshold value. In second step, impaired edge but noise free image obtained in step-I is considered as the input. Here in this step, the corrupted pixels identified in step-I are interpolated by using cubic B-Spline edge preservation process and hence, the edge preserved and de-noised image is obtained. The robustness of the scheme is validated in noise situations of different strengths. In all situations, the results have been compared with standard schemes. It is in general observed that the proposed method outperforms in terms of restoration parameter and visual quality.