Uočavanje i izlučivanje rubova puteva u sekvencijalnim slikama

master's thesis

• Title (english): Detection and extraction of path edges in sequential images
• Author: Mario Pavečić
• Mentor: Ivan Petrović (mentor)
• Committee member: Ivan Petrović (predsjednik povjerenstva)
• Committee member: Marija Seder (član povjerenstva)
• Committee member: Ivan Marković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Autonomno upravljanje gibanjem vozila je kompleksni problem koji ovisi o zadatku. Uz pretpostavku da se vozilo kreće rubom kolnika, određivanje putanje vozila moguće je svesti na izlučivanje rubova istoga. U tom slučaju potrebno je iz slike kamere, postavljene na prednji dio vozila, izlučiti rubove ceste kako bi se informacija o tim rubovima mogla iskoristiti za računanje naredbi upravljanja vozilom.
U ovom radu predstavljena su dva algoritma izlučivanja rubova ceste, prvi se temelji na Cannyjevom detektoru rubova, a drugi na semantičkoj segmentaciji. Prikazane su implementacije oba algoritma zajedno sa rezultatima svakog koraka algoritma. U zadnjem dijelu rada, razmotrena su moguća poboljšanja opisanih algoritama uvođenjem dodatnih informacija.

Abstract (english)

Autonomous vehicle motion control is a complex problem that depends on the task. With assumption that vehicle is moving along the edge of the road, determining direction of the vehicle at given moment can be reduced to exctrating road edge. In that case, it is necessary to determine road edges from camera image frames on the front of the vehicle, so they could be used as information in controlling vehicle movement.
This master thesis presents two different algorithms for detecting road edges. First one is based on Canny edge detection and the second on semantic segmentation. Both implementations are shown together along with results of each step in both algorithms. Last part of this master thesis considers possible improvements on both algorithms with taking in consideration additional resources.

• Keywords: semantička segmentacija; rubovi ceste; poluautomatska vozila; gradijent slike; Cannyjev detektor rubova
• Keywords (english): semantic segmentation; road edges; semi-automatic vehicles; image gradient; Canny edge detection
• Language: english
• URN:NBN: urn:nbn:hr:168:544816
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-15 10:58:44