|
|
Crowd Behavior Anomaly Detection in Drone Videodata
Bažout, David ; Herout, Adam (referee) ; Beran, Vítězslav (advisor)
There have been lots of new drone applications in recent years. Drones are also often used in the field of national security forces. The aim of this work is to design and implement a tool intended for crowd behavior analysis in drone videodata. This tool ensures identification of suspicious behavior of persons and facilitates its localization. The main benefits include the design of a suitable video stabilization algorithm to stabilize small jitters, as well as trace back of the lost scene. Furthermore, two anomaly detectors were proposed, differing in the method of feature vector extraction and background modeling. Compared to the state of the art approaches, they achieved comparable results, but at the same time they brought the possibility of online data processing.
|
|
|
Visualization of Crowd Behavior Analysis Using Drone Videodata
Dobeš, Zdeněk ; Španěl, Michal (referee) ; Bažout, David (advisor)
The purpose of this work is to design and implement an user interface for visualizing videodata from a drone. The document consists of a description of the construction of a design until its implementation in the form of a web application. It uses the Django application framework, the React web framework and variety of APIs with Google Map API at the top. It aims to create an intuitive environment for monitoring sectors on a map with drones and offers useful tools to make it easier.
|
|
|
Smart Home: Learning Control of Heating
Milostný, Tomáš ; Bažout, David (referee) ; Materna, Zdeněk (advisor)
This bachelor thesis deals with the implementation of a prototype smart home ecosystem focused on home heating. The goal is to create software running on a Raspberry Pi device that, with the help of machine learning predictions, will be able to control and monitor a network of electric direct heating convectors controlled by switching a simple Wi-Fi relay Shelly 1PM. The resulting solution should also have a lower price than conventional smart thermostats.
|
|
|
Detection of Wanted People in Video
Bažout, David ; Musil, Petr (referee) ; Beran, Vítězslav (advisor)
The aim of this work is to create a software tool for searching of wanted people in video recordings from surveillance cameras. Wanted people are identified to the system using multiple facial photos. The output consists of information on the occurrence of wanted persons in specific frames. The problem consists of face detection and its subsequent identification task. Experiments with existing approaches on appropriate datasets provide relevant comparisons of method performance under different conditions. Appropriate methods and their optimal settings for this particular task are chosen according to the results of the experiments. The thesis also deals with the design of suitable architecture, research of existing libraries implementing the tested methods and other ways of optimizing the calculation. The result is the implementation of a user application that meets the specified parameters. The application's functionality has been tested on the own dataset simulating real-world conditions.
|
|
|
Smart Home: Learning of Routines
Junák, Petr ; Bažout, David (referee) ; Materna, Zdeněk (advisor)
This thesis focuses on the design, development, and implementation of an innovative smart system that uses user sleep data and machine learning for intelligent control of lights in a household. The main goal is to improve the user’s sleep quality through the optimization of light exposure, particularly by reducing exposure to cool daylight before sleep. The work is characterized by an emphasis on the use of readily available data, cost-effective hardware devices, and open-source technologies. The primary source of data was the FitBit Charge 5 smartwatch, which provides detailed information about the user’s sleep patterns. The created machine learning model, specifically linear regression, is hosted on the Raspberry Pi platform, which also serves as the central unit for controlling lights and managing the overall system. In addition to light control, the system also supports sending notifications to the user about the approaching recommended sleep start time contributing to the overall improvement of sleep hygiene and sleep quality.
|
|
| |
|
|
Application for Plant Disease Detection
Kozub, Tadeáš ; Vaško, Marek (referee) ; Bažout, David (advisor)
This thesis deals with the design and creation of a mobile application that serves to help with plant disease identification. It presents the process of designing the app and its UI, user testing, creation of REST API and also the training of models using machine learning methods. The result is a React Native application, trained model for image-based recognition of tomato diseases and a set up remote server with backend implementation. The achieved accuracy of tomato disease model is 98.76%.
|
|
|
Web Application for Managing a Smart Greenhouse with Gamification Elements
Janeček, Martin ; Materna, Zdeněk (referee) ; Bažout, David (advisor)
The aim of this work is to create an intuitive user interface for monitoring and controlling conditions in a smart greenhouse. This is complemented by a system of devices, compared to a normal greenhouse, which maintain an ideal environment for the plants using remote access from the application. The web application is equipped by a gamification system, which aims to motivate the user to use all the features. At the same time, this system guides the user through the initial introduction to the smart greenhouse. The application is implemented using the Django framework, the presentation side is provided by the React library. The developed solution allows users to easily control the conditions in the smart greenhouse and analyze the history of measured data in a chart or statistics.
|
|
|
UI for the Cooperation of a Surveillance System with a Human
Klem, Václav ; Špaňhel, Jakub (referee) ; Bažout, David (advisor)
This bachelor's thesis discusses the topic of the design and development of an effective user interface for the cooperation of a user and a surveillance system. It offers a study of currently available tools for the annotation of videos and detection/tracking of people in these videos to create an annotation application intended for new way of detecting anomalies. This innovative way is based on detecting anomalies on the level of frames. Anomaly in this sense means any illegal behavior of people in a video. This thesis aims to create a proposition of UI design for the annotation application, confirm the validity of the concept and then implement the full-fledged version of this application based on the testing results. During the testing of the prototype, the annotation success rate of 89% and the rating of overall UI clarity of 83% respectively were reached. Following the results of the testing, the final version of the application was implemented reflecting the user feedback.
|
|
|
Conversion of a Robot Program from Code to an AR-Compatible Representation
Kadnár, Šimon ; Bažout, David (referee) ; Materna, Zdeněk (advisor)
The demand for robots and their development is constantly growing. Along with robots, various environments are being created to facilitate their programming. One such environment is the ARCOR2 application. The application has implemented a functionality that allows internally represented data, which control the robot, to be converted into the widely known Python language. The aforementioned conversion was introduced because Python is used as an interpreter for the robot program in which it is possible to run the program. The aim of the work is to utilize the conversion from the internal representation of data to the Python language for modifying the robot's code. The benefit of the work is enabling modifications in the Python language by converting the modified code back into the internal representation of data. As a result, the workstation operator can create a program using augmented reality, and an experienced programmer can subsequently modify the created program in the Python language. The solution allows for effective collaboration between regular users and programmers, with each working with a program form that corresponds to their abilities.
|
guest ::
