Browsing by Author "Gavrilova, Marina L"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    A multimethod approach to the differentiation of enthesis bone microstructure based on soft tissue type
    (Wiley, 2021-06) Whitebone, S Amber; Bari, A S M Hossain; Gavrilova, Marina L; Anderson, Jason S
    Whereas there is a wealth of research studying the nature of various soft tissues that attach to bone, comparatively little research focuses on the bone's microscopic properties in the area where these tissues attach. Using scanning electron microscopy to generate a dataset of 1600 images of soft tissue attachment sites, an image classification program with novel convolutional neural network architecture can categorize images of attachment areas by soft tissue type based on observed patterns in microstructure morphology. Using stained histological thin section and liquid crystal cross-polarized microscopy, it is determined that soft tissue type can be quantitatively determined from the microstructure. The primary diagnostic characters are the orientation of collagen fibers and heterogeneity of collagen density throughout the attachment area thickness. These determinations are made across broad taxonomic sampling and multiple skeletal elements.
  • Thumbnail Image
    ItemOpen Access
    Open-set Speaker Recognition with Bounded Laguerre Voronoi Clustering
    (2024-08-19) Ohi, Abu Quwsar; Gavrilova, Marina L; Sousa, Mario Costa; Bezdek, Karoly
    Speaker recognition is a challenging problem in behavioral biometrics. It has been rigorously investigated over the last decade. Although numerous supervised closed-set systems successfully harvest the power of deep neural networks, limited studies have been made on open-set speaker recognition. This thesis proposes a self-supervised open-set speaker recognition that leverages the geometric properties of speaker distribution for accurate and robust speaker identification. The proposed framework consists of a deep neural network incorporating a wider viewpoint of temporal speech features and Laguerre–Voronoi diagram-based speech feature extraction. The deep neural network is trained with a specialized clustering criterion that only requires positive pairs during training. The framework further incorporates a novel approach of clustering by integrating concepts from Voronoi diagrams in Laguerre geometry. This approach offers flexibility by necessitating only one hyperparameter, an upper-bound value for the number of centroids. The experiments validated that the proposed system outperformed current state-of-the-art methods in open-set speaker verification and identification.