Machine vision algorithms in IT business
Bogomolov, Denis (2022)
2022
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-2022121328427
https://urn.fi/URN:NBN:fi:amk-2022121328427
Tiivistelmä
The purpose of this applied research was to develop machine vision algorithm and image processing model for specific mobile testing software. Firstly, modern AI vision technologies were reviewed and used in three development steps, and developed cropping algorithm based on OpenCV library. Afterwards, training custom vision model was presented and ran into mobile Kotlin application.
The application presented developed machine vision and cropping algorithm which might be embedded as corporate product of the company. Algorithms improve accuracy and quality of mobile diagnostic services.
The research was initiated by Piceasoft Oy company which develops mobile testing software. Company is moving towards smart and high-precision testing methods including machine vision. As a result, new AI algorithm was developed to prove its efficiency. This thesis work can be used by students for learning AI technologies. Source code written by company employees is classified except code presented in the research.
The application presented developed machine vision and cropping algorithm which might be embedded as corporate product of the company. Algorithms improve accuracy and quality of mobile diagnostic services.
The research was initiated by Piceasoft Oy company which develops mobile testing software. Company is moving towards smart and high-precision testing methods including machine vision. As a result, new AI algorithm was developed to prove its efficiency. This thesis work can be used by students for learning AI technologies. Source code written by company employees is classified except code presented in the research.