Determination Of Rolling Returns From A Rasterized Graph Using Image Processing

• Main Author: Chew, Jia Hao
• Format: Final Year Project / Dissertation / Thesis
• Published: 2020
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utar.edu.my/4038/1/3E_1502778_FYP_report_%2D_JIA_HAO_CHEW.pdf; http://eprints.utar.edu.my/4038/

This project is to develop a software application that is able to extract raw data from a rasterized performance graph of a mutual fund using image processing. The extracted raw data are the percentage return of mutual fund with the date. The software is able to calculate rolling return of mutual fund using the extracted raw data. This software application will make investor’s work becomes easier and also provides them better insight into the mutual fund performance. The main programming language used in this project is Python. OpenCV and Tkinter are used in this program as image processing library and graphical user interface library respectively. Users will get raw data in an excel file and graph of rolling return of mutual fund after users insert the graph image and some inputs into the software application. Validation of results was carried out after every simulation during developing the program. The average percentage difference obtained by comparing actual value with calculated value is 1.84% for one-year, 2.09% for three-year return and 2.22% for fiveyear return. There are more works and efforts needed on improving the accuracy of results while reducing the number of inputs provided by users so that the software application will become more efficient and effective.