Sökning: "Binarization"

Visar resultat 1 - 5 av 17 uppsatser innehållade ordet Binarization.

  1. 1. Gaussian Process Multiclass Classification : Evaluation of Binarization Techniques and Likelihood Functions

    Kandidat-uppsats, Linnéuniversitetet/Institutionen för matematik (MA)

    Författare :Benjamin Ringdahl; [2019]
    Nyckelord :Machine learning; Gaussian process classification; Laplace s approximation; Likelihood functions; Binarization techniques; Class imbalance;

    Sammanfattning : In binary Gaussian process classification the prior class membership probabilities are obtained by transforming a Gaussian process to the unit interval, typically either with the logistic likelihood function or the cumulative Gaussian likelihood function. Multiclass classification problems can be handled by any binary classifier by means of so-called binarization techniques, which reduces the multiclass problem into a number of binary problems. LÄS MER

  2. 2. Cluster-based Sample Selection for Document Image Binarization

    Master-uppsats, Blekinge Tekniska Högskola/Institutionen för datavetenskap

    Författare :Amandus Krantz; [2019]
    Nyckelord :document image binarization; sample selection; neural networks; computer vision;

    Sammanfattning : The current state-of-the-art, in terms of performance, for solving document image binarization is training artificial neural networks on pre-labelled ground truth data. As such, it faces the same issues as other, more conventional, classification problems; requiring a large amount of training data. LÄS MER

  3. 3. Semantic Segmentation of Historical Document Images Using Recurrent Neural Networks

    Master-uppsats, Blekinge Tekniska Högskola/Institutionen för programvaruteknik; Blekinge Tekniska Högskola/Institutionen för programvaruteknik

    Författare :Jakob Ahrneteg; Dean Kulenovic; [2019]
    Nyckelord :semantic segmentation; page segmentation; recurrent neural network; layout analysis; semantisk segmentering; dokument segmentering; recurrent neural network; layout analys;

    Sammanfattning : Background. This thesis focuses on the task of historical document semantic segmentation with recurrent neural networks. Document semantic segmentation involves the segmentation of a page into different meaningful regions and is an important prerequisite step of automated document analysis and digitisation with optical character recognition. LÄS MER

  4. 4. Designing Variational Autoencoders for Image Retrieval

    Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Sara Torres Fernandez; [2018]
    Nyckelord :;

    Sammanfattning : The explosive growth of acquired visual data on the Internet has raised interestin developing advanced image retrieval systems. The main problem relies on thesearch of a specic image among large collections or databases, and this issue isshared by lots of users from a variety of domains, like crime prevention, medicineor journalism. LÄS MER

  5. 5. Development of Registration and Fusion Methods for the Jonasson Medical Imaging Center MiniPET-microCT

    Master-uppsats, KTH/Skolan för kemi, bioteknologi och hälsa (CBH)

    Författare :Dimitrios Gkotsoulias; [2018]
    Nyckelord :miniPET; microCT; Medical imaging; Multimodality imaging; Registration; Fusion; Elastix; Mutual Information;

    Sammanfattning : Multimodal image registration is essential when combining functional and structural imaging modalities. Among the most common combinations, numerous methods have been developed for co-registration of CT and PET, typically validated on human size scanners. LÄS MER