Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method

This paper presents the retrieval method of geological images or their fragments using Query by Sketch method. The sketch can be created manually, for instance using a graphics editor, and may show the shape of objects or their distribution within an image. This query is then used to search the imag...

Full description

Bibliographic Details
Main Authors: Magdalena Habrat, Mariusz Młynarczuk
Format: Article
Language:English
Published: MDPI AG 2019-12-01
Series:Applied Sciences
Subjects:
Online Access:https://www.mdpi.com/2076-3417/10/1/278
id doaj-8d2015aa8b5c418280ba4eb7c301eb06
record_format Article
spelling doaj-8d2015aa8b5c418280ba4eb7c301eb062020-11-25T02:21:16ZengMDPI AGApplied Sciences2076-34172019-12-0110127810.3390/app10010278app10010278Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch MethodMagdalena Habrat0Mariusz Młynarczuk1Faculty of Geology, Geophysics and Environmental Protection, AGH University of Science and Technology, 30-059 Kraków, PolandFaculty of Geology, Geophysics and Environmental Protection, AGH University of Science and Technology, 30-059 Kraków, PolandThis paper presents the retrieval method of geological images or their fragments using Query by Sketch method. The sketch can be created manually, for instance using a graphics editor, and may show the shape of objects or their distribution within an image. This query is then used to search the image database for objects showing the greatest similarity. As an example of the proposed method, the detection of porosity in microscopic images of carbonate rock and sandstone was presented. An approach was described which is founded on the designation of parameters of selected properties of the query image and images in databases, as well as on the conformity analysis of these parameters. Two methods were proposed: the first one searches for the most similar object in the image database with respect to the set criteria. The second method performs a search based on a sketch of images which are similar in terms of object distribution (i.e., porosity). The presented research results confirm that database search using the query by sketch method forms an interesting and modern approach and may constitute one of the functionalities of IT systems intended for use in geology and mining industry.https://www.mdpi.com/2076-3417/10/1/278rock analysisreservoir rocks featuresquery by sketchimage analysiscomputer graphics
collection DOAJ
language English
format Article
sources DOAJ
author Magdalena Habrat
Mariusz Młynarczuk
spellingShingle Magdalena Habrat
Mariusz Młynarczuk
Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
Applied Sciences
rock analysis
reservoir rocks features
query by sketch
image analysis
computer graphics
author_facet Magdalena Habrat
Mariusz Młynarczuk
author_sort Magdalena Habrat
title Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
title_short Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
title_full Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
title_fullStr Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
title_full_unstemmed Object Retrieval in Microscopic Images of Rocks Using the Query by Sketch Method
title_sort object retrieval in microscopic images of rocks using the query by sketch method
publisher MDPI AG
series Applied Sciences
issn 2076-3417
publishDate 2019-12-01
description This paper presents the retrieval method of geological images or their fragments using Query by Sketch method. The sketch can be created manually, for instance using a graphics editor, and may show the shape of objects or their distribution within an image. This query is then used to search the image database for objects showing the greatest similarity. As an example of the proposed method, the detection of porosity in microscopic images of carbonate rock and sandstone was presented. An approach was described which is founded on the designation of parameters of selected properties of the query image and images in databases, as well as on the conformity analysis of these parameters. Two methods were proposed: the first one searches for the most similar object in the image database with respect to the set criteria. The second method performs a search based on a sketch of images which are similar in terms of object distribution (i.e., porosity). The presented research results confirm that database search using the query by sketch method forms an interesting and modern approach and may constitute one of the functionalities of IT systems intended for use in geology and mining industry.
topic rock analysis
reservoir rocks features
query by sketch
image analysis
computer graphics
url https://www.mdpi.com/2076-3417/10/1/278
work_keys_str_mv AT magdalenahabrat objectretrievalinmicroscopicimagesofrocksusingthequerybysketchmethod
AT mariuszmłynarczuk objectretrievalinmicroscopicimagesofrocksusingthequerybysketchmethod
_version_ 1724867265240760320