Saufiah Abdul Rahim
Thumbnail Image
Preferred name
Saufiah Abdul Rahim
Official Name
Saufiah, Abdul Rahim
Alternative Name
Rahim, Saufiah Binti Abdul
Rahim, S. Abdul
Rahim, Saufiah Abd
Rahim, Saufiah A.
Rahim, S. A.
Main Affiliation
Scopus Author ID
57210438798

Research Output

Filters

Reset filters

Settings
Now showing 1 - 2 of 2
  • Publication
    Performance analysis on denoising filters with new edge-directed interpolation for fingerprint images
    (Springer, 2025)
    Haniza Yazid
    ;
    Saufiah Abdul Rahim
    In image processing such as image interpolation, denoising filters are one of the most fundamental requirements. It is important in order to preserve the image quality and its edges’ properties. In this paper, we analyse the performance of seven different image denoising filters on fingerprint images. The fingerprint images were filtered and interpolated using New Edge-Directed Interpolation (NEDI) method. The image quality assessment (IQA) metrics that were used to assess the denoising filters quality are Edge-Based Image Quality Assessment (EBIQA), Non-Shift Edge Based Ratio (NSER), and Gradient Conduction Mean Square Error (GCMSE). The best image denoising filter is the Conservative Smoothing filter while the worst is the Laplacian filter.
  • Publication
    Performance Analysis on Interpolation-based Methods for Fingerprint Images
    (Institute of Electrical and Electronics Engineers Inc., 2022-01-01)
    Azam N.Z.F.N.
    ;
    Haniza Yazid
    ;
    Saufiah Abdul Rahim
    Image interpolation methods are a good approach to enhance or super resolve a digital image. But, there are surprisingly lack of use of these methods in biometric images especially, fingerprint images. In this research, we analysed is the most commonly known interpolation methods namely; nearest neighbour, bilinear, bicubic, and Lanczos interpolations to super resolve fingerprint images and compare the performance of each interpolation methods. Based on the results, the best interpolation is the Lanczos interpolation, followed by bicubic, bilinear and nearest neighbour interpolations.
      4  5