Variable Illumination and Invariant Features for Detecting and Classifying Varnish Defects

Variable Illumination and Invariant Features for Detecting and Classifying Varnish Defects
Author: Ana Pérez Grassi
Publisher: KIT Scientific Publishing
Total Pages: 170
Release: 2014-08-14
Genre: Technology (General)
ISBN: 3866445377

This work presents a method to detect and classify varnish defects on wood surfaces. Since these defects are only partially visible under certain illumination directions, one image doesn't provide enough information for a recognition task. A classification requires inspecting the surface under different illumination directions, which results in image series. The information is distributed along this series and can be extracted by merging the knowledge about the defect shape and light direction.

Proceedings of the 2010 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory

Proceedings of the 2010 Joint Workshop of Fraunhofer IOSB and Institute for Anthropomatics, Vision and Fusion Laboratory
Author: Jürgen Beyerer
Publisher: KIT Scientific Publishing
Total Pages: 254
Release: 2014-08-18
Genre: Computers
ISBN: 3866446098

On the annual Joint Workshop of the Fraunhofer IOSB and the Karlsruhe Institute of Technology (KIT), Vision and Fusion Laboratory, the students of both institutions present their latest research findings on image processing, visual inspection, pattern recognition, tracking, SLAM, information fusion, non-myopic planning, world modeling, security in surveillance, interoperability, and human-computer interaction. This book is a collection of 16 reviewed technical reports of the 2010 Joint Workshop.

Machine Vision

Machine Vision
Author: Jürgen Beyerer
Publisher: Springer
Total Pages: 802
Release: 2015-10-01
Genre: Technology & Engineering
ISBN: 3662477947

The book offers a thorough introduction to machine vision. It is organized in two parts. The first part covers the image acquisition, which is the crucial component of most automated visual inspection systems. All important methods are described in great detail and are presented with a reasoned structure. The second part deals with the modeling and processing of image signals and pays particular regard to methods, which are relevant for automated visual inspection.

Probabilistic Models and Inference for Multi-View People Detection in Overlapping Depth Images

Probabilistic Models and Inference for Multi-View People Detection in Overlapping Depth Images
Author: Wetzel, Johannes
Publisher: KIT Scientific Publishing
Total Pages: 204
Release: 2022-07-12
Genre: Technology & Engineering
ISBN: 3731511770

In this work, the task of wide-area indoor people detection in a network of depth sensors is examined. In particular, we investigate how the redundant and complementary multi-view information, including the temporal context, can be jointly leveraged to improve the detection performance. We recast the problem of multi-view people detection in overlapping depth images as an inverse problem and present a generative probabilistic framework to jointly exploit the temporal multi-view image evidence.

Light Field Imaging for Deflectometry

Light Field Imaging for Deflectometry
Author: Uhlig, David
Publisher: KIT Scientific Publishing
Total Pages: 284
Release: 2023-07-14
Genre:
ISBN: 3731513064

Optical measurement methods are becoming increasingly important for high-precision production of components and quality assurance. The increasing demand can be met by modern imaging systems with advanced optics, such as light field cameras. This work explores their use in the deflectometric measurement of specular surfaces. It presents improvements in phase unwrapping and calibration techniques, enabling high surface reconstruction accuracies using only a single monocular light field camera.

Reconstruction from Spatio-Spectrally Coded Multispectral Light Fields

Reconstruction from Spatio-Spectrally Coded Multispectral Light Fields
Author: Schambach, Maximilian
Publisher: KIT Scientific Publishing
Total Pages: 238
Release: 2022-10-17
Genre: Technology & Engineering
ISBN: 3731512106

In dieser Arbeit werden spektral kodierte multispektrale Lichtfelder untersucht, wie sie von einer Lichtfeldkamera mit einem spektral kodierten Mikrolinsenarray aufgenommen werden. Für die Rekonstruktion der kodierten Lichtfelder werden zwei Methoden entwickelt, eine basierend auf den Prinzipien des Compressed Sensing sowie eine Deep Learning Methode. Anhand neuartiger synthetischer und realer Datensätze werden die vorgeschlagenen Rekonstruktionsansätze im Detail evaluiert. -In this work, spatio-spectrally coded multispectral light fields, as taken by a light field camera with a spectrally coded microlens array, are investigated. For the reconstruction of the coded light fields, two methods, one based on the principles of compressed sensing and one deep learning approach, are developed. Using novel synthetic as well as a real-world datasets, the proposed reconstruction approaches are evaluated in detail.

Computational, label, and data efficiency in deep learning for sparse 3D data

Computational, label, and data efficiency in deep learning for sparse 3D data
Author: Li, Lanxiao
Publisher: KIT Scientific Publishing
Total Pages: 256
Release: 2024-05-13
Genre:
ISBN: 3731513463

Deep learning is widely applied to sparse 3D data to perform challenging tasks, e.g., 3D object detection and semantic segmentation. However, the high performance of deep learning comes with high costs, including computational costs and the effort to capture and label data. This work investigates and improves the efficiency of deep learning for sparse 3D data to overcome the obstacles to the further development of this technology.

Fusion of Data from Heterogeneous Sensors with Distributed Fields of View and Situation Evaluation for Advanced Driver Assistance Systems

Fusion of Data from Heterogeneous Sensors with Distributed Fields of View and Situation Evaluation for Advanced Driver Assistance Systems
Author: Otto, Carola
Publisher: KIT Scientific Publishing
Total Pages: 262
Release: 2014-07-29
Genre: Technology & Engineering
ISBN: 3731500736

In order to develop a driver assistance system for pedestrian protection, pedestrians in the environment of a truck are detected by radars and a camera and are tracked across distributed fields of view using a Joint Integrated Probabilistic Data Association filter. A robust approach for prediction of the system vehicles trajectory is presented. It serves the computation of a probabilistic collision risk based on reachable sets where different sources of uncertainty are taken into account.

Characterization and Emulation of Low-Voltage Power Line Channels for Narrowband and Broadband Communication

Characterization and Emulation of Low-Voltage Power Line Channels for Narrowband and Broadband Communication
Author: Han, Bin
Publisher: KIT Scientific Publishing
Total Pages: 194
Release: 2017-06-06
Genre: Technology (General)
ISBN: 3731506548

The demand for smart grid and smart home applications has raised the recent interest in power line communication (PLC) technologies, and has driven a broad set of deep surveys in low-voltage (LV) power line channels. This book proposes a set of novel approaches, to characterize and to emulate LV power line channels in the frequency range from0.15to 10 MHz, which closes gaps between the traditional narrowband (up to 500 kHz) and broadband (above1.8 MHz) ranges.

Hyperspectral Image Unmixing Incorporating Adjacency Information

Hyperspectral Image Unmixing Incorporating Adjacency Information
Author: Bauer, Sebastian
Publisher: KIT Scientific Publishing
Total Pages: 236
Release: 2018-07-18
Genre: Technology & Engineering
ISBN: 3731507889

While the spectral information contained in hyperspectral images is rich, the spatial resolution of such images is in many cases very low. Many pixel spectra are mixtures of pure materials' spectra and therefore need to be decomposed into their constituents. This work investigates new decomposition methods taking into account spectral, spatial and global 3D adjacency information. This allows for faster and more accurate decomposition results.