Sökning: "visual sensor"

Visar resultat 21 - 25 av 114 uppsatser innehållade orden visual sensor.

1. 21. Deep Visual Inertial-Aided Feature Extraction Network for Visual Odometry : Deep Neural Network training scheme to fuse visual and inertial information for feature extraction

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

   Författare :Franco Serra; [2022]
   Nyckelord :Feature extraction network; Visual Odometry; IMU; Neural Network; Pose estimation; Feature extraction; Visuell Odometri; IMU; Neuralt nätverk; Poseuppskattning;

   Sammanfattning : Feature extraction is an essential part of the Visual Odometry problem. In recent years, with the rise of Neural Networks, the problem has shifted from a more classical to a deep learning approach. This thesis presents a fine-tuned feature extraction network trained on pose estimation as a proxy task. LÄS MER

3. 22. Operational data extraction using visual perception

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

   Författare :Nagarajan Shunmugam; [2021]
   Nyckelord :Visual perception; camera; convolutional neural networks; classification; object detection; semantic segmentation; depth estimation; gradient descent with restarts; cosine annealing.; Visuell uppfattning; kamera; neurologiska nätverk; klassificering; objektdetektering; semantisk segmentering; djupberäkning; gradientnedstigning med omstart; cosinusglödgning.;

   Sammanfattning : The information era has led the manufacturer of trucks and logistics solution providers are inclined towards software as a service (SAAS) based solutions. With advancements in software technologies like artificial intelligence and deep learning, the domain of computer vision has achieved significant performance boosts that it competes with hardware based solutions. LÄS MER

4. 23. Efficient Recycling Of Non-Ferrous Materials Using Cross-Modal Knowledge Distillation

   Master-uppsats, Linköpings universitet/Datorseende

   Författare :Sebastian Brundin; Adam Gräns; [2021]
   Nyckelord :Recycling; Ferrous; Non-Ferrous; Knowledge Distillation; Hallucinogenic; Hyperspectral; HSI ;

   Sammanfattning : This thesis investigates the possibility of utilizing data from multiple modalities to enable an automated recycling system to separate ferrous from non-ferrous debris. The two methods sensor fusion and hallucinogenic sensor fusion were implemented in a four-step approach of deep CNNs. LÄS MER

5. 24. Benchmarking VisualInertial Odometry Filterbased Methods for Vehicles

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

   Författare :Muhammad Zahid; [2021]
   Nyckelord :visual inertial odometry; VIO; sensor fusion; state estimation; recursive filter; navigation; autonomous vehicles; visuell tröghetsodometri; VIO; sensorfusion; tillståndsuppskattning; rekursivt filter; navigering; autonoma fordon;

   Sammanfattning : Autonomous navigation has the opportunity to make roads safer and help perform search and rescue missions by reducing human error. Odometry methods are essential to allow for autonomous navigation because they estimate how the robot will move based on the available sensors. LÄS MER

7. 25. Improving visual comfort and energy efficiency in a classroom : A comparative approach of evaluating a lighting design technique and a sensor positioning method

   Master-uppsats, Umeå universitet/Institutionen för tillämpad fysik och elektronik

   Författare :Anton Sanaei; [2021]
   Nyckelord :Visual comfort; Energy efficiency; Light sensor;

   Sammanfattning : The reduction of energy demand and increased energy efficiency is believed to play a major role in tackling the global climate change. Artificial lighting systems in commercial buildings accounts for a substantial proportion of the total energy consumption. LÄS MER