|
|
Recognition of Handwritten Digits
Štrba, Miroslav ; Španěl, Michal (referee) ; Herout, Adam (advisor)
Recognition of handwritten digits is a problem, which could serve as model task for multiclass recognition of image patterns. This thesis studies different kinds of algoritms (Self-Organizing Maps, Randomized tree and AdaBoost) and methods for increasing accuracy using fusion (majority voting, averaging log likelihood ratio, linear logistic regression). Fusion methods were used for combine classifiers with indentical train parameters, with different training methods and with multiscale input.
|
|
|
Recognition of Handwritten Digits
Dobrovolný, Martin ; Mlích, Jozef (referee) ; Herout, Adam (advisor)
Recognition of handwritten digits is one of computer vision problematics that can not be solved with 100 % success these days. This document describes a method for handwritten digits recognizing based on shape features and randomized tree classifiers. These methods are known for their long time machine learning and quick characters recognizing. This method is due to use of relative angles among key locations and is nearly invariant to substantial affine and nonlinear deformations.
|
|
|
Recognition of Handwritten Digits
Štrba, Miroslav ; Španěl, Michal (referee) ; Herout, Adam (advisor)
Recognition of handwritten digits is a problem, which could serve as model task for multiclass recognition of image patterns. This thesis studies different kinds of algoritms (Self-Organizing Maps, Randomized tree and AdaBoost) and methods for increasing accuracy using fusion (majority voting, averaging log likelihood ratio, linear logistic regression). Fusion methods were used for combine classifiers with indentical train parameters, with different training methods and with multiscale input.
|
|
|
Recognition of Handwritten Digits
Dobrovolný, Martin ; Mlích, Jozef (referee) ; Herout, Adam (advisor)
Recognition of handwritten digits is one of computer vision problematics that can not be solved with 100 % success these days. This document describes a method for handwritten digits recognizing based on shape features and randomized tree classifiers. These methods are known for their long time machine learning and quick characters recognizing. This method is due to use of relative angles among key locations and is nearly invariant to substantial affine and nonlinear deformations.
|
guest ::
