Duboko učenje za detekciju pisanih matematičkih izraza korištenjem mobilnih uređaja

master's thesis

• Title (english): Deep Learning for Detection of Handwritten Mathematical Expressions Using Mobile Devices
• Author: Luka Zuanović
• Mentor: Sven Lončarić (mentor)
• Committee member: Sven Lončarić (predsjednik povjerenstva)
• Committee member: Marko Subašić (član povjerenstva)
• Committee member: Zvonko Kostanjčar (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-09-14, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Detekcija objekata je postupak kojim se na slici pronalaze područja koja predstavljaju potencijalne objekte, dok je klasifikacija postupak u kojem se tim područjima dodijeli neki od mogućih razreda objekata. Konvolucijske neuronske mreže pokazale su se kao dobar izbor za izlučivanje značajki za klasifikaciju. Faster R-CNN model objedinjuje postupak detekcije i klasifikacije objekata na slici te je pokazao zavidne rezultate u tom području. Korištenjem Faster R-CNN modela izgrađen je sustav za detekciju pisanih matematičkih izraza te su raspravljeni rezultati detekcije.

Abstract (english)

Object detection is a method in which regions that represent potential objects are extracted from an image, while classification is a method in which one of the potential object classes is assigned to those regions. Convolutional neural networks have been shown as a good choice for extracting features for the classification task. Faster R-CNN model unifies object detection and classification on an image and has demonstrated exceptional results in that filed. Using the Faster R-CNN model, a system for detection of handwritten mathematical expressions has been built and the detection results are discussed.

• Keywords: detekcija objekata; konvolucijska neuronska mreža; Faster R-CNN
• Keywords (english): object detection; convolutional neural network; Faster R-CNN
• Language: croatian
• URN:NBN: urn:nbn:hr:168:626505
• 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: 2019-03-14 02:21:14