Abstract
Though coins are less used in the informational era, a significant amount of the population still uses physical coins for financial transactions. Also, when the countrys government collects the coins for equivalent digital currency transactions, coin identification is still vital yet tedious. Moreover, the dated coins and the difference in illuminations. This novel presents a coin identification and validation model based on the Alex Net convolutional model. It identifies the value of a coin through the numbers, animals, and plants on the two sides. The model employs data enhancement, feature attention layers, max pooling, and residue groups. We collected 2826 images of Brazilian coins with reverse motifs, and the experimental accuracy of our model reached 0.97. The code part has shown here: https://github.com/Erik-Xie/-Brazilian-Coin-counter-Research-Report.git
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
