Abstract:
For a person with normal vision, recognition of paper currency is an easy task, but this is not the
case for a visually impaired person as the latter faces a lot of difficulties in their day-to-day
involvements with monetary transactions. They not only have difficulty in recognizing the paper
currencies due to the similarity of paper texture, but also the size between different categories of
currency notes. Financial institutions like banks can afford expensive hardware such as
Automatic Teller Machines (ATMs), automatic banknote sorters, to resolve the issue of currency
recognition, but for common people, especially the visually impaired persons, accessing such
expensive hardware is a daunting challenge. The aim of the research was to provide visually
impaired persons with a cost-effective android application solution to detect currencies and
objectives to establish the effectiveness of the current currency detector applications, challenges
faced by users of current currency detectors, and develop a viable currency detector for now and
future generations. The study centered on currency recognition software that helps distinguish
different currency notes. Development techniques utilized incorporated image foreground
segmentation, histogram enhancement, area of interest (ROI) extraction, and template matching
primarily based on the cross-correlation among the captured picture and the records set. The
system will reduce cases of visually impaired persons being coned and limit transactions
involving fake currencies, while acting as a benchmarking tool for emerging research and
discoveries.
