Swamy, Gokul ; Balasubramanian, Krishnan (2007) Directional filter bank-based segmentation for improved evaluation of nondestructive evaluation images NDT & E International, 40 (3). pp. 250-257. ISSN 0963-8695
Full text not available from this repository.
Official URL: https://doi.org/10.1016/j.ndteint.2006.10.005
Related URL: http://dx.doi.org/10.1016/j.ndteint.2006.10.005
Abstract
This paper introduces a directional filter bank (DFB) for segmentation of NDE images containing directional information. The DFB is used to split an image into a desired number of sub-band images with each sub-band image containing features belonging only to a given angular range. The DFB is a two-channel decomposition employing the Quincunx sampling matrix and the diamond half band filter pair. The DFB is also designed to incorporate the property of perfect reconstruction or alias free reconstruction. Applications of the DFB towards segmenting C-scan images of fiber-reinforced composites, magnetic flux leakage (MFL) images of seamless tubes, IR images of solar cell panels and optical images for the computation of area coverage in a shot-peening process are discussed.
| Item Type: | Article |
|---|---|
| Source: | Copyright of this article belongs to Elsevier Science. |
| ID Code: | 141648 |
| Deposited On: | 22 Jan 2026 18:14 |
| Last Modified: | 22 Jan 2026 18:14 |
Repository Staff Only: item control page
