|
|
Fingerprint Recognition with Graph Neural Networks
Pospíšil, Ondřej ; Špaňhel, Jakub (referee) ; Hradiš, Michal (advisor)
This thesis deals with the verification of fingerprints based on their graph representation. The proposed method uses a graph neural network and a combinatorial solver to obtain the matching between the minutae points of a pair of fingerprints. The matched minutae points are used to align the fingerprints using an estimated transformation by the RANSAC algorithm. The aligned fingerprints are processed by the SimGNN model. The resulting similarity score is then combined with the metrics obtained from the aligned fingerprints. The experiments summarize the selection of method parameters and the evaluation of fingerprint matching and verification accuracy. The contribution of this work is a new stable method of fingerprint alignment by solving the graph matching problem. The proposed verification method does not achieve high accuracy due to too few minutae attributes and poor discriminating power of the metrics used.
|
|
|
Neural Networks for Automatic Table Recognition
Piwowarski, Lukáš ; Španěl, Michal (referee) ; Hradiš, Michal (advisor)
Tato práce seznamuje čtenáře se současnými technikami rozpoznávání tabulek, které se používají především k získávání informací z ručně psaných nebo tištěných historických tabulek. Představujeme také metodu založenou na grafové neuronové síti, která je inspirována představenými přístupy. Metoda se skládá ze tří fází: fáze inicializace grafu, fáze klasifikace uzlů/hran a fáze transformace grafu na text. Ve fázi inicializace grafu používáme algoritmus viditelnosti uzlů a OCR k vytvoření počáteční grafové reprezentace vstupní tabulky. Ve fázi klasifikace uzlů a hran jsou uzly a hrany klasifikovány a ve fázi transformace grafu na text zarovnáváme uzly grafu do mřížky, která je pak použita k vytvoření konečné textové reprezentace tabulky. Náš implementovaný model byl schopen dosáhnout přesnosti 68 % u detekce horizontálních sousedů, přesnosti 71 % u detekce vertikálních sousedů a přesnosti 83 % u detekce buněk na datové sadě ABP.
|
|
|
Neural Networks for Automatic Equation Recognition
Halva, Vladislav ; Zemčík, Pavel (referee) ; Hradiš, Michal (advisor)
This thesis deals with automatic mathematical expressions recognition using deep neural networks. It contains an overview of existing approaches and focuses mainly on handwritten mathematical expressions and the use of graph neural networks. The core of the proposed system for handwritten mathematical expressions recognition is an encoder-decoder neural network model using graph neural networks to exploit the hierarchical structure of mathematical expressions. The designed system is evaluated on the CROHME dataset, which was published within the competition of the same name on mathematical expression recognition. The work also includes description of experiments performed with the designed model. The proposed solution achieves an exact expression recognition rate of 13.34% on the CROHME 2019 test dataset. The contribution of this work is mainly a method of using graph neural networks for mathematical expression recognition from images and their processing in the graph domain.
|
|
|
Machine Learning Methods for Web Documents
Katrňák, Josef ; Bartík, Vladimír (referee) ; Burget, Radek (advisor)
This work aims to use machine learning techniques for the classification of specific parts of web page content. First, current methods for representing and classifying web page content using machine learning methods are described. For web page representation, the thesis focuses on the experimental tool FitLayout, whose visual representation of web pages serves as input for further processing and subsequent training of machine learning models. The work results in trained models that classify specific parts of the web page content. The model architecture is based on graph neural networks. For the experiments, a dataset of publicly available websites containing pages of products sold online is used. The advantage of the proposed and implemented approach is information extraction independent of the structure and language of a web page.
|
|
|
Fingerprint Recognition with Graph Neural Networks
Pospíšil, Ondřej ; Špaňhel, Jakub (referee) ; Hradiš, Michal (advisor)
This thesis deals with the verification of fingerprints based on their graph representation. The proposed method uses a graph neural network and a combinatorial solver to obtain the matching between the minutae points of a pair of fingerprints. The matched minutae points are used to align the fingerprints using an estimated transformation by the RANSAC algorithm. The aligned fingerprints are processed by the SimGNN model. The resulting similarity score is then combined with the metrics obtained from the aligned fingerprints. The experiments summarize the selection of method parameters and the evaluation of fingerprint matching and verification accuracy. The contribution of this work is a new stable method of fingerprint alignment by solving the graph matching problem. The proposed verification method does not achieve high accuracy due to too few minutae attributes and poor discriminating power of the metrics used.
|
|
|
Neural Networks for Automatic Table Recognition
Piwowarski, Lukáš ; Španěl, Michal (referee) ; Hradiš, Michal (advisor)
Tato práce seznamuje čtenáře se současnými technikami rozpoznávání tabulek, které se používají především k získávání informací z ručně psaných nebo tištěných historických tabulek. Představujeme také metodu založenou na grafové neuronové síti, která je inspirována představenými přístupy. Metoda se skládá ze tří fází: fáze inicializace grafu, fáze klasifikace uzlů/hran a fáze transformace grafu na text. Ve fázi inicializace grafu používáme algoritmus viditelnosti uzlů a OCR k vytvoření počáteční grafové reprezentace vstupní tabulky. Ve fázi klasifikace uzlů a hran jsou uzly a hrany klasifikovány a ve fázi transformace grafu na text zarovnáváme uzly grafu do mřížky, která je pak použita k vytvoření konečné textové reprezentace tabulky. Náš implementovaný model byl schopen dosáhnout přesnosti 68 % u detekce horizontálních sousedů, přesnosti 71 % u detekce vertikálních sousedů a přesnosti 83 % u detekce buněk na datové sadě ABP.
|
|
|
Neural Networks for Automatic Equation Recognition
Halva, Vladislav ; Zemčík, Pavel (referee) ; Hradiš, Michal (advisor)
This thesis deals with automatic mathematical expressions recognition using deep neural networks. It contains an overview of existing approaches and focuses mainly on handwritten mathematical expressions and the use of graph neural networks. The core of the proposed system for handwritten mathematical expressions recognition is an encoder-decoder neural network model using graph neural networks to exploit the hierarchical structure of mathematical expressions. The designed system is evaluated on the CROHME dataset, which was published within the competition of the same name on mathematical expression recognition. The work also includes description of experiments performed with the designed model. The proposed solution achieves an exact expression recognition rate of 13.34% on the CROHME 2019 test dataset. The contribution of this work is mainly a method of using graph neural networks for mathematical expression recognition from images and their processing in the graph domain.
|
guest ::
