|
|
Using structural method for objects recognition
Valsa, Vít ; Heriban, Pavel (referee) ; Šťastný, Jiří (advisor)
This diploma thesis deals with posibilities of using structural methods for recognition objects in a picture. The first part of this thesis describes methods for preparing the picture before processing. The core of the whole thesis is in chapter 3, where is analyzed in details the problem of the formation of deformation grammars for parsing and their using. In the next part is space for syntactic parser describing the deformation grammar. The conclusion is focused on testing the suggested methods and their results.
|
|
|
The effect of the background and dataset size on training of neural networks for image classification
Mikulec, Vojtěch ; Kolařík, Martin (referee) ; Rajnoha, Martin (advisor)
This bachelor thesis deals with the impact of background and database size on training of neural networks for image classification. The work describes techniques of image processing using convolutional neural networks and the influence of background (noise) and database size on training. The work proposes methods which can be used to achieve faster and more accurate training process of convolutional neural networks. A binary classification of Labeled Faces in the Wild dataset is selected where the background is modified with color change or cropping for each experiment. The size of dataset is crucial for training convolutional neural networks, there are experiments with the size of training set in this work, which simulate a real problem with the lack of data when training convolutional neural networks for image classification.
|
|
|
Vehicle Make and Model Recognition in Image
Buchta, Martin ; Špaňhel, Jakub (referee) ; Herout, Adam (advisor)
This thesis deals with classification of a car model from an image. It describes several methods, such as convolutional neural networks, methods limited to the fron/rear view and methods using 3D CAD models. From these approaches it chooses convolutional neural networks, which it further deals with. The work contains a description of the individual layers of which such a network consists. The practical part describes the procedure by which the classifier, that has an accuracy of 80.7\,\%, was created. A dataset containing 1\,034 photos was created to verify functionality. The work further experiments with different architectures and evaluates their accuracy. The work contains a program which, thanks to the car detector, finds the vehicle in the video and marks it with a square and a description of the car model in the given video.
|
|
|
Methodology for the solution of massive tasks in GIS
Opatřilová, Irena ; Hanzl, Vlastimil (referee) ; Cajthaml,, Jiří (referee) ; Řezník,, Tomáš (referee) ; Bartoněk, Dalibor (advisor)
This doctoral thesis deals with the issue of solving massive tasks in GIS. These tasks process large volumes of geographic data with different formats. The thesis describes a theoretical analysis of the complexity of tasks and the possibilities to optimize sub-processes which lead to an acceptable solution. It considers the possibility of using parallelism in GIS, which leads to an acceleration in the processing of large volumes of geographic data. It also proposes a method for the optimization of processes through an algorithm which determines the number of means necessary for the successful solution of a task at a specified time and assigns processes to these means. Additionally, there is a proposed algorithm for the optimization of the preparation of data for extensive GIS projects. The algorithms have been validated by the results of a research project, the aim of which was to analyse the terrain surface above a gas line in the Czech Republic. The primary method of analysis was the classification of an orthophoto image, which was further refined through filtration using the ZABAGED layers. Therefore, the thesis deals with the possibility of improving the results of image classification using GIS instruments as well as dealing with the determination of the error rate in analysis results. The results of the analysis are now used for the strategic planning of maintenance and the development of gas facilities in the Czech Republic. The results of the work have general importance regarding the performance of other operations of the same class in GIS.
|
|
|
Image based flower recognition
Jedlička, František ; Kříž, Petr (referee) ; Přinosil, Jiří (advisor)
This paper is focus on flowers recognition in an image and class classification. Theoretical part is focus on problematics of deep convolutional neural networks. The practical part if focuse on created flowers database, with which it is further worked on. The database conteins it total 13000 plant pictures of 26 spicies as cornflower, violet, gerbera, cha- momile, cornflower, liverwort, hawkweed, clover, carnation, lily of the valley, marguerite daisy, pansy, poppy, marigold, daffodil, dandelion, teasel, forget-me-not, rose, anemone, daisy, sunflower, snowdrop, ragwort, tulip and celandine. Next is in the paper described used neural network model Inception v3 for class classification. The resulting accuracy has been achieved 92%.
|
|
|
Image classification using deep learning
Hřebíček, Zdeněk ; Přinosil, Jiří (referee) ; Mašek, Jan (advisor)
This thesis deals with image object detection and its classification into classes. Classification is provided by models of framework for deep learning BVLC/Caffe. Object detection is provided by AlpacaDB/selectivesearch and belltailjp/selective_search_py algorithms. One of results of this thesis is modification and usage of deep convolutional neural network AlexNet in BVLC/Caffe framework. This model was trained with precision 51,75% for classification into 1 000 classes. Then it was modified and trained for classification into 20 classes with precision 75.50%. Contribution of this thesis is implementation of graphical interface for object detction and their classification into classes, which is implemented as aplication based on web server in Python language. Aplication integrates object detection algorithms mentioned abowe with classification with help of BVLC/Caffe. Resulting aplication can be used for both object detection (and classification) and for fast verification of any classification model of BVLC/Caffe. This aplication was published on server GitHub under license Apache 2.0 so it can be further implemented and used.
|
|
|
Image Classification Using Genetic Programming
Jašíčková, Karolína ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
This thesis deals with image classification based on genetic programming and coevolution. Genetic programming algorithms make generating executable structures possible, which allows us to design solutions in form of programs. Using coevolution with the fitness prediction lowers the amount of time consumed by fitness evaluation and, therefore, also the execution time. The thesis describes a theoretical background of evolutionary algorithms and, in particular, cartesian genetic programming. We also describe coevolutionary algorithms properties and especially the proposed method for the image classifier evolution using coevolution of fitness predictors, where the objective is to find a good compromise between the classification accuracy, design time and classifier complexity. A part of the thesis is implementation of the proposed method, conducting the experiments and comparison of obtained results with other methods.
|
|
|
Vision Transformery pre vstavané platformy
Nemčeková, Barbora
This work focuses on investigation of Vision Transformers for the task of image classification, their optimization and deployment on selected embedded devices. Until now, convolutional neural networks have been used for image classification on the selected embedded devices, but with the revolution in natural language processing, there has been an interest in investigating transformers for computer vision tasks as well. The work experiments with different kinds of model quantization methods, such as int8 quantization, int16x8 quantization, dynamic quantization, and SmoothQuant method. The results show that not all transformers for computer vision can be quantized with sufficient accuracy, even when using the specialized SmoothQuant method. It also turned out that the quantized transformer model cannot be accelerated on the NPUs of selected devices. From the investigated factors, such as accuracy after model optimization, latency and memory usage on the embedded device, it emerged that for the task of image classification and model deployment on embedded devices, convolutional neural networks still outperform transformer models.
|
|
|
MLOSINT: Classifying Vehicle Losses in Ukraine
Kanát, Antonín ; Špelda, Petr (advisor) ; Střítecký, Vít (referee)
This thesis explores the potential of applying machine learning (ML) to assist with open source intelligence (OSINT) analysis. As the shared input of both disciplines, data is the primary lens through which the topic is examined. To understand the entire process of deploying an ML model from data collection to analysis, an image classifier of Russian vehicle losses in the invasion of Ukraine was trained and tested. Trained on a dataset of over 50,000 labelled images from the WarSpotting database, the classifier achieved a decent accuracy of 79% on evaluation data on the five most populous categories of images. On testing data from a later period, the performance dropped to 62%. One explanation offered is that the static frontlines and the prominence of drones led to most of the recent imagery being aerial, while the training data was captured mainly from the ground. That result demonstrated how inevitable changes, even in seemingly well-curated data, can lead to the low performance of ML models in deployment. Beyond changes on the battlefield, deeper data issues came to light, including the cascading effects of early data management decisions and dataset imbalance. Overall, current image classification methods do not work well on the noisy data available.
|
|
|
Supporting Board Game Nemesis on Android Mobile Phone
Štěpánek, Miroslav ; Švec, Tomáš (referee) ; Smrž, Pavel (advisor)
The aim of this thesis is to create a mobile application for the board game Nemesis designed for the Android system, which will allow the user to find out information about the game components during the game. The solution consists of two main parts the first is a model created with the help of the Tensorflow library, which is responsible for the detection of these components. The second is the application itself, which receives results from the model and displays the resulting information to the user. This makes the game easier for the user and helps to speed it up. The resulting system can be modified so that the application can be used for other games.
|
guest ::
