1. Cephalometric landmark detection in lateral skull X-ray images by using improved spatialconfiguration-netMartin Šavc, Gašper Sedej, Božidar Potočnik, 2022, izvirni znanstveni članek Opis: Accurate automated localization of cephalometric landmarks in skull X-ray images is the
basis for planning orthodontic treatments, predicting skull growth, or diagnosing face discrepancies.
Such diagnoses require as many landmarks as possible to be detected on cephalograms. Today’s
best methods are adapted to detect just 19 landmarks accurately in images varying not too much.
This paper describes the development of the SCN-EXT convolutional neural network (CNN), which
is designed to localize 72 landmarks in strongly varying images. The proposed method is based
on the SpatialConfiguration-Net network, which is upgraded by adding replications of the simpler
local appearance and spatial configuration components. The CNN capacity can be increased without
increasing the number of free parameters simultaneously by such modification of an architecture.
The successfulness of our approach was confirmed experimentally on two datasets. The SCN-EXT
method was, with respect to its effectiveness, around 4% behind the state-of-the-art on the small ISBI
database with 250 testing images and 19 cephalometric landmarks. On the other hand, our method
surpassed the state-of-the-art on the demanding AUDAX database with 4695 highly variable testing
images and 72 landmarks statistically significantly by around 3%. Increasing the CNN capacity
as proposed is especially important for a small learning set and limited computer resources. Our
algorithm is already utilized in orthodontic clinical practice.
Ključne besede: detection of cephalometric landmarks, skull X-ray images, convolutional neural networks, deep learning, SpatialConfiguration-Net architecture, AUDAX database
Objavljeno v DKUM: 27.03.2025; Ogledov: 0; Prenosov: 0
 Celotno besedilo (2,46 MB)
Gradivo ima več datotek! Več... |
2. Deeply-supervised 3D convolutional neural networks for automated ovary and follicle detection from ultrasound volumesBožidar Potočnik, Martin Šavc, 2022, izvirni znanstveni članek Opis: Automated detection of ovarian follicles in ultrasound images is much appreciated when
its effectiveness is comparable with the experts’ annotations. Today’s best methods estimate follicles
notably worse than the experts. This paper describes the development of two-stage deeply-supervised
3D Convolutional Neural Networks (CNN) based on the established U-Net. Either the entire U-Net
or specific parts of the U-Net decoder were replicated in order to integrate the prior knowledge into
the detection. Methods were trained end-to-end by follicle detection, while transfer learning was
employed for ovary detection. The USOVA3D database of annotated ultrasound volumes, with its
verification protocol, was used to verify the effectiveness. In follicle detection, the proposed methods
estimate follicles up to 2.9% more accurately than the compared methods. With our two-stage CNNs
trained by transfer learning, the effectiveness of ovary detection surpasses the up-to-date automated
detection methods by about 7.6%. The obtained results demonstrated that our methods estimate
follicles only slightly worse than the experts, while the ovaries are detected almost as accurately as
by the experts. Statistical analysis of 50 repetitions of CNN model training proved that the training
is stable, and that the effectiveness improvements are not only due to random initialisation. Our
deeply-supervised 3D CNNs can be adapted easily to other problem domains.
Ključne besede: 3D deep neural networks, 3D ultrasound images of ovaries, deep supervision, detection of follicles and ovaries, U-Net based architecture
Objavljeno v DKUM: 27.03.2025; Ogledov: 0; Prenosov: 0
Celotno besedilo (1,28 MB)
Gradivo ima več datotek! Več... |
3. Event square and multipurpose center with library in the city of StrugaJeta Ajdari, 2024, magistrsko delo Opis: This thesis addresses the redevelopment of a centrally located site in Struga, currently occupied by a large, unadorned parking lot devoid of greenery. Surrounded by key state institutions, including the Health Insurance Office, Municipal Archives, and the Main Post Office, as well as public amenities, such as restaurants and pharmacies, the site offers a unique opportunity for urban revitalization.Until 2016, this location served as the central square of Struga. Following the restoration of the "Monument of Revolution," it was expected that the area would be transformed into a modern, attractive space for local residents. However, contrary to these expectations, it has since been repurposed as a parking lot. This project envisions reclaiming the site as a multifunctional space that serves as a cultural hub and social gathering point for both residents and visitors.Struga, situated on the shores of Lake Ohrid in the Republic of North Macedonia, has a population of approximately 50,000 and is a popular summer tourist destination. Despite this, the city faces a deficit of social gathering places and environments conducive to student collaboration and study. The existing city library, with its outdated facilities and limited collection of 8,000 books, fails to meet modern standards. Furthermore, Struga’s renowned Poetry Nights festival underscores the lack of dedicated public spaces for cultural events, highlighting the need for proper outdoor venues. The absence of adequate library resources and diverse social spaces for various age groups further exacerbates these challenges.This project proposes a comprehensive solution by transforming the parking lot into a multifaceted community facility. The envisioned development includes an open square for events, a gallery and exhibition hall, an innovative city library, study areas, communal spaces, green zones, and a children’s playground. Outdoor reading areas and organized play activities will be integrated into the design, promoting relaxation and leisure within an inviting, green environment.By combining diverse indoor and outdoor programs, the project seeks to foster social interaction across all age groups, enhancing community cohesion. This multi-generational event center and library will become a landmark for Struga, enriching the city’s cultural, recreational, and educational landscape while significantly improving the quality of life for its residents.
Ključne besede: contemporary architecture, library, multigeneration building, square, public space, common space, green space, Struga.
Objavljeno v DKUM: 13.01.2025; Ogledov: 0; Prenosov: 55
Celotno besedilo (94,06 MB) |
4. Analysis of Methodologies and Tools for Software Development in Different ArchitecturesMaksim Nikitashin, 2024, magistrsko delo Opis: In modern software development, three of the most popular application architectures are commonly used: monolithic, service-oriented, and microservices. At the same time, plenty of different methodologies, primarily agile ones, and tools are in use to plan and manage software projects. The research studies relations between these two entities to define the most popular and suitable methodologies and tools for projects, implementing each specific architecture. The research consists of 2 parts: qualitative and quantitative. The first one includes the study of the theoretical basis and related sources to find out the most popular modern methodologies and tools considering their characteristics and each specific architecture. Based on these characteristics, a definition of the most suitable ones is made. The second part represents preliminary statistical research among people, currently employed in software development companies about their opinions on the thematic. The second part includes 36 professionals, employed in different roles. In this part, descriptive statistics and correlation analysis are used. The results have some limitations related to the limited scope of literature studied and also to the size and representativeness of the sample. However, they are useful both for people working in software development, especially for leading personnel, making decisions about project management, and for researchers as a good starting point for continuing research. The results show the necessity of further investigation of the topic due to inconsistency between the scientific literature research and real business world research. Also, we have found out possible correlation between employees’ satisfaction and tools used in project management.
In reference to IEEE copyrighted material which is used with permission in this thesis, the IEEE does not endorse any of University of Maribor’s products or services. Internal or personal use of this material is permitted. If interested in reprinting/republishing IEEE copyrighted material for advertising or promotional purposes or for creating new collective works for resale or redistribution, please go to http://www.ieee.org/publications_standards/publications/rights/rights_link.html to learn how to obtain a License from RightsLink. If applicable, University Microfilms and/or ProQuest Library, or the Archives of Canada may supply single copies of the dissertation.
Ključne besede: agile methodologies, software, monolithic architecture, service-oriented architecture, microservices
Objavljeno v DKUM: 25.09.2024; Ogledov: 0; Prenosov: 23
Celotno besedilo (1,01 MB) |
5. A review of federated learning in agricultureKrista Rizman Žalik, Mitja Žalik, 2023, pregledni znanstveni članek Opis: Federated learning (FL), with the aim of training machine learning models using data and computational resources on edge devices without sharing raw local data, is essential for improving agricultural management and smart agriculture. This study is a review of FL applications that address various agricultural problems. We compare the types of data partitioning and types of FL (horizontal partitioning and horizontal FL, vertical partitioning and vertical FL, and hybrid partitioning and transfer FL), architectures (centralized and decentralized), levels of federation (cross-device and cross-silo), and the use of aggregation algorithms in different reviewed approaches and applications of FL in agriculture. We also briefly review how the communication challenge is solved by different approaches. This work is useful for gaining an overview of the FL techniques used in agriculture and the progress made in this field.
Ključne besede: federated learning, agriculture, architecture, data partitioning, federation scal, aggregation algorithms, communication bottleneck
Objavljeno v DKUM: 05.06.2024; Ogledov: 142; Prenosov: 26
Celotno besedilo (839,33 KB)
Gradivo ima več datotek! Več... |
6. Knowledge Graph Completion with Triple Structure and Text RepresentationShuang Liu, Yufeng Qin, Man Xu, Simon Kolmanič, 2023, izvirni znanstveni članek Opis: Knowledge Graphs (KGs) describe objective facts in the form of RDF triples, each triple contains sufficient semantic information and triple structure information. Knowledge Graph Completion (KGC) is to acquire new knowledge by predicting hidden relationships between entities and adding the new knowledge to the KG. At present, the mainstream KGC approaches only applied the triple structure information or only utilized the semantic information of the text. This paper proposes an approach (TSTR) using BERT and deep neural networks to fully extract the semantic information of knowledge, and designs an aggregated re-ranking scheme that incorporates existing graph embedding approach to learn the structural information of triples. In experiments, the approach achieves state-of-the-art performance on three benchmark datasets, and outperforms recent KGC approaches on sparsely connected datasets.
Ključne besede: knowledge graph completion, BERT, deep convolutional architecture, re-ranking
Objavljeno v DKUM: 19.02.2024; Ogledov: 229; Prenosov: 21
Celotno besedilo (1,03 MB)
Gradivo ima več datotek! Več... |
7. Signing blockchain transactions using qualified certificatesMuhamed Turkanović, Blaž Podgorelec, izvirni znanstveni članek Opis: Blockchain technology is increasingly being considered among both private
enterprises and public services. However, it poses a challenge with regard to aligning its identity management scheme with the Public Key Infrastructure and the Qualified Digital Certificates issued by Qualified Trust Service Providers. To solve this challenge, we will present a solution in the form of an architecture reference model, which enables enterprises and public services to leverage blockchain technology by integrating Qualified Electronic Signatures with blockchain transactions. The evaluation of the architecture reference model is provided through the design of a Blockchain-based Trusted Public Service and a use-case scenario example. The proposed architecture reference model is based on the CEF building blocks EBSI, eSignature, and eID compliant with eIDAS
Ključne besede: blockchain, computer architecture, distributed ladger, fabrics, elliptic curve cryptography, internet, law
Objavljeno v DKUM: 08.12.2022; Ogledov: 712; Prenosov: 175
Celotno besedilo (215,95 KB)
Gradivo ima več datotek! Več... |
8. Construction of deep neutral networks using swarm intelligence to detect anomalies : master's thesisSašo Pavlič, 2021, magistrsko delo Opis: The design of neural network architecture is becoming more difficult as the complexity of the problems we tackle using machine learning increases. Many variables influence the performance of a neural model, and those variables are often limited by the researcher's prior knowledge and experience. In our master's thesis, we will focus on becoming familiar with evolutionary neural network design, anomaly detection techniques, and a deeper knowledge of autoencoders and their potential for application in unsupervised learning. Our practical objective will be to build a neural architecture search based on swarm intelligence, and construct an autoencoder architecture for anomaly detection in the MNIST dataset.
Ključne besede: neural architecture search, machine learning, swarm intelligence
Objavljeno v DKUM: 18.10.2021; Ogledov: 1269; Prenosov: 111
Celotno besedilo (3,18 MB)
Gradivo ima več datotek! Več... |
9. Cultural complex by the river Morača, Podgorica, Montenegro : master thesisFilip Radojević, 2021, magistrsko delo Opis: The masters’ thesis aims to examine the practice of appropriation of inherent qualities of certain sites in the city and the necessity of their preservation against dramatically developing trend of private real-estate and consumer-capitalism driven land-claim. The latter, in combination with inadequate response from government institutions can contribute to a total loss of areas of public interest in the city. The location of the former ”Kasarna Morača” military barracks in Podgorica makes for a valuable target for the greedy investors, but also to the public, which could benefit from a large mass of greenery in the very centre of the capital next to the Morača river. For this reason, questions have been raised sparking public debates on how this location, currently an asset of the state, ought to be managed. This project aims to develop an urban regeneration solution for the brownfield area by means of two interconnected segments. The first one is the recreation area along the Morača riverbank which incorporates a large part of the existing greenery and extends to the Mausoleum of the Partisan fighter on the Gorica hill. The second is situated at the Eastern edge of the site. With the adaptive reuse of the existing barrack structures and a new adaptable and flexible structural system, it proposes a set of public and cultural programs along the”Vaka Đurovića” boulevard. Both the green belt and the cultural complex are connected via pedestrian and cycling paths which also lead to a proposed pedestrian bridge connecting the public institutions, neighbourhoods and recreational areas along Morača riverbanks.
Ključne besede: culture, recreation, complex, urban connector, ”weak” architecture, Morača, bridge
Objavljeno v DKUM: 01.10.2021; Ogledov: 993; Prenosov: 201
Celotno besedilo (60,86 MB) |
10. Revitalization of the Walter Perić Power Plant in Sarajevo : master thesisMerima Tica, 2021, magistrsko delo Opis: The master thesis focuses on the analysis and research of Sarajevo’s central area of Marijin Dvor, and specifically the revitalization of the electrical power plant Walter Perić. The focus is on understanding the proper ways in which degraded zone and dilapidated buildings ought to be revitalized, but also the benefits of revitalizing city cores. It is a multifaceted discourse, thus, the research focuses on industrial heritage and brownfield development in the context of Sarajevo’s City Planning Strategy, as well as, the concept of ruins. The analyses culminate in project offering a new urbanism proposal for the degraded zones in Marijin Dvor, with a focus on the adaptive reuse of the former electrical power plant complex with a new cultural program.
Ključne besede: industrial heritage architecture, adaptive reuse, ruins, Marijin Dvor, Sarajevo
Objavljeno v DKUM: 28.09.2021; Ogledov: 989; Prenosov: 138
Celotno besedilo (102,03 MB) |
