Iskanje po katalogu digitalne knjižnice

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...

Opis: Zaključno delo obravnava problem kompleksnosti konfiguracije produktnih katalogov v sistemu Monetization podjetja Tridens, kjer se uporabniki pogosto soočajo z izzivi pri razumevanju in nastavitvi zapletenih obračunskih modelov. Cilj dela je razviti umetno inteligenco, ki bo s pomočjo velikih jezikovnih modelov (LLM), kot je ChatGPT, omogočila enostavnejšo in bolj intuitivno konfiguracijo produktnih katalogov. V okviru raziskave smo analizirali obstoječo tehnično dokumentacijo in bazo znanja podjetja Tridens ter izvedli integracijo modela ChatGPT v zaledni sistem za avtomatizirano generiranje in prilagajanje produktnih katalogov. Rešitev smo preizkusili s testnim sklopom povpraševanj, pri čemer smo primerjali rezultate generiranih konfiguracij z referenčnimi primeri. Rezultati so pokazali, da umetna inteligenca uspešno izboljšuje natančnost in hitrost konfiguracije produktnih katalogov, hkrati pa zmanjšuje potrebo po ročnem delu. Zaključek naloge potrjuje, da uporaba velikih jezikovnih modelov prispeva k izboljšanju uporabniške izkušnje in poenostavitvi konfiguracijskih procesov, kar lahko podjetjem, kot je Tridens, prinese pomembne prednosti v konkurenčnem okolju.
Ključne besede: veliki jezikovni modeli, umetna inteligenca, produktni katalog, chatGPT, java
Objavljeno v DKUM: 06.02.2025; Ogledov: 0; Prenosov: 23
Celotno besedilo (3,25 MB)

Opis: V diplomskem delu smo preučevali in analizirali rezultate arhitektur konvolucijskih nevronskih mrež na problemu binarne segmentacije. V teoretičnem delu smo preučili arhitekture konvolucijskih nevronskih mrež SegFormer, DeepLapV3+, Gated-SCNN, Segmenter, FastFCN in TopFormer. V praktičnem delu diplomskega dela smo mreže SegFormer, Segmenter, FastFCN in TopFormer učili segmentirati slike na podatkovnih množicah Cityscapes in ADE20K. Mreže smo učili binarne segmentacije vozil. Mrežo SegFormer smo še dodatno učili na problemu binarnega segmentiranja ljudi. Dobljene rezultate smo analizirali in jih ovrednotili z evalvacijskimi metrikami. Za ocenjevanje uspešnosti smo uporabili metrike točnost, priklic, natančnost, IoU in F1 oceno. Najboljše rezultate za problem binarnega segmentiranja vozil smo dobili s pomočjo mreže SegFormer na podatkovni zbirki Cityscapes. Na podatkovni zbirki ADE20K smo za enak problem segmentacije dobili najboljše rezultate za mrežo Segmenter. Najslabše sta se izkazali mreža TopFormer na podatkovni zbirki Cityscapes in mreža FastFCN na podatkovni zbirki ADE20K.
Ključne besede: binarna segmentacija, konvolucijske nevronske mreže, primerjava mrež, evalvacijske metrike, računalniški vid
Objavljeno v DKUM: 26.04.2024; Ogledov: 506; Prenosov: 90
Celotno besedilo (3,19 MB)

Opis: A few promising solutions for thermal imaging Unexploded Ordnance (UXO) detection were proposed after the start of the military conflict in Ukraine in 2014. At the same time, most of the landmine clearance protocols and practices are based on old, 20th-century technologies. More than 60 countries worldwide are still affected by explosive remnants of war, and new areas are contaminated almost every day. To date, no automated solutions exist for surface UXO detection by using thermal imaging. One of the reasons is also that there are no publicly available data. This research bridges both gaps by introducing an automated UXO detection method, and by publishing thermal imaging data. During a project in Bosnia and Herzegovina in 2019, an organisation, Norwegian People's Aid, collected data about unexploded ordnances and made them available for this research. Thermal images with a size of 720 x 480 pixels were collected by using an Unmanned Aerial Vehicle at a height of 3 m, thus achieving a very small Ground Sampling Distance (GSD). One of the goals of our research was also to verify if the explosive war remnants' detection accuracy could be improved further by using Convolutional Neural Networks (CNN). We have experimented with various existing modern CNN architectures for object identification, whereat the YOLOv5 model was selected as the most promising for retraining. An eleven-class object detection problem was solved primarily in this study. Our data were annotated semi-manually. Five versions of the YOLOv5 model, fine-tuned with a grid-search, were trained end-to-end on randomly selected 640 training and 80 validation images from our dataset. The trained models were verified on the remaining 88 images from our dataset. Objects from each of the eleven classes were identified with more than 90% probability, whereat the Mean Average Precision (mAP) at a 0.5 threshold was 99.5%, and the mAP at thresholds from 0.5 to 0.95 was 87.0% up to 90.5%, depending on the model's complexity. Our results are comparable to the state-of-the-art, whereat these object detection methods have been tested on other similar small datasets with thermal images. Our study is one of the few in the field of Automated UXO detection by using thermal images, and the first that solves the problem of identifying more than one class of objects. On the other hand, publicly available thermal images with a relatively small GSD will enable and stimulate the development of new detection algorithms, where our method and results can serve as a baseline. Only really accurate automatic UXO detection solutions will help to solve one of the least explored worldwide life-threatening problems.
Ključne besede: unmanned aerial vehicle, unexploded ordnance, thermal imaging, UXOTi_NPA dataset, convolutional neural networks, deep learning
Objavljeno v DKUM: 12.02.2024; Ogledov: 389; Prenosov: 26
Celotno besedilo (16,94 MB)
Gradivo ima več datotek! Več...

Opis: V tej diplomski nalogi se ukvarjamo s kombiniranjem različnih prosto dostopnih modelov konvolucijskih nevronskih mrež za reševanje problema ocenjevanja starosti oseb iz digitalnih posnetkov. Naš cilj je bil implementirati ansambelske metode, ki so bolj uspešne od vsakega posameznega modela v našem ansamblu. Kombinirali smo pet osnovnih modelov konvolucijskih nevronskih mrež, in sicer Xception, ResNet152V2, InceptionV3, InceptionResNetV2 in EfficientNetV2B0. Modele smo kombinirali na nivoju rezultatov in na nivoju značilnic, pri čemer smo implementirali štiri metode kombiniranja na nivoju rezultatov in pet metod kombiniranja na nivoju značilnic. Za učenje posameznih modelov in testiranje metod smo uporabili bazo podatkov UTKFace. Najboljši rezultat smo dosegli z metodo večslojnega »stackinga«, in sicer srednjo absolutno napako 4,526 let nad našo testno množico. Rezultat le malenkost zaostaja za rezultati najboljših sodobnih metod ocenjevanja starosti iz digitalnih posnetkov.
Ključne besede: ocenjevanje starosti oseb, konvolucijske nevronske mreže, ansambelsko učenje
Objavljeno v DKUM: 05.10.2023; Ogledov: 424; Prenosov: 33
Celotno besedilo (1,75 MB)