38th Annual Student Research Conference
Automated Mold Growth Assessment: ]Buidling a Kivy-Based Mobile Application for Accurate Fungal Analysis
Mohammed Ayan Mahmood
Dr. Kafi Rahman and Dr. Hajeewaka Mendis, Faculty Mentors
This project aims to develop a Kivy-based mobile application to automate mold (including Zygomycota and Ascomycota fungi) growth assessment on agar plates, addressing limitations in manual methods like measuring colony diameter or using grids. Existing tools such as ImageJ require complex setups, limiting accessibility.
The proposed app uses computer vision algorithms to capture images, convert them to grayscale, and apply thresholding techniques to distinguish mold colonies from the background. Plate dimensions can be input manually or selected from ISO-standard sizes, ensuring precision. To reduce parallax error, the app leverages the device’s accelerometer and gyroscope. A Raspberry Pi setup enables continuous monitoring, sending notifications when colonies reach user-defined growth stages. Data is stored on a MySQL server, supporting session-based analysis and generating reports in PDF/CSV formats.
The app shall offer an accessible, automated measurement method, and will be accessible on Google Playstore/Apple Appstore for free, intending to revolutionize microbiological research.
Keywords: Zygomycota, Ascomycota, Fungal Growth Assessment, Computer Vision, Mobile Application, Microbiological Research, Automated Measurement, Session-Based Analysis
Topic(s):Biochemistry and Molecular Biology
Computer Science
Presentation Type: Oral Presentation
Session: 208-3
Location: MG 1000
Time: 10:45