|
|
Statistical Analysis of Anomalies in Sensor Data
Gregorová, Kateřina ; Čmiel, Vratislav (referee) ; Sekora, Jiří (advisor)
This thesis deals with the failure mode detection of aircraft engines. The main approach to the detection is searching for anomalies in the sensor data. In order to get a comprehensive idea of the system and the particular sensors, the description of the whole system, namely the aircraft engine HTF7000 as well as the description of the sensors, are dealt with at the beginning of the thesis. A proposal of the anomaly detection algorithm based on three different detection methods is discussed in the second chapter. The above-mentioned methods are SVM (Support Vector Machine), K-means a ARIMA (Autoregressive Integrated Moving Average). The implementation of the algorithm including graphical user interface proposal are elaborated on in the next part of the thesis. Finally, statistical analysis of the results,the comparison of efficiency particular models and the discussion of outputs of the proposed algorithm can be found at the end of the thesis.
|
|
|
System for Detection of APT Attacks
Hujňák, Ondřej ; Kačic, Matej (referee) ; Barabas, Maroš (advisor)
The thesis investigates APT attacks, which are professional targeted attacks that are characterised by long-term duration and use of advanced techniques. The thesis summarises current knowledge about APT attacks and suggests seven symptoms that can be used to check, whether an organization is under an APT attack. Thesis suggests a system for detection of APT attacks based on interaction of those symptoms. This system is elaborated further for detection of attacks in computer networks, where it uses user behaviour modelling for anomaly detection. The detector uses k-nearest neighbors (k-NN) method. The APT attack recognition ability in network environment is verified by implementing and testing this detector.
|
|
|
Detection of Anomalies in Pedestrian Walking
Pokorný, Ondřej ; Orság, Filip (referee) ; Goldmann, Tomáš (advisor)
The goal of this work was to create a system that would be able to detect anomalies in pedestrian walking. As the core of my application, I have used OpenPose, which is an application for detecting human skeletons. Then I used a bidirectional LSTM neural network to detect anomalies in video sequences. This architecture was chosen during the experiment because it outperformed other solutions. I trained my model to detect three types of anomalies. The output of my application is a video with marked sequences of anomalies. The whole system is implemented in Python.
|
|
|
Detection of Network Attacks Based on NetFlow Data
Kulička, Vojtěch ; Tobola, Jiří (referee) ; Žádník, Martin (advisor)
With rising popularity of the internet there is also rising number of people misusing it. This thesis analyzes the problem of network attack detection based on NetFlow data. A program is designed to point out anomalous behaviour by analyzing the flow records using data mining techniques. The method of TCM-KNN utilizing the fact that attacks statistically deviate is implemented. Thus even new types of attacks are detected
|
|
|
RQA System Anomaly Detection
Lorenc, Jan ; Jeřábek, Kamil (referee) ; Pluskal, Jan (advisor)
The aim of the theses is to design and implement a machine learning model for anomaly detection in Y Soft's RQA system. Owing to the microservice architecture, an anomaly is considered to be a recurring occurrence of outliers in durations of service requests or a considerable variance in error rate. The thesis outlines the current data collection process in the system and defines what kind of data describe the state of the system. It devises a suitable format of data storage for its subsequent analysis. It presents algorithms commonly used to solve anomaly detection problems. The anomaly detection is designed and implemented using cluster analysis and statistical methods. Finally, the thesis evaluates the quality of the detection and the achieved results.
|
|
|
Unsupervised Anomaly Detection in Image
Salvet, Lukáš ; Herout, Adam (referee) ; Juránek, Roman (advisor)
This thesis deals with anomaly detection on industrial products. The main requirement was that the method required as little data with anomalies as possible at the time of construction and that it was easily applicable to different types of products. Neural network that is indirectly taught to find differences between two pictures is designed and described in this thesis. The anomaly detection itself should take place based on the representation of input data in latent space or in combination with a reconstruction loss. Four different method modifications have been designed and tested. The testing was mainly carried out on the MVTec AD dataset, which contains industrial products. Unfortunately the assumption that if the network is taught to look for differences the latent space will be interpreted better was not confirmed. Therefore the method was evaluated in a reconstructive error mode in~which it achieves comparable results with other methods. The result is insufficient for use in practice.
|
|
|
Identifying Attacks in Home IoT Networks
Matuš, Tomáš ; Pluskal, Jan (referee) ; Ryšavý, Ondřej (advisor)
The aim of this bachelor thesis is to design and implement an experimental environment for a home IoT network based on the use of Home Assistant, testing various selected attacks on this system using penetration testing tools, designing a way to monitor this environment using logs from Home Assistant, analysis of collected monitoring data, design of detection method for the selected attacks and evaluation of the proposed attack detection method.
|
|
|
Anomaly detection in video sequences on devices with low computing power
Bílek, František ; Ježek, Štěpán (referee) ; Sikora, Pavel (advisor)
The bachelor’s thesis focuses on the problem of anomaly detection in video sequences on devices with low computational power. Traditional and current approaches to anomaly detection are described from the perspective of machine learning and neural networks. The goal of the thesis is to implement an efficient and reliable algorithm capable of detecting anomalies in real-time. Emphasis is placed on minimizing computational requirements and optimizing memory usage to achieve efficiency on devices with limited computational capacities.
|
|
|
Applications of Machine Learning in Predictive Maintenance of Industry 4.0
Navrátil, Tadeáš ; Richter, Miloslav (referee) ; Horák, Karel (advisor)
The thesis develops machine learning algorithms for use in the Industry 4.0 concept. The main focus is on predictive maintenance and visual inspection. In the theoretical part, the thesis focuses on a literature search of machine learning methods in the field of anomaly detection in time series and image data. The practical part deals with the reimplementation of the selected methods and their evaluation using the confusion matrix and metrics based on it
|
|
|
Visual Anomaly Detection in Industrial Production
Lukaszczyk, Jakub ; Petyovský, Petr (referee) ; Horák, Karel (advisor)
This work deals with the detection of anomalies in image data taken on an industrial product. The first part outlines the problem and approaches to its solution using deep learning. Then, some of the architectures that can be used for this task are discussed. The practical part then describes the platform for industrial inspection, the software used and the creation of the annotated dataset. The software is extended with features for controlling the platform and working with multiple cameras. The last section deals with experiments designed to investigate the effect of the dataset on the resulting model and the estimation of its performance. The experiments evaluate the influence in both training and testing phases.
|
guest ::
