izpis_h1_title_alt

Podobnostna mreža slik
ID Hartman, Jan (Author), ID Zupan, Blaž (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (3,86 MB)
MD5: DD083B227DF6A90BB36B74DC1869058F

Abstract
V diplomskem delu smo razvili novo orodje za interaktivno pregledovanje slik. Orodje za izboljšan prikaz slik kombinira uporabo globokega učenja z drugimi tehnikami podatkovne analitike. Predlagana rešitev je sestavljena iz več zaporedno povezanih metod. Prvi korak je obdelava slik z uporabo konvolucijskih nevronskih mrež za pridobitev značilk, s katerimi lahko slike med seboj primerjamo. Nato z optimizacijskimi algoritmi slike razporedimo v dvodimenzionalno pravokotno podobnostno mrežo, v kateri so podobne slike postavljene bližje skupaj. Rezultat naloge je interaktivni pregledovalnik slik, oblikovan kot podobnostna mreža. Implementiran je v programskem jeziku Python in vključen v orodje Orange kot gradnik z imenom Image Grid, kjer se enostavno povezuje z ostalimi gradniki v dodatku Image Analytics.

Language:Slovenian
Keywords:podobnostna mreža slik, interaktivni pregledovalnik slik, konvolucijske nevronske mreže, vložitve slik, zmanjševanje dimenzij, Orange
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2018
PID:20.500.12556/RUL-102408 This link opens in a new window
Publication date in RUL:28.08.2018
Views:1001
Downloads:333
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Image Similarity Grid
Abstract:
The thesis describes the development of an interactive image visualization tool. To improve the display of images, the tool combines the use of deep learning with other data analytics techniques. The proposed solution consists of several sequentially connected methods. The first step is using convolutional neural networks for image processing to acquire features which can be used for image comparison. Then, optimization algorithms are used to place images into a two-dimensional rectangular similarity grid where similar images are placed closer together. The result of the thesis is an interactive image viewer, designed as a similarity grid. It is implemented in the Python programming language and included in the Orange toolbox as a widget named Image Grid, where it is seamlessly integrated with other widgets in the Image Analytics add-on.

Keywords:image similarity grid, interactive image viewer, convolutional neural networks, image embeddings, dimensionality reduction, Orange

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back