A Mobile Application Using AI-Based Image Recognition to Support Informal E-Waste Workers
Berko-Boateng, Anna (2025)
2025
Tietotekniikan DI-ohjelma - Master's Programme in Information Technology
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2025-12-15
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-2025121211579
https://urn.fi/URN:NBN:fi:tuni-2025121211579
Tiivistelmä
This thesis explores the use of artificial intelligence (AI) to improve occupational safety for informal electronic waste (e-waste) workers in the global South. Informal e-waste recycling is often carried out in hazardous conditions without access to safety information or protective equipment. To address this, a mobile application was designed, implemented and tested. The application allows users to take a photo of e-waste, after which an AI model analyzes the image and identifies potentially hazardous components. Based on the identified parts, the app provides safety instructions and information on harmful substances.
The development process followed user-centered design principles and was informed by a needs assessment conducted through interviews with informal e-waste workers in Ghana. The majority of the respondents were already familiar with smartphones and mobile applications, suggesting that a mobile-based solution is a feasible intervention in this context. A minimum viable product (MVP) of the application was implemented for Android devices. It uses a cloud-based AI model through an external API to analyze e-waste images, keeping the local application lightweight in terms of processing requirements.
The application was evaluated through field testing with informal e-waste workers in Abossey Okai, Ghana. Participants completed all test tasks and rated the core functionality as easy to use. The safety instructions were found to be understandable and relevant to daily work practices. Users indicated that the application would be beneficial to their occupational health and safety. The results demonstrate the potential of AI-powered mobile tools to support sustainable and inclusive innovation in the e-waste sector. They also identify areas for improvement in user interface design and risk communication.
The development process followed user-centered design principles and was informed by a needs assessment conducted through interviews with informal e-waste workers in Ghana. The majority of the respondents were already familiar with smartphones and mobile applications, suggesting that a mobile-based solution is a feasible intervention in this context. A minimum viable product (MVP) of the application was implemented for Android devices. It uses a cloud-based AI model through an external API to analyze e-waste images, keeping the local application lightweight in terms of processing requirements.
The application was evaluated through field testing with informal e-waste workers in Abossey Okai, Ghana. Participants completed all test tasks and rated the core functionality as easy to use. The safety instructions were found to be understandable and relevant to daily work practices. Users indicated that the application would be beneficial to their occupational health and safety. The results demonstrate the potential of AI-powered mobile tools to support sustainable and inclusive innovation in the e-waste sector. They also identify areas for improvement in user interface design and risk communication.