Vrednovanje postupka semantičke segmentacije temeljenog na slučajnim šumama

undergraduate thesis

• Title (english): Experimental evaluation of semantic segmentation by random forests
• Author: Ivan Fabijanić
• Mentor: Siniša Šegvić (mentor)
• Committee member: Siniša Šegvić (predsjednik povjerenstva)
• Committee member: Mladen Karan (član povjerenstva)
• Committee member: Goran Glavaš (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Rad razmatra i evaluira postupak semantičke segmentacije slika temeljene na slučajnim šumama. Objašnjeni su koncepti stabla odlučivanja i slučajne šume te način rada semantičke segmentacije slika. Evaluirana je metoda segmentacije temeljena na slučajnim šumama te je unakrsnom evaluacijom određena preciznost detekcije u ovisnosti o promjeni parametara poput broja stabala, veličine prozora i dubine stabala. Za evaluaciju metode korišten je skup slika „KITTI road“ koji je objavio Karlsruhe Institute of Tehnology i koji se koristi kao referenca za kvalitetu algoritama namijenjenih za detekciju ceste. Također je paralelno sa radom razvijen i softver za označavanje razreda slika koji se može koristiti za buduća istraživanja povezana sa semantičkom segmentacijom.

Abstract (english)

This paper examines and evaluates algorithm for semantic image segmentation by random forests. Concepts of decision trees and random forest have been explained along with the semantic image segmentation algorithm. Semantic segmentation by random forests precision has been evaluated and cross-validated with changing parameters such as number of decision trees, window size and tree depth. Picture set that has been used for evaluation is „KITTI road“ which was published by Karlsruhe Institute of Tehnology and which is used as a reference for detecting quality of road detecting algorithms. New software for image labeling has also been developed alongside with this paper which can be used for future semantic segmentation scientific papers.

• Keywords: semantička segmentacija; slučajne šume; stablo odluke; integralna slika; vrednovanje postupka semantičke segmentacije; OpenCV
• Keywords (english): semantic segmentation; random forests; decision tree; integral image; experimental evaluation of semantic segmentation; OpenCV
• Language: croatian
• URN:NBN: urn:nbn:hr:168:699022
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 08:41:26