1. A novel system for quasi-continuous THz signal transmission and receptionAndrej Sarjaš, Blaž Pongrac, Dušan Gleich, 2022, izvirni znanstveni članek Opis: This paper presents a novel system for generating and receiving quasi-continuous (QC)
TeraHertz (THz) waves. A system design and theoretical foundation for QC-THz signal generation
are presented. The proposed QC-THz system consists of commercially available photo-conductive
antennas used for transmission and reception of THz waves and a custom-designed QC optical
signal generator, which is based on a fast optical frequency sweep of a single telecom distributedfeedback laser diode and unbalanced optical fiber Michelson interferometer used for a high-frequency
modulation. The theoretical model for the proposed system is presented and experimentally evaluated. The experimental results were compared to the state-of-the-art continuous-wave THz system. The comparison between the continuous-wave THz system and the proposed QC-THz system
showed the ability to transmit and receive QC-THz waves up to 300 GHz. The upper-frequency limit
is bounded by the length of the used Michelson interferometer. The presented design of THz signal
generation has a potential for industrial application because it is cost-efficient and can be built using
commercially available components.
Ključne besede: quasi-continuous, terahertz, photoconductive antenna, wave emittance, wave detection
Objavljeno v DKUM: 01.04.2025; Ogledov: 0; Prenosov: 2
 Celotno besedilo (1,87 MB)
Gradivo ima več datotek! Več... |
2. Using a region-based convolutional neural network (R-CNN) for potato segmentation in a sorting processJaka Verk, Jernej Hernavs, Simon Klančnik, 2025, izvirni znanstveni članek Opis: This study focuses on the segmentation part in the development of a potato-sorting system that utilizes camera input for the segmentation and classification of potatoes. The key challenge addressed is the need for efficient segmentation to allow the sorter to handle a higher volume of potatoes simultaneously. To achieve this, the study employs a region-based convolutional neural network (R-CNN) approach for the segmentation task, while trying to achieve more precise segmentation than with classic CNN-based object detectors. Specifically, Mask R-CNN is implemented and evaluated based on its performance with different parameters in order to achieve the best segmentation results. The implementation and methodologies used are thoroughly detailed in this work. The findings reveal that Mask R-CNN models can be utilized in the production process of potato sorting and can improve the process.
Ključne besede: image segmentation, potato sorting, neural network, mask RCNN, object detection, production process, machine learning, AI
Objavljeno v DKUM: 27.03.2025; Ogledov: 0; Prenosov: 9
Celotno besedilo (5,97 MB)
Gradivo ima več datotek! Več... |
3. 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: 5
Celotno besedilo (2,46 MB)
Gradivo ima več datotek! Več... |
4. 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: 5
Celotno besedilo (1,28 MB)
Gradivo ima več datotek! Več... |
5. A cloud-based system for the optical monitoring of tool conditions during milling through the detection of chip surface size and identification of cutting force trendsUroš Župerl, Krzysztof Stępień, Goran Munđar, Miha Kovačič, 2022, izvirni znanstveni članek Opis: This article presents a cloud-based system for the on-line monitoring of tool conditions in
end milling. The novelty of this research is the developed system that connects the IoT (Internet of
Things) platform for the monitoring of tool conditions in the cloud to the machine tool and optical
system for the detection of cutting chip size. The optical system takes care of the acquisition and
transfer of signals regarding chip size to the IoT application, where they are used as an indicator
for the determination of tool conditions. In addition, the novelty of the presented approach is in
the artificial intelligence integrated into the platform, which monitors a tool’s condition through
identification of the current cutting force trend and protects the tool against excessive loading by
correcting process parameters. The practical significance of the research is that it is a new system for
fast tool condition monitoring, which ensures savings, reduces investment costs due to the use of
a more cost-effective sensor, improves machining efficiency and allows remote process monitoring
on mobile devices. A machining test was performed to verify the feasibility of the monitoring
system. The results show that the developed system with an ANN (artificial neural network) for the
recognition of cutting force patterns successfully detects tool damage and stops the process within
35 ms. This article reports a classification accuracy of 85.3% using an ANN with no error in the
identification of tool breakage, which verifies the effectiveness and practicality of the approach.
Ključne besede: machining, end milling, tool condition monitoring, chip size detection, cutting force trend identification, visual sensor monitoring, cloud manufacturing technologies
Objavljeno v DKUM: 26.03.2025; Ogledov: 0; Prenosov: 2
Celotno besedilo (5,65 MB)
Gradivo ima več datotek! Več... |
6. Improved relation extraction through key phrase identification using community detection on dependency treesShuang Liu, Xunqin Chen, Jiana Meng, Niko Lukač, 2025, izvirni znanstveni članek Opis: A method for extracting relations from sentences by utilizing their dependency trees to identify key phrases is presented in this paper. Dependency trees are commonly used in natural language processing to represent the grammatical structure of a sentence, and this approach builds upon this representation to extract meaningful relations between phrases. Identifying key phrases is crucial in relation extraction as they often indicate the entities and actions involved in a relation. The method uses community detection algorithms on the dependency tree to identify groups of related words that form key phrases, such as subject-verb-object structures. The experiments on the Semeval-2010 task8 dataset and the TACRED dataset demonstrate that the proposed method outperforms existing baseline methods.
Ključne besede: community detection algorithms, dependency trees, relation extraction
Objavljeno v DKUM: 17.01.2025; Ogledov: 0; Prenosov: 5
Celotno besedilo (3,12 MB) |
7. Dimethoate detection through a fluorescent coumarin dyeEdoardo Donà, Gerhard J. Mohr, Aleksandra Lobnik, 2024, izvirni znanstveni članek Opis: In this study, we present a straightforward and innovative approach utilizing a coumarin fluorescent dye for the detection of dimethoate in green tea. Initially, the pesticide undergoes hydrolysis in a NaOH solution, yielding our target analyte, methylamine. Following neutralization to pH 9, methylamine reacts with the dye in CH3CN for 20 min. After a careful optimization, we achieved an outstanding linear correlation (R2 = 0.999) for dimethoate, spanning concentrations from 7.8 to 292 µg/L and LOD of 3.2 µg/L. Moreover, we successfully detected dimethoate in green tea, with a recovery of 95.4% (σ = 5.7%). Organophosphates pesticides (OPs), which dimethoate is one of the most used, pose a significant threat due to their toxicity upon both high direct exposure and prolonged low-level exposure, which has been linked to cancer. Therefore, the development of a detection method that is both selective and sensitive is imperative for safeguarding both the population and the environment. This method effectively addresses the stability challenges encountered by enzyme-based fluorescent sensors, thereby opening new avenues for the detection of organophosphate pesticides.
Ključne besede: organophosphate pesticide, methylamine, coumarin, fluorescence detection, dimethoate, TICT, twisted intramolecular charge transfer
Objavljeno v DKUM: 20.12.2024; Ogledov: 0; Prenosov: 3
Celotno besedilo (1,77 MB)
Gradivo ima več datotek! Več... |
8. Recent developments in electrochemical-impedimetric biosensors for virus detectionZala Štukovnik, Urban Bren, 2022, pregledni znanstveni članek Opis: Viruses, including influenza viruses, MERS-CoV (Middle East respiratory syndrome coronavirus), SARS-CoV (severe acute respiratory syndrome coronavirus), HAV (Hepatitis A virus), HBV (Hepatitis B virus), HCV (Hepatitis C virus), HIV (human immunodeficiency virus), EBOV (Ebola virus), ZIKV (Zika virus), and most recently SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2), are responsible for many diseases that result in hundreds of thousands of deaths yearly. The ongoing outbreak of the COVID-19 disease has raised a global concern and intensified research on the detection of viruses and virus-related diseases. Novel methods for the sensitive, rapid, and on-site detection of pathogens, such as the recent SARS-CoV-2, are critical for diagnosing and treating infectious diseases before they spread and affect human health worldwide. In this sense, electrochemical impedimetric biosensors could be applied for virus detection on a large scale. This review focuses on the recent developments in electrochemical-impedimetric biosensors for the detection of viruses.
Ključne besede: electrochemical impedance spectroscopy, impedimetric biosensor, genosensor, aptasensor, immunosensor, virus detection, SARS-CoV-2, HIV, influenza virus, hepatitis virus
Objavljeno v DKUM: 05.12.2024; Ogledov: 0; Prenosov: 12
Celotno besedilo (1,56 MB)
Gradivo ima več datotek! Več... |
9. Tilt correction toward building detection of remote sensing imagesKang Liu, Zhiyu Jiang, Mingliang Xu, Matjaž Perc, Xuelong Li, 2021, izvirni znanstveni članek Opis: Building detection is a crucial task in the field of remote sensing, which can facilitate urban construction planning, disaster survey, and emergency landing. However, for large-size remote sensing images, the great majority of existing works have ignored the image tilt problem. This problem can result in partitioning buildings into separately oblique parts when the large-size images are partitioned. This is not beneficial to preserve semantic completeness of the building objects. Motivated by the above fact, we first propose a framework for detecting objects in a large-size image, particularly for building detection. The framework mainly consists of two phases. In the first phase, we particularly propose a tilt correction (TC) algorithm, which contains three steps: texture mapping, tilt angle assessment, and image rotation. In the second phase, building detection is performed with object detectors, especially deep-neural-network-based methods. Last but not least, the detection results will be inversely mapped to the original large-size image. Furthermore, a challenging dataset named Aerial Image Building Detection is contributed for the public research. To evaluate the TC method, we also define an evaluation metric to compute the cost of building partition. The experimental results demonstrate the effects of the proposed method for building detection.
Ključne besede: building detection, cost of building partition, deep neural network, remote sensing, tilt correction
Objavljeno v DKUM: 26.09.2024; Ogledov: 0; Prenosov: 1
Celotno besedilo (8,62 MB)
Gradivo ima več datotek! Več... |
10. Background purification framework with extended morphological attribute profile for hyperspectral anomaly detectionJu Huang, Kang Liu, Mingliang Xu, Matjaž Perc, Xuelong Li, 2021, izvirni znanstveni članek Opis: Hyperspectral anomaly detection has attracted extensive interests for its wide use in military and civilian fields, and three main categories of detection methods have been developed successively over past few decades, including statistical model-based, representation-based, and deep-learning-based methods. Most of these algorithms are essentially trying to construct proper background profiles, which describe the characteristics of background and then identify the pixels that do not conform to the profiles as anomalies. Apparently, the crucial issue is how to build an accurate background profile; however, the background profiles constructed by existing methods are not accurate enough. In this article, a novel and universal background purification framework with extended morphological attribute profiles is proposed. It explores the spatial characteristic of image and removes suspect anomaly pixels from the image to obtain a purified background. Moreover, three detectors with this framework covering different categories are also developed. The experiments implemented on four real hyperspectral images demonstrate that the background purification framework is effective, universal, and suitable. Furthermore, compared with other popular algorithms, the detectors with the framework perform well in terms of accuracy and efficiency.
Ključne besede: detectors, anomaly detection, image reconstruction, hyperspectral imaging, training, optics, dictionaries, background purification, extended attribute profile, sparse representation, stacked autoencoder
Objavljeno v DKUM: 19.08.2024; Ogledov: 92; Prenosov: 9
Celotno besedilo (5,36 MB)
Gradivo ima več datotek! Več... |
