Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 14 of 14 for “"Detectron2"”.
-
Hledání městkých prvků v satelitních snímcích
… model neurónovej siete RetinaNet a knižnica Detectron2. Model, ktorý sa podarilo vytrénovať, dokáže detekovať objekty s priemernou presnosťou (AP50) na úrovni 63,402 %. Práca môže byť prínosom v oblasti automatizovania získavania štatistík o povrchu zeme.
-
MACHINE LEARNING BASED MORPHOLOGIC CLASSIFICATION OF PROTON INDUCED CELL DEATH MODES
… a machine learning algorithm, based on the detectron2 Mask R-CNN, the YOLOv8-cls and YOLOv10 code packages.
-
Magnetic Resonance Image Algorithm to Identify Demyelination and Ischemia Brain Diseases
… U-Net, Segmenting Anything Model, YOLOv8 y Detectron2. Además, se propuso un modelo utilizando redes ResNet18 para la clasificación de Regiones de Interés (ROIs). Los resultados experimentales indicaron que el modelo U-Net logró un coeficiente Dice medio de 0.95 para la segmentación. El …
-
Improving test case diversity for functional testing in computer vision-based systems
… using two different CV algorithms: YOLOv8 and Detectron2, with a recommendation to explore AWS Rekognition in future research. To enhance the system's adaptability, robustness, and efficiency, we applied the Predictive Selection with Secondary engines pattern to further optimize its design. The …
-
The role of surface topology on reactive wetting of SAC305 solder on Ag substrates
… A Mask R-CNN model, developed using Detectron2 on hot-stage microscopy images of SAC305 solder on Ag substrates. The model achieved 99% accuracy and an average precision of 76.25%
-
Non-destructive visual quantification of food breakdown in human and robotic mastication using instance segmentation and structural descriptors
… mixing. The proposed pipeline combined Detectron2 foreground extraction, Segment Anything Model (SAM) fragment-mask generation, and six interpretable candidate descriptors of fragment density, compactness, dominance, boundary complexity, spatial separation, and size heterogeneity. …
-
Laser Surface Treatment and Laser Powder Bed Fusion Additive Manufacturing Study Using Custom Designed 3D Printer and the Application of Machine Learning in Materials Science
… performed using a simple CNN model, YOLOv4 and Detectron2, respectively. Furthermore, defects detection is successfully applied on video by using a sequence of CT scan images. It demonstrates that Machine Learning (ML) can enable more efficient and economical materials science research.
-
Modelos de estimativa de produtividade do cafeeiro por planta utilizando índices fenológicos
… contagem automática de frutos de café baseado no Detectron2, uma biblioteca de código aberto para detecção de objetos. O modelo em questão obteve na etapa de treinamento uma precisão média de 93,24% para o período de dezembro/janeiro e 95,69% para o período de janeiro/fevereiro. Com as variáveis …
-
Mapping Cardiac Adipose Tissue Using Spectral Analysis of Ultrasound Radiofrequency Data and Machine Learning
… the Segment Anything Model (SAM), Detectron2, and U-Net with ResNet backbones—were evaluated. These models achieved an average Dice Similarity Coefficient of 0.82 and a Hausdorff Distance of 4.93, indicating strong performance in delineating epicardial boundaries. For CAT mapping, …
-
Automating Manufacturing Surveillance Processes Using External Observers
… of all parts within the bowl-feeder, we train Detectron2 with data augmentation. We see that supervised segmentation outperforms pixel-based and interactive-based segmentation. To address illumination variance within datasets, we apply color-based segmentation by conversion of image datasets to …
-
Weed Recognition in Agriculture: A Mask R-CNN Approach
Recent interdisciplinary collaboration on deep learning has led to a growing interest in its application in the agriculture domain. Weed control and management are some of the crucial tasks in agriculture to maintain high crop productivity. The inception phase of weed control and management is to …
-
Crop Yield using Novel Parametric L-system plant modelling
This research considers a system for the recognition of real plant parts through image analysis based upon synthetic plant modelling. It aims to use data pipelines and synthetic datasets to define recognizable features that assist in the efficient analysis of real plants and plant images. This …
-
Detection, Extraction and Measurement of the Contour and Circumference of the Metacarpal Bones in X-rays of the Human Hand
Cieľom tejto práce je detekovať a následne extrahovať kontúru tretej metakarpálnej kosti ľudskej ruky z röntgenových snímkov a zmerať jej šírku. Práca popisuje segmentáciu obrazu pomocou metód na detekciu objektov, ktoré sa následne využijú za účelom konečných meraní šírky kosti.
-
Použití hlubokého učení a unsupervised learning pro segmentaci obrazu
Metody učení bez učitele v oblasti počítačového vidění představují slibný směr pro automatickou detekci objektů a instanční segmentaci, který nevyžaduje manuální anotaci dat. Tato práce zkoumá možnosti přenosu metody CutLER a jejích komponent na segmentaci anomálií pro zemědělské letecké snímky …