Bearbeitung der Übungsaufgaben des Kurses Maschinelles Sehen bei Dr.-Ing. Anko Börner an der Beuth Hochschule für Technik Berlin.
Die Projekte sind in C++11 (soweit von meinem verwendeten Compiler unterstützt) mittels OpenMP bearbeitet. OpenMP-Unterstützung ist nicht zwingend notwendig, aber empfohlen; Ebenso ist es nicht ratsam, die Beispiele in DEBUG-Konfiguration zu starten.
Die Projekte verwenden OpenCV~2.4.1 für die Darstellung der Ergebnisse und verwendet die Libraries OpenCV Core und OpenCV HighGUI (mas01 zusätzlich OpenCV Image Processing.)
mas01 is about the general use of OpenCV for loading and storing images, as well as basic image transformations through pixel manipulation.
mas02 is about dynamic memory management and simple statistics (minimum and maximum, mean, standard deviation) of an image. This project also covers radiometric transformations in the context of high dynamic range imaging.
mas03 is about feature detection by using image/template cross-correlation and absolute differences.
mas04 is about image filtering by using high- and lowpass kernels, as well as a median filter.
Copyright © 2013 Markus Mayer
State-Space Sandbox is licensed under the EUPL, Version 1.1 or - as soon they will be approved by the European Commission - subsequent versions of the EUPL (the "Licence"); you may not use this work except in compliance with the Licence. You may obtain a copy of the Licence at:
http://joinup.ec.europa.eu/software/page/eupl/licence-eupl
Unless required by applicable law or agreed to in writing, software distributed under the Licence is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the Licence for the specific language governing permissions and limitations under the Licence.
I do not claim copyright on any of the images used in these examples; They are used here solely on an scientific fair use basis (wissenschaftliches Zitat bzw. Kleinzitat gemäß §51 UrhG). Concerning the images bild0.raw to bild1.raw and mask_32_32.raw, as well as rued_corr_flt.img, consider them to be copyrighted by Dr.-Ing. Anko Börner. lena.jpg and lena.raw is our good ol' first lady o' the interwebs, Lena.
The image lenaml.raw is a histogram adjusted 512x512 pixel greyscale crop from an image of Lena Meyer-Landrut copyrighted by Sandra Luedwig (source), again used under a scientific fair use (wissenschaftliches Zitat bzw. Kleinzitat gemäß §51 UrhG).