|
|
Recognition of Handwritten Digits
Hekrdla, Michal ; Hradiš, Michal (referee) ; Herout, Adam (advisor)
This bachelors thesis inspects an issue of recognition of handwritten digits with decision trees method. It describes principle method, usage database NITS (National Institute of Standards and Technology) for purposes teaching algorithm, construction tags tree and decision tree. It describes too implementation those method on demonstrational program, which is its programme part. Finally it deal with testing recognition program and its estimation.
|
|
|
Machine Learning in Image Classification
Král, Jiří ; Španěl, Michal (referee) ; Hradiš, Michal (advisor)
This project deals vith analysis and testing of algorithms and statistical models, that could potentionaly improve resuts of FIT BUT in ImageNet Large Scale Visual Recognition Challenge and TRECVID. Multinomial model was tested. Phonotactic Intersession Variation Compensation (PIVCO) model was used for reducing random e ffects in image representation and for dimensionality reduction. PIVCO - dimensionality reduction achieved the best mean average precision while reducing to one-twenyth of original dimension. KPCA model was tested to approximate Kernel SVM. All statistical models were tested on Pascal VOC 2007 dataset.
|
|
|
Texture-Based Object Recognition
Wozniak, Jan ; Hradiš, Michal (referee) ; Španěl, Michal (advisor)
This thesis is focused on analysis of texture-based features and classi cation of known objects. The technical report provides basic outline of commonly used texture features and principles of their classifi cation, whereas narrower attention is dedicated to extraction of Local Binary Patterns and Support Vector Machine algorithm based classi er. This work also includes evaluation of attained results by statistical methods Jackkni ng and F-measure.
|
|
|
Face Detection
Štrba, Miroslav ; Juránek, Roman (referee) ; Hradiš, Michal (advisor)
This bachelor thesis contains overview of actual face detection methods using classifier. It also contains description of creating system for face detection. There are described different methods for classifier training in first part. There is analysis, which preceded creation of system focused on black-and-white picture, in second part. Implemented system is using WaldBoost algorithm and Haar features. There is option to use particle filter in video.
|
|
|
Multi Object Class Learning and Detection in Image
Chrápek, David ; Hradiš, Michal (referee) ; Beran, Vítězslav (advisor)
This paper is focused on object learning and recognizing in the image and in the image stream. More specifically on learning and recognizing humans or theirs parts in case they are partly occluded, with possible usage on robotic platforms. This task is based on features called Histogram of Oriented Gradients (HOG) which can work quite well with different poses the human can be in. The human is split into several parts and those parts are detected individually. Then a system of voting is introduced in which detected parts votes for the final positions of found people. For training the detector a linear SVM is used. Then the Kalman filter is used for stabilization of the detector in case of detecting from image stream.
|
|
|
Road Surface Detection
Melichar, Jiří ; Motlíček, Petr (referee) ; Hradiš, Michal (advisor)
This bachelor`s thesis deals with a method for road surface detection in picture and is inspired by work of S. Thrun and H. Dahlkamp. The method works with color models of road, which are adjusted to changing environment. Then these models are used to classify the picture. Output of this is the detected road. The method is thoroughly analyzed, implemented and tested. Test results are discussed and proposals for improvements are presented.
|
|
|
Convolutional Networks for Handwriting Recognition
Sladký, Jan ; Kišš, Martin (referee) ; Hradiš, Michal (advisor)
This thesis deals with handwriting recognition using convolutional neural networks. From the current methods, a network model was chosen to consist of convolutional and recurrent neural networks with the Connectist Temporal Classification. The Vertical Attention Module, which selects the relevant information in each column corresponding to the text in the figure was subsequently implemented in such a model. Then, this module was compared with other possibilities of vertical aggregation between convolutional and recurrent networks. The experiments took place on a data set containing over 80,000 lines of text from Czech letters from the 20th century. The results show that the Vertical Attention Module almost always achieves the best results on all used types of convolution networks. The resulting network achieved the best result with 8,9% of the character error rate. The contribution of this work is a neural network with a newly introduced element that can recognize lines of text.
|
|
|
Deep Learning for Medical Image Analysis
Trávníčková, Kateřina ; Hradiš, Michal (referee) ; Španěl, Michal (advisor)
This bachelor thesis deals with medical volume data analysis using convolutional neural networks. The input of the analysis is a CT scan of human limbs and the output are segmented countours of long bones, humerus and tibia. The goal of this work is to find suitable convolutional neural network settings to achieve the best possible analysis output while the area under the Precision-Recall curve is used as the precision metric. The best accuracy reaches almost 88 % (0.8778 AUC). The implementation is based on Caffe framework, or python caffe module.
|
|
|
Holistic License Plate Recognition Based on Convolution Neural Networks
Le, Hoang Anh ; Hradiš, Michal (referee) ; Špaňhel, Jakub (advisor)
Main goal of this work was to create a holistic license plate reader, with an emphasis on achieving the highest possible accuracy on low quality images. Combination of convolutional and recurrent neural networks was designed and implemented, with usage of LSTM and CTC, where the inputs are cut-outs from the entire license plate. Competitive networks were also implemented to compare results. Networks were compared on a total of 4 datasets and the results were, that my design has achieved the best results with a recognition accuracy of 97.6%.
|
|
|
Image-Based Licence Plate Recognition
Vacek, Michal ; Hradiš, Michal (referee) ; Beran, Vítězslav (advisor)
In first part thesis contains known methods of license plate detection. Preprocessing-based methods, AdaBoost-based methods and extremal region detection methods are described.Finally, there is a described and implemented own access using local detectors to creating visual vocabulary, which is used to plate recognition. All measurements are summarized on the end.
|
guest ::
