|
|
Speed of learning multilayer network
Maceček, Aleš ; Zámečník, Dušan (referee) ; Jirsík, Václav (advisor)
Theoretical study about neural networks, especially their types of topologies and networks learning. Special attention is attended to multilayer neural network with learning backpropagation. Introduced learning algorithm backpropagation of simple networks in conjunction with descriptions of parameters affecting network learning also methods to exaluation quality of network learning. Definition moment invariants to rotation, translation and scaling. Optimalization parameters of neural networks to find the network which has the fastest learning and also the networks with the best value of recognition patterns of letters from testing set.
|
|
|
Artificial neural network RCE
Maceček, Aleš ; Klusáček, Jan (referee) ; Jirsík, Václav (advisor)
This paper is focused on an artificial neural network RCE, especially describing the topology, properties and learning algorithm of the network. This paper describes program uTeachRCE developed for learning the RCE network and program RCEin3D, which is created to visualize the RCE network in 3D space. The RCE network is compared with a multilayer neural network with a learning algorithm backpropagation in the practical application of recognition letters. For a descriptions of the letters were chosen moments invariant to rotation, translation and scaling image.
|
|
|
Advanced Moment-Based Methods for Image Analysis
Höschl, Cyril ; Flusser, Jan (advisor) ; Papakostas, George (referee) ; Jiřík, Radovan (referee)
The Thesis consists of an introduction and four papers that contribute to the research of image moments and moment invariants. The first two papers focus on rectangular decomposition algorithms that rapidly speed up the moment calculations. The other two papers present a design of new moment invariants. We present a comparative study of cutting edge methods for the decomposition of 2D binary images, including original implementations of all the methods. For 3D binary images, finding the optimal decomposition is an NP-complete problem, hence a polynomial-time heuristic needs to be developed. We propose a sub-optimal algorithm that outperforms other state of the art approximations. Additionally, we propose a new form of blur invariants that are derived by means of projection operators in a Fourier domain, which improves mainly the discrimination power of the features. Furthermore, we propose new moment-based features that are tolerant to additive Gaussian image noise and we show by extensive image retrieval experiments that the proposed features are robust and outperform other commonly used methods.
|
|
|
Advanced Moment-Based Methods for Image Analysis
Höschl, Cyril ; Flusser, Jan (advisor) ; Papakostas, George (referee) ; Jiřík, Radovan (referee)
The Thesis consists of an introduction and four papers that contribute to the research of image moments and moment invariants. The first two papers focus on rectangular decomposition algorithms that rapidly speed up the moment calculations. The other two papers present a design of new moment invariants. We present a comparative study of cutting edge methods for the decomposition of 2D binary images, including original implementations of all the methods. For 3D binary images, finding the optimal decomposition is an NP-complete problem, hence a polynomial-time heuristic needs to be developed. We propose a sub-optimal algorithm that outperforms other state of the art approximations. Additionally, we propose a new form of blur invariants that are derived by means of projection operators in a Fourier domain, which improves mainly the discrimination power of the features. Furthermore, we propose new moment-based features that are tolerant to additive Gaussian image noise and we show by extensive image retrieval experiments that the proposed features are robust and outperform other commonly used methods.
|
|
|
Speed of learning multilayer network
Maceček, Aleš ; Zámečník, Dušan (referee) ; Jirsík, Václav (advisor)
Theoretical study about neural networks, especially their types of topologies and networks learning. Special attention is attended to multilayer neural network with learning backpropagation. Introduced learning algorithm backpropagation of simple networks in conjunction with descriptions of parameters affecting network learning also methods to exaluation quality of network learning. Definition moment invariants to rotation, translation and scaling. Optimalization parameters of neural networks to find the network which has the fastest learning and also the networks with the best value of recognition patterns of letters from testing set.
|
|
|
Artificial neural network RCE
Maceček, Aleš ; Klusáček, Jan (referee) ; Jirsík, Václav (advisor)
This paper is focused on an artificial neural network RCE, especially describing the topology, properties and learning algorithm of the network. This paper describes program uTeachRCE developed for learning the RCE network and program RCEin3D, which is created to visualize the RCE network in 3D space. The RCE network is compared with a multilayer neural network with a learning algorithm backpropagation in the practical application of recognition letters. For a descriptions of the letters were chosen moments invariant to rotation, translation and scaling image.
|
guest ::
