|
|
The design study of nature-friendly erosion control measures in the basin
Mrázek, Vojtěch ; Sobotková, Veronika (referee) ; Dumbrovský, Miroslav (advisor)
The work is focused to design of complex erosionand flood control measures close to nature in the Bravantice Regional Office. Erosion and runoff conditions were analyzed in the given area. The DesQ-MaxQ program was used to calculate the runoff ratios in the critical point profiles. The analysis of erosion conditions was performed using GIS tools together with the USLE2D program. ArcGis was used to create graphical outputs. Based on the performed analyzes, protective measures improving erosion and runoff conditions were proposed. The evaluation of their efficiency was performed by comparing the values of the basic characteristics of direct runoff and soil erosion before and after the design of measures.
|
|
|
Evolutionary Optimization of Convolutional Neural Networks
Roreček, Pavel ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
This Master's Thesis is focused on the principles of neural networks, primarily convolutional neural networks (CNN). It introduces the evolutionary optimization in the context of neural networks. One of existing libraries devoted to the CNN design was chosen (Keras), analysed and used in image classification tasks. An optimization technique based on cartesian genetic programming that should reduce the complexity of CNN's computation was proposed and implemented. The impact of the proposed technique on CNN behaviour was evaluated in a case study.
|
|
|
Estimation of the Probability of Occurrence of Persons in an Area
Koprda, Peter ; Vašíček, Zdeněk (referee) ; Mrázek, Vojtěch (advisor)
The main goal of this bachelor thesis is to implement application that creates estimation of the probability of occurrence of persons in an area from given map data. The estimation of the probability is displayed by using a heat map. Heat map is created by using defined tags which determine how many people are in the area. During the implementation map data from OpenStreetMap were used. The created application can be used during the planning of the electricity equipment construction, which can be life-threatening if people are close to the area of this equipment.
|
|
|
Cellular Modem Emulator for Acceleration of Embedded System Development
Horký, Vít ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
The purpose of this work is to create an emulator intended for testing and development of libraries for communication with cellular modems with support for AT protocol. This emulator is implemented as a python library and supports basic AT commands and GSM multiplexing. Support for testing other devices and libraries is achieved by user specified testing scenarios. By default, support for emulating modem Bg96 by Quectel is implemented but an emphasis is placed on the ability to support other modems.
|
|
|
Autopilot of RC Plane
Denk, Filip ; Bidlo, Michal (referee) ; Mrázek, Vojtěch (advisor)
This thesis is focused on stabilization of the radio-controlled aircraft model. The aim is to create a system that could controll the direction of flight (without pilot intervention) itself. This system also includes determination of the aircraft’s position using a 3-axis gyroscope. The Arduino Uno microcontroller was used as a controll unit.
|
|
|
Exploiting Approximate Arithmetic Circuits in Neural Networks Inference
Matula, Tomáš ; Mrázek, Vojtěch (referee) ; Češka, Milan (advisor)
Táto práca sa zaoberá využitím aproximovaných obvodov v neurónových sieťach so zámerom prínosu energetických úspor. K tejto téme už existujú štúdie, avšak väčšina z nich bola príliš špecifická k aplikácii alebo bola demonštrovaná v malom rozsahu. Pre dodatočné preskúmanie možností sme preto skrz netriviálne modifikácie open-source frameworku TensorFlow vytvorili platformu umožňujúcu simulovať používanie approximovaných obvodov na populárnych a robustných neurónových sieťach ako Inception alebo MobileNet. Bodom záujmu bolo nahradenie väčšiny výpočtovo náročných častí konvolučných neurónových sietí, ktorými sú konkrétne operácie násobenia v konvolučnách vrstvách. Experimentálne sme ukázali a porovnávali rozličné varianty a aj napriek tomu, že sme postupovali bez preučenia siete sa nám podarilo získať zaujímavé výsledky. Napríklad pri architektúre Inception v4 sme získali takmer 8% úspor, pričom nedošlo k žiadnemu poklesu presnosti. Táto úspora vie rozhodne nájsť uplatnenie v mobilných zariadeniach alebo pri veľkých neurónových sieťach s enormnými výpočtovými nárokmi.
|
|
|
Visualization of CPU Cache System
Peřina, Daniel ; Sekanina, Lukáš (referee) ; Mrázek, Vojtěch (advisor)
The goal of this work is to design and implement CPU cache simulator. In today's computers there is a difference of orders of magnitude between performance of CPUs and the main memory and thus it is necessary to use caches as an interlayer. The simulator will demonstrate effect of caches on different algorithms and problems that can occur if they are used inappropriately. For ease of use the simulator is implemented as a web application using framework Vue.js. User can enter assembly code and then execute it on several different types of caches. The application visualizes data flow between main memory and cache. Several sample programs were also created, which demonstrate various properties and problems of caches. With this application it is possible to clearly show significance of cache memory.
|
|
|
Wireless Communication in Fire Sport
Pelka, Tomáš ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
This work describes the design of wireless timer for fire sport disciplines. Its aim is to replace long wires with wireless communication while preserving equal or better reliability. Important part is the need for clock synchronization between wireless modules. Related topic is selection of suitable synchronization algorithm and communication protocol. The proposed system was realized in form of prototype using Texas Instruments's development platform LaunchPad EXP 430 FR4133 and wireless modules with Texas Instruments's chip CC1101. Parameters of designed system (especially the accuracy of time synchronization and measurement) were verified in laboratory.
|
|
|
Compression of Device Location Data
Morong, Lukáš ; Mrázek, Vojtěch (referee) ; Vašíček, Zdeněk (advisor)
The goal of this work is to analyze the properties of the data used to locate IoT devices using AoA/AoD based on BLE (Bluetooth low energy) technology. The work aims to identify suitable algorithms for lossless compression, while taking into account the computing and memory resources of the target microcontroller architecture. This involves methods such as VLQ, Rice-Golomb coding or predictive coding. Then implement the selected algorithms in the C language for a platform with an ARM processor and evaluate their parameters in terms of both computational and memory complexity as well as the level of compression.
|
|
|
Evolutionary Optimization of Convolutional Neural Networks
Čoupek, Vojtěch ; Mrázek, Vojtěch (referee) ; Sekanina, Lukáš (advisor)
This thesis deals with the problem of neural network weights compression using the technique of Weight-Sharing and parameter optimization of this technique by unconventional optimization algorithms. The reason for the optimization is decreasing the memory or energy demands of the neural network response calculation. The aim is to design a system that accepts a neural network and reduces its memory demands. Its functionality is demonstrated with the help of several experiments. The thesis investigates the use of various optimization algorithms, additional compression using the quantization above the Weight-Sharing technique, and proposes the quantization results tuning method to improve accuracy. These procedures are first tested on the Le-Net-5 network and then applied for the MobileNet\_v2. network compression.
|
guest ::
RSS feed.