A B S T R A C T title: Multimodal Image Processing in Art Investigation author: Jan Blažek department: Department of Image Processing, IITA of the CAS supervisor: RNDr. Barbara Zitová PhD., Institute of Information Theory and Automation supervisor's e-mail address: zitova@utia.cas.cz abstract: Art investigation and digital image processing demar- cate an interdisciplinary field of the presented thesis. Over the past 8 years we have published thirteen papers belonging to this field of research. This thesis presents the current state of the art and puts these papers into context. Our research is focused on modalities in the visible and near-infrared parts of the spectrum and affects vari- ous tasks of art investigation. For studying the spectral response of paint materials, we suggest a low-cost mobile multi-band acquisition system and a calibration method extended by a light source with an adjustable wavelength. We created the m3art database of the spectral responses of pigments, available for comparison and public use. The central point of our research is underdrawing detection and visual- ization. For this purpose we have developed: acquisition guidelines based on optical properties of the topmost non-transparent layer, a visualization technique for comparison of modalities, and a signal separation technique...

This project will explore machine learning approaches that are used in creditscoring. In this study we consider consumer credit scoring instead of corporatecredit scoring and our focus is on methods that are currently used in practiceby banks such as logistic regression and decision trees and also compare theirperformance against machine learning approaches such as support vector machines (SVM), neural networks and random forests. In our models we addressimportant issues such as dataset imbalance, model overfitting and calibrationof model probabilities. The six machine learning methods we study are support vector machine, logistic regression, k-nearest neighbour, artificial neuralnetworks, decision trees and random forests. We implement these models inpython and analyse their performance on credit dataset with 30000 observations from Taiwan, extracted from the University of California Irvine (UCI)machine learning repository.

No description available.

Recent developments in the field of object detection have highlighted a significant variation in quality between visual datasets. As a result, there is a need for a standardized approach of validating visual dataset features and their performance contribution. With a focus on vehicle detection, this thesis aims to develop an evaluation method utilized for comparing visual datasets. This method was utilized to determine the dataset that contributed to the detection model with the greatest ability to detect vehicles. The visual datasets compared in this research were BDD100K, KITTI and Udacity, each one being trained on individual models. Applying the developed evaluation method, a strong indication of BDD100K's performance superiority was determined. Further analysis and feature extraction of dataset size, label distribution and average labels per image was conducted. In addition, real-world experimental conduction was performed in order to validate the developed evaluation method. It could be determined that all features and experimental results pointed to BDD100K's superiority over the other datasets, validating the developed evaluation method. Furthermore, the TensorFlow Object Detection API's ability to improve performance gain from a visual dataset was studied. Through the use of augmentations, it was concluded that the TensorFlow Object Detection API serves as a great tool to increase performance gain for visual datasets. / Inom fältet av objektdetektering har ny utveckling demonstrerat stor kvalitetsvariation mellan visuella dataset. Till följd av detta finns det ett behov av standardiserade valideringsmetoder för att jämföra visuella dataset och deras prestationsförmåga. Detta examensarbete har, med ett fokus på fordonsigenkänning, som syfte att utveckla en pålitlig valideringsmetod som kan användas för att jämföra visuella dataset. Denna valideringsmetod användes därefter för att fastställa det dataset som bidrog till systemet med bäst förmåga att detektera fordon. De dataset som användes i denna studien var BDD100K, KITTI och Udacity, som tränades på individuella igenkänningsmodeller. Genom att applicera denna valideringsmetod, fastställdes det att BDD100K var det dataset som bidrog till systemet med bäst presterande igenkänningsförmåga. En analys av dataset storlek, etikettdistribution och genomsnittliga antalet etiketter per bild var även genomförd. Tillsammans med ett experiment som genomfördes för att testa modellerna i verkliga sammanhang, kunde det avgöras att valideringsmetoden stämde överens med de fastställda resultaten. Slutligen studerades TensorFlow Object Detection APIs förmåga att förbättra prestandan som erhålls av ett visuellt dataset. Genom användning av ett modifierat dataset, kunde det fastställas att TensorFlow Object Detection API är ett lämpligt modifieringsverktyg som kan användas för att öka prestandan av ett visuellt dataset.

With today’s focus on Continuous Integration, test cases are used to ensure the software’s reliability when integrating and developing code. Test cases that behave in an undeterministic manner are known as flaky tests, which threatens the software’s reliability. Because of flaky test’s undeterministic nature, they can be troublesome to detect and correct. This is causing companies to spend great amount of resources on flaky tests since they can reduce the quality of their products and services. The aim of this thesis was to develop a usable tool that can automatically detect flakiness in the Randomness category. This was done by initially locating and rerunning flaky tests found in public Git repositories. By scanning the resulting pytest logs from the tests that manifested flaky behaviour, noting indicators of how flakiness manifests in the Randomness category. From these findings we determined tracing to be a viable option of detecting Randomness as a cause of flakiness. The findings were implemented into our proposed tool FlakyReporter, which reruns flaky tests to determine if they pertain to the Randomness category. Our FlakyReporter tool was found to accurately categorise flaky tests into the Randomness category when tested against 25 different flaky tests. This indicates the viability of utilizing tracing as a method of categorizing flakiness.

Power grids have rapidly evolved into Smart grids and are heavily dependent on Supervisory Control and Data Acquisition (SCADA) systems for monitoring and control. However, this evolution increases the susceptibility of the remote (VMs, VPNs) and physical interfaces (sensors, PMUs LAN, WAN, sub-stations power lines, and smart meters) to sophisticated cyberattacks. The continuous supply of power is critical to power generation plants, power grids, industrial grids, and nuclear grids; the halt to global power could have a devastating effect on the economy's critical infrastructures and human life. Machine Learning and Deep Learning-based cyberattack detection modeling have yielded promising results when combined as a Hybrid with an Intrusion Detection System (IDS) or Host Intrusion Detection Systems (HIDs). This thesis proposes two cyberattack detection techniques; one that leverages Machine Learning algorithms and the other that leverages Artificial Neural networks algorithms to classify and detect the cyberattack data held in a foundational dataset crucial to network intrusion detection modeling. This thesis aimed to analyze and evaluate the performance of a Hybrid Machine Learning (ML) and a Hybrid Deep Learning (DL) during ingress packet filtering, class classification, and anomaly detection on a Smart grid network.

Dans cette dissertation, nous discutons comment utiliser les données du regard humain pour améliorer la performance du modèle d'apprentissage supervisé faible dans la classification des images. Le contexte de ce sujet est à l'ère de la technologie de l'information en pleine croissance. En conséquence, les données à analyser augmentent de façon spectaculaire. Étant donné que la quantité de données pouvant être annotées par l'humain ne peut pas tenir compte de la quantité de données elle-même, les approches d'apprentissage supervisées bien développées actuelles peuvent faire face aux goulets d'étranglement l'avenir. Dans ce contexte, l'utilisation de annotations faibles pour les méthodes d'apprentissage à haute performance est digne d'étude. Plus précisément, nous essayons de résoudre le problème à partir de deux aspects: l'un consiste à proposer une annotation plus longue, un regard de suivi des yeux humains, comme une annotation alternative par rapport à l'annotation traditionnelle longue, par exemple boîte de délimitation. L'autre consiste à intégrer l'annotation du regard dans un système d'apprentissage faiblement supervisé pour la classification de l'image. Ce schéma bénéficie de l'annotation du regard pour inférer les régions contenant l'objet cible. Une propriété utile de notre modèle est qu'elle exploite seulement regardez pour la formation, alors que la phase de test est libre de regard. Cette propriété réduit encore la demande d'annotations. Les deux aspects isolés sont liés ensemble dans nos modèles, ce qui permet d'obtenir des résultats expérimentaux compétitifs. / In this dissertation, we discuss how to use the human gaze data to improve the performance of the weak supervised learning model in image classification. The background of this topic is in the era of rapidly growing information technology. As a consequence, the data to analyze is also growing dramatically. Since the amount of data that can be annotated by the human cannot keep up with the amount of data itself, current well-developed supervised learning approaches may confront bottlenecks in the future. In this context, the use of weak annotations for high-performance learning methods is worthy of study. Specifically, we try to solve the problem from two aspects: One is to propose a more time-saving annotation, human eye-tracking gaze, as an alternative annotation with respect to the traditional time-consuming annotation, e.g. bounding box. The other is to integrate gaze annotation into a weakly supervised learning scheme for image classification. This scheme benefits from the gaze annotation for inferring the regions containing the target object. A useful property of our model is that it only exploits gaze for training, while the test phase is gaze free. This property further reduces the demand of annotations. The two isolated aspects are connected together in our models, which further achieve competitive experimental results.

Le sens d’un mot est sujet à des variations au cours du temps. Nombre de phénomènes motivent ces modifications comme l’apparition de nouveaux objets ou les changements d’habitudes. Ainsi, un même mot peut se voir assigner un nouveau sens, retirer un sens, ou encore rester stable entre deux dates. L’étude de la diachronie sémantique est un domaine s’intéressant à ces changements de sens. Les récents travaux sur la diachronie sémantique proposent des méthodologies pour le repérage de diachronies. Pour ce faire, ils s’appuient sur des textes issus de plusieurs périodes temporelles différentes, et grâce auxquels sont entrainés des modèles de langue. Un alignement des représentations obtenues, et une comparaison de celles de mots-cibles leur permet de conclure quant à leur changement de sens. Néanmoins, l’absence de jeu de données (dataset) de référence pour la validation de ces méthodes conduit au développement de méthodes de validation alternatives, suggérant notamment de s’appuyer sur les changements de sens recensés dans les dictionnaires traditionnels. Le travail réalisé au cours de ma maitrise s’attache à exposer une réflexion sur les méthodes existantes de repérage des diachronies. En nous appuyant sur un corpus journalistique couvrant l’ensemble du XXème siècle, nous proposons des méthodes complémentaires grâce auxquelles nous démontrons que les évaluations proposées font l’objet d’ambiguïtés. Celles-ci ne permettent dès lors pas de conclure quant à la qualité des méthodes. Nous nous sommes ensuite attachés à développer une méthodologie pour la construction d’un jeu de données de validation. Cette méthodologie tire parti d’un algorithme de désambiguïsation afin d’associer à tous les sens recensés d’un mot une date d’apparition au cours du temps. Nous proposons un jeu de données composé de 151 mots permettant d’évaluer le repérage de diachronies en français entre 1910 et 1990. / The meaning of a word is subject to variations over time. Many phenomena motivate these modifications such as the appearance of new objects or changes in habits. Thus, the same word can be assigned a new meaning, or have a meaning withdrawn, or remain stable between two dates. The study of semantic diachrony is a field that focuses on these changes in meaning. Recent work on semantic diachrony proposes methodologies for the detection of diachronies. In order to do so, they rely on texts from several different temporal periods, and through which language models are trained. An alignment of the obtained representations, and a comparison of those of target words enables one to infer the change of meaning. Nevertheless, the absence of a reference dataset for the validation of these methods leads to the development of alternative validation methods, suggesting in particular to rely on the changes of meaning identified in traditional dictionaries. The work carried out during my master's degree aims at presenting a reflection on the existing methods of diachrony detection. Based on a corpus of newspapers covering the whole 20th century, we propose complementary methods thanks to which we demonstrate that the proposed evaluations are subject to ambiguities. These ambiguities do not allow us to ensure the quality of the methods. We then develop a methodology for the construction of a validation dataset. This methodology takes advantage of a disambiguation algorithm in order to associate a date of appearance in the course of time to all the senses of a word. We propose a dataset composed of 151 words allowing one to evaluate the identification of diachronies in French between 1910 and 1990.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第21397号 / 情博第683号 / 新制||情||118(附属図書館) / 京都大学大学院情報学研究科社会情報学専攻 / (主査)教授 吉川 正俊, 教授 黒田 知宏, 教授 守屋 和幸 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

No description available.