Primjena HALCON sučelja za robotizirano sortiranje predmeta

master's thesis

• Title (english): Application of the HALCON interface for robotic item sorting
• Author: Josip Kozumplik
• Mentor: Petar Ćurković (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2023-12-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

U ovom radu uvodno je objašnjen koncept strojnog vida, grane računalnog vida koja strojevima omogućuje interpretaciju i razumijevanje vizualnih informacija iz stvarnog svijeta temeljem obrade podataka dobivenih iz slike. Potom su navedene i objašnjene temeljne komponente sustava strojnog vida.
Glavna tema ovog rada je primjena HALCON programskog paketa za robotizirano sortiranje predmeta. Detaljno su prikazane i objašnjene metode HALCON-a koje su korištene pri rješavanju diplomskog zadatka, a to su metode: segmentacije slike, podudaranja oblika, dimenzijske analize te je objašnjen modul za duboko učenje.
Rješavanje zadatka ovog diplomskog rada prikazano je kroz tri primjera. Prvi primjer prikazuje slaganje pločica sa segmentima slova iz neuređenog stanja. Cilj ovog primjera bio je razviti algoritam strojnog vida u HALCON-u pomoću kojeg se prepoznaje, odnosno podudara određeni segment slova sa referentnim segmentom na svakoj od osamnaest pločica. Potom se svaka pločica mjeri kako bi se odredilo njezino ishodište i orijentacija. Dobivene informacije o položaju i orijentaciji šalju se robotskoj ruci kako bi vakuumskom hvataljkom mogla izuzeti određenu pločicu te je odložiti na za to predviđeno mjesto.
Drugi primjer se sastoji od sortiranja plastike ravne površine prema boji. Ideja je bila razviti algoritam prepoznavanja boje otpadne plastike te njezino sortiranje robotskom rukom pomoću vakuumske hvataljke.
Treći primjer koji je prikazan u ovom diplomskom radu sastoji se od sortiranja plastike nepravilnog oblika prema boji. Cilj ovog primjera isti je kao i prethodnog, razviti algoritam prepoznavanja boje otpadne plastike, no u ovom se slučaju za izuzimanje plastike koristila univerzalna hvataljka. Postupak izrade univerzalne hvataljke prikazan je u zadnjem poglavlju ovog rada.
Svi primjeri koji su prikazani u ovom radu realizirani su u proizvodnom pogonu tvrtke ASEMO d.o.o. koja je također omogućila i svu potrebnu opremu za uspješno rješavanje zadatka.
Plastiku koja je korištena pri razvoju algoritama prepoznavanja boje ustupila je na korištenje tvrtka Pos-Plast d.o.o.

Abstract (english)

This master thesis introduces the concept of machine vision, a branch of computer vision that enables machines to interpret and understand visual information from the real world based on the processing of data obtained from images. The basic components of the machine vision system are then listed and explained.
The main topic of this work is the application of the HALCON program package for robotic item sorting. The HALCON methods that were used in the solution of master thesis task were presented and explained in detail, namely: image segmentation, shape matching, dimensional analysis, and deep learning module was explained.
Solving the task of this thesis is presented through three examples. The first example shows palletization of tiles with letter segments from an unordered state. The goal of this example was to develop a machine vision algorithm in HALCON, by means of which a certain segment of letters is recognized, that is, matched with a reference segment on each of the eighteen tiles. Each tile is then measured to determine its origin and orientation. The obtained information about the position and orientation is sent to the robotic arm so that it can pick a specific tile with a vacuum gripper and place it in the designated place.
The second example consists of sorting flat plastic by color. The idea was to develop an algorithm for recognizing the color of waste plastic and sorting it with a robotic arm using a vacuum gripper.
The third example presented in this thesis consists of sorting plastic of irregular shape according to color. The goal of this example is the same as the previous one, to develop an algorithm for recognizing the color of waste plastic, but in this case, a universal gripper was used to pick the plastic. The process of making a universal gripper is presented in the last chapter of this thesis.
All the examples presented in this thesis were realized in the production plant of the company ASEMO d.o.o. which also provided all the necessary equipment to successfully solve the task.
The plastic that was used in the development of color recognition algorithms was provided for use by the company Pos-Plast d.o.o.

• Keywords: strojni vid; kamera; MVTec HALCON; Fanuc Lr Mate 200iD; robotska ruka; segmentacija slike; podudaranje oblika; dimenzijska analiza; vakuumska hvataljka; univerzalna hvataljka; sortiranje
• Keywords (english): machine vision; camera; MVTec HALCON; Fanuc Lr Mate 200iD; robotic arm; image segmentation; shape matching; dimensional analysis; vacuum gripper; universal gripper; sorting
• Language: croatian
• URN:NBN: urn:nbn:hr:235:369113
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Embargoed access; Embargo expiration date: 2025-11-30
• Created on: 2023-11-30 13:06:43