{"id":990,"date":"2017-11-10T10:47:16","date_gmt":"2017-11-10T09:47:16","guid":{"rendered":"https:\/\/www.hsu-hh.de\/ant\/?page_id=990"},"modified":"2020-03-30T10:55:29","modified_gmt":"2020-03-30T08:55:29","slug":"dipl-ing-andre-kelm","status":"publish","type":"page","link":"https:\/\/www.hsu-hh.de\/ant\/mitarbeiter\/dipl-ing-andre-kelm","title":{"rendered":"Dipl.-Ing. Andr\u00e9 Kelm"},"content":{"rendered":"

 <\/p>\n

<\/div>\n
\n
\n
\n
<\/div>\n<\/div>\n
\n
\n

\"APK\"<\/p>\n<\/div>\n<\/div>\n

\n
\n
\n
\n
Raum:<\/div>\n
117<\/div>\n<\/div>\n<\/div>\n
\n
\n
Telefon:<\/div>\n
(040) 6541-3738<\/div>\n<\/div>\n<\/div>\n
\n
\n
Fax:<\/div>\n
(040) 6541-2834<\/div>\n<\/div>\n<\/div>\n
\n
\n
E-Mail:<\/div>\n
Andre.Kelm@hsu-hh.de<\/a><\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Besucheranschrift<\/div>\n
Helmut-Schmidt-Universit\u00e4t<\/div>\n
Geb\u00e4ude H3<\/div>\n
Holstenhofweg 85<\/div>\n
22043 Hamburg<\/div>\n<\/div>\n
\n
Postanschrift<\/div>\n
Helmut-Schmidt-Universit\u00e4t<\/div>\n
Fakult\u00e4t f\u00fcr Elektrotechnik<\/div>\n
Allgemeine Nachrichtentechnik<\/div>\n
Postfach 70 08 22<\/div>\n
22008 Hamburg<\/div>\n<\/div>\n<\/div>\n<\/div>\n
<\/div>\n

 <\/p>\n

 <\/p>\n

Forschung<\/h3>\n

Bildverarbeitung mit Deep Learning<\/h3>\n

Ziel ist die Segmentierung und Klassifikation von Bildern im maritimen Umfeld. Daf\u00fcr werden haupts\u00e4chlich Neuronale Faltungsnetzwerke genutzt. Neuartige Verfahren werden untersucht und gegebenenfalls angepasst oder entwickelt. Klassische Bildverarbeitungsalgorithmen werden unterst\u00fctzend f\u00fcr die Segmentierung eingesetzt.<\/p>\n

 <\/p>\n

\"FT\"
Segmentierungs-CNN<\/figcaption><\/figure>\n

 <\/p>\n

Um die Szene des Bildes zu erfassen, werden neue interessante Klassen, wie Wasser, Himmel, … und weitere gelernt.<\/p>\n

\"Ftraeg\"
Schiff-Szenen-Segmentierung-CNN<\/figcaption><\/figure>\n

 <\/p>\n

Um die Klassifikationsaufgabe zu unterst\u00fctzen, wird eine Kontur-Extraktions-Methode mit Deep Learning implementiert.<\/p>\n

\"A1\"
Kontur-Extraktions-CNN<\/figcaption><\/figure>\n

Das RefineContourNet nutzt ein etabliertes Segmentierungsnetzwerk und erreicht State-of-the-Art Edge-Detection auf dem BSDS500 Benchmark: BSDS500 SoA<\/a><\/p>\n

Test-Code verf\u00fcgbar auf https:\/\/github.com\/AndreKelm\/RefineContourNet<\/a><\/p>\n

 <\/p>\n

Betreute Projektarbeiten<\/strong><\/p>\n

Vijesh Sao Rao \u2013 Training State-of-the-Art CNNs for Segmentation and Contour Detection on Ship Images, Student Project<\/p>\n

Christina Sander \u2013 SSD Analysis and Development of Adaption Proposal for a CNN Ship Detector, Student Project<\/p>\n

Supradeep Chikaballapur Manjunath – Contour Detection using SRCNN,\u00a0 Student Project<\/p>\n

Karthik Kadur Manjunath – Analysing Hypercolumn Features for Object Contour detection and Semantic Segmentation,\u00a0 Student Project<\/p>\n

Betreute Masterarbeiten<\/strong><\/p>\n

Vijesh Sao Rao \u2013 Generating synthetic NIR images from RGB images using GANs, Master Thesis<\/p>\n

Roc\u00edo Aldana Figueroa \u2013 CNN for Detailed Ship Image Segmentation using Contour and Segmentation Feature Maps, Master Thesis<\/p>\n

Sous-Lieutenant Jean Fissot \u2013 Using Phase-Stretch-Transform Algorithm as an Image-Feature-Extractor, Master Thesis<\/p>\n

Ver\u00f6ffentlichungen<\/strong><\/p>\n

A. Kelm, V. Rao, U. Z\u00f6lzer:\u00a0Object Contour and Edge Detection with RefineContourNet<\/a> – Computer Analysis of Images and Patterns, 18th International Conference, CAIP 2019, Salerno, Italy, September 3\u20135, 2019, Proceedings, Part I<\/p>\n

P. Bhattacharya, G. Simkus, C. de Obald\u00eda, A. Kelm, U. Z\u00f6lzer: Convolutional Neural Networks for Digital Signal Processing, LSA2017 – L\u00fcbeck Summer Academy on Medical Technology, July 2017, L\u00fcbeck, Germany.<\/p>\n

 <\/p>\n","protected":false},"excerpt":{"rendered":"

  Raum: 117 Telefon: (040) 6541-3738 Fax: (040) 6541-2834 E-Mail: Andre.Kelm@hsu-hh.de Besucheranschrift Helmut-Schmidt-Universit\u00e4t Geb\u00e4ude H3 Holstenhofweg 85 22043 Hamburg Postanschrift Helmut-Schmidt-Universit\u00e4t Fakult\u00e4t f\u00fcr Elektrotechnik Allgemeine Nachrichtentechnik Postfach 70 08 22 […]<\/p>\n","protected":false},"author":142,"featured_media":0,"parent":120,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"footnotes":""},"categories":[],"tags":[],"class_list":["post-990","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/pages\/990","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/users\/142"}],"replies":[{"embeddable":true,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/comments?post=990"}],"version-history":[{"count":56,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/pages\/990\/revisions"}],"predecessor-version":[{"id":1805,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/pages\/990\/revisions\/1805"}],"up":[{"embeddable":true,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/pages\/120"}],"wp:attachment":[{"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/media?parent=990"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/categories?post=990"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.hsu-hh.de\/ant\/wp-json\/wp\/v2\/tags?post=990"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}