Vgrajen sistem za razpoznavo bakterijskih kolonij na osnovi strojnega vida

magistrsko delo

Andraž Vergles (Author), Iztok Kramberger (Mentor)

Abstract

Magistrsko delo opisuje zasnovo prototipnega vgrajenega sistema za avtomatsko prepoznavo bakterijskih kolonij v petrijevkah. Razvoj je potekal s pomočjo razvojne plošče, s katero smo zajemali vzorčne slike. Algoritme za razpoznavo smo najprej razvili na osebnem računalniku, nato pa jih prenesli na vgrajen sistem. Pomagali smo si z odprtokodnimi knjižnicami, ki imajo večino potrebnih algoritmov za obdelavo slik že implementiranih. Pri prenosu pa je bilo te algoritme potrebno implementirati v obliki, primerni za izvajanje na vgrajenem sistemu. Da smo dobili slike, primerljive s tistimi, kot bi jih v inkubatorju, smo izdelali maketo z osvetlitvijo. Izmerili smo nekatere optične parametre kamere — popačenje leče in modulacijsko prenosno funkcijo

Keywords

vgrajen sistemi;strojni vid;kamera;petrijevka;štetje kolonij;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2023
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[A. Vergles]
UDC:	004.932.031.6(043.2)
COBISS:	169958659
Views:	48
Downloads:	5
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Embedded system for recognizing bacterial colonies based on machine vision
Secondary abstract:	The master’s thesis describes the design of a prototype embedded system for the automatic recognition of bacterial colonies in Petri dishes. Development took place with the help of a development board, which was used to capture sample images. We first developed the recog- nition algorithms on a personal computer, and then transfer them to the embedded system. We used open source libraries, which have most of the necessary image processing algorithms al- ready implemented. When porting, however, it was necessary to implement these algorithms in a suitable form, for execution on the embedded system. To obtain images comparable to those as they would be in an incubator, we made a model with lighting. We measured some optical parameters of the camera—lens distortion and modulation transfer function.
Secondary keywords:	embedded system;machine vision;camera;petri dish;colony counting;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Elektrotehnika
Pages:	1 spletni vir (1 datoteka PDF (IX, 62 f.))
ID:	19858761