Detekcija objekata u samoposlužnim aparatima

master's thesis

• Title (english): Self-service machine object detection
• Author: Barbara Koščević
• Mentor: Mislav Grgić (mentor)
• Committee member: Mislav Grgić (predsjednik povjerenstva)
• Committee member: Sonja Grgić (član povjerenstva)
• Committee member: Tomislav Kos (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-04, Croatia

Abstract

Napredak u području računalnog vida i dubokom učenju imaju svoj utjecaj u prodaji i industriji samoposlužnih aparata. Duboko učenje i računalni vid često se primjenjuju za poboljšanje ponude, optimizaciju rasporeda proizvoda na policama, praćenje zaliha, analizu kupovnih navika i raspoznavanje proizvoda. U ovom radu predstavljamo sustav za detekciju objekata u samoposlužnim aparatima koji se sastoji od četiri koraka: detekcija samoposlužnih aparata, detekcija proizvoda, klasifikacija proizvoda i verifikacija planograma. Kroz ispitivanje sustava, otkrivamo da kvaliteta ulazne slike ima veliki utjecaj na točnost rada sustava. Poboljšanja u točnosti modela mogu pružiti velike prednosti ovog sustava u industriji samoposlužnih aparata, smanjujući ljudske pogreške i poboljšavajući korisničko iskustvo.

Abstract (english)

Advancements in computer vision and deep learning have made a significant impact on the sales and self-service machine industry. Deep learning and computer vision are frequently used to enhance product offerings, optimize shelf placement, monitor inventory, analyze purchasing patterns, and enable product recognition. In this study, we present a system for object detection in self-service machines, comprising four steps: self-service machine detection, product detection, product classification, and planogram verification. Through system testing, we reveal that the quality of input images significantly influences the system's accuracy. Improving model accuracy can provide substantial benefits to the self-service machine industry by reducing human errors and enhancing user experience.

• Keywords: računalni vid; duboko učenje; detekcija objekata; samoposlužni aparat; prodaja; planogram; skupovi podataka
• Keywords (english): computer vision; deep learning; object detection; self-service machine; sales industry; planogram; datasets
• Language: croatian
• URN:NBN: urn:nbn:hr:168:323156
• 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: 2023-07-13 10:12:55