AItem Recognition
Loading...
Date
2024
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In collaboration with the company ITPRO, this project aimed to extend the existing “Project AR”, which was developed by us in a school project, by developing an extension called “AItem Recognition”. The primary objective of the school project is to support communication between ITPRO technicians and installers. This is done by placing visual indicators inside of an Augmented Reality environment. The extension “AItem Recognition” empowers workers with the ability to automatically identify machine parts in real-time and access relevant information to speed up the time it takes to complete tasks. The main reason this system was developed, is so help can be offered even in absence of an ITPRO technician.
The implementation phase involved the enlargement of the already existing mobile application. A backend system was developed and deployed on a cloud platform to facilitate seamless communication between the mobile application and backend services. Additionally, another component was deployed to enable the generation of 3D models from photos, allowing users to scan an object, annotate it with a title and a description and finally upload them within the application. Afterwards those objects must be trained by the “Vuforia Target Model API” to be able to recognize them afterwards.
The outcome of the project is a fully functional mobile application that supports assembly tasks through automatic object recognition and information overlay. The backend system ensures seamless communication between the application and backend services, facilitating functionality extension and object management. From an implementation perspective, the application comprises a mobile app for field workers and a backend consisting of multiple components handling general business logic, data storage, and additional functionalities. Additionally, a web application simplifies management tasks.