Only the fusion of the results of the analysis algorithms based on global (histogram of intensity levels) and local image features (cross-sectional and topological characteristics) improves discriminative features applicable to complete the information about the properties of the HFUS images with layered structure and to develop a method assessing the thickness of the skin layers. The knowledge gathered from such layers checks can improve understanding of the nature of the human skin and provide a more objective conditions for HFUS diagnostic imaging with speeding up the diagnostic process for dermatologists. We proposed a new method for automatic segmentation on HFUS images using fusion of discriminative information based on nonlinear segmentation with a reasonable threshold setting, boundary selecting and linking, and false boundary point removing strategies for intensity distributions. |
ACCESS THE FULL ARTICLE
No SPIE Account? Create one
CITATIONS
Cited by 2 scholarly publications.
Skin
Image segmentation
Image fusion
Visualization
Ultrasonography
Image processing
Speckle