Sökning: "filter networks"

Visar resultat 1 - 5 av 128 uppsatser innehållade orden filter networks.

1. 1. Despeckling Echocardiograms Using Generative Adversarial Networks

   Master-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

   Författare :Falk DIppel; [2023-10-23]
   Nyckelord :Generative adversarial network; deep learning; echocardiography; speckle noise; denoising; segmentation;

   Sammanfattning : Previous research had shown that generative adversarial networks (GANs) are capable of despeckling echocardiograms (echos) through image-to-image translation in real-time once trained. However, only limited information regarding the quality of denoised echos and explainability of useful GAN components is provided. LÄS MER

3. 2. Comparison of Hebbian Learning and Backpropagation for Image Classification in Convolutional Neural Networks

   Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Teodor Morfeldt Gadler; [2023]
   Nyckelord :;

   Sammanfattning : Current commonly used image recognition convolutional neural networks share some similarities with the human brain. However, the differences are many and the well established backpropagation learning algorithm is not biologically plausible. LÄS MER

4. 3. Segment Routing Based Traffic Engineering : A QoS adaptive rerouting using segment routing approach based on IPv6 to mitigate network congestion

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Sepehr Javid; [2023]
   Nyckelord :Segment Routing with IPv6; Quality of Service; Congestion; Extended Berkeley Packet Filter; Segmentering av vägval med IPv6; Tjänstekvalitet; Överbelastning; Utökad Berkeley Packet Filter;

   Sammanfattning : In modern networks, the increasing volume of network traffic and the diverse range of services with varying requirements necessitate the implementation of more advanced routing decisions and traffic engineering. This academic study proposes a QoS adaptive mechanism called "Sepitto", which utilizes Segment routing protocols, specifically SRv6, to address network-traffic control and congestion avoidance. LÄS MER

5. 4. Detection of local motion artifacts and image background in laser speckle contrast imaging

   Master-uppsats, Linköpings universitet/Institutionen för medicinsk teknik

   Författare :Johannes Nyhlén; Märta Sund; [2023]
   Nyckelord :laser speckle contrast imaging; lsci; multi exposure laser speckle contrast imaging; melsci; machine learning; deep learning; statistical analysis; detection; local motion artifacts; 1D; 2D; 3D;

   Sammanfattning : Laser speckle contrast imaging (LSCI) and its extension, multi-exposure laser speckle contrast imaging (MELSCI) are non-invasive techniques to monitor peripheral blood perfusion. One of the main drawbacks of LSCI and MELSCI in clinical use is that the techniques are sensitive to tissue movement. LÄS MER

7. 5. Trainable Region of Interest Prediction: Hard Attention Framework for Hardware-Efficient Event-Based Computer Vision Neural Networks on Neuromorphic Processors

   Master-uppsats, Lunds universitet/Institutionen för elektro- och informationsteknik

   Författare :Cina Arjmand; [2023]
   Nyckelord :Artifical Intelligence; Machine Learning; Neuromorphic Engineering; Computer Vision; Technology and Engineering;

   Sammanfattning : Neuromorphic processors are a promising new type of hardware for optimizing neural network computation using biologically-inspired principles. They can effectively leverage information sparsity such as in images from event-based cameras, and are well-adapted to processing event-based data in an energy-efficient fashion. LÄS MER