Brazilian Banknote Recognition based on CNN for Blind People
The most used payment method worldwide is through banknotes and it still constitutes an essential part in monetary transactions. Banknotes recognition is a non-trivial task for visually impaired and blind people. The mentioned difficulty makes people with visual impairment vulnerable to fraud. Moreover, it creates a barrier to daily consumption activities, like purchases and other banking tasks.
Studies carried out in 2015 demonstrated that there were an estimated 36 million blind people globally. Meanwhile, moderate and severe vision impairment affected 216.6 million people. According to the Brazilian National Health Survey in 2019, 6.978 million Brazilian people have severe vision impairment or blindness.
Banking institutions around the world, responsible for the banknotes manufacture, provide some strategies to facilitate banknote recognition by the visually impaired, such as: different sizes, coded tactile structure, braille markings, irregular edges or tangible features. However, the tactile analysis represents a considerable difficulty due to intense banknotes turnover, promoting damages in their original state resulting in incorrect identification. Figure 1 presents an example of Brazilian banknote damage, regarding tactile analysis.
Figure 1: Example of Brazilian banknote with damage in tactile marks.
This paper presents an approach to recognizing Brazilian banknotes and assisting people with visual impairment in financial operations. The banknote denomination (value) is represented and classified through the proposed CNN architecture, with a specific learning process for the banknote domain. Experiments in real-world scenarios demonstrate the proposed approach scalability and show that the obtained results are accurate and reliable.
Our main contribution is to provide a robust and efficient approach, based on Deep Learning, to classify Brazilian banknotes, assisting the visually impaired in daily monetary transactions. Based on the proposed approach it is possible to embed the learning model into a mobile application to support visually impaired people, in monetary operations. We also highlight the dataset used in this research work, consisting of Brazilian banknotes in different environments and luminosity conditions. Furthermore, it is important to mention the variety of images, including crumpled banknotes and real acquisition scenarios.
NETO, ODALISIO ; OLIVEIRA, FELIPE ; CAVALCANTI, JOÃO ; PIO, JOSÉ . Brazilian Banknote Recognition Based on CNN for Blind People. In: 18th International Conference on Computer Vision Theory and Applications, 2023, Lisbon. Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications, 2023. p. 846.
This work was developed with support from the Motorola, through the IMPACT-Lab R&D project, in the Institute of Computing (ICOMP) of the Federal University of Amazonas (UFAM).
Team:
Odalisio L. S. Neto, Graduate Student at Universidade Federal do Amazonas (UFAM)
Felipe G. Oliveira, Adjunct Professor at Universidade Federal do Amazonas (UFAM)
João M. B. Cavalcanti, Associate Professor at Universidade Federal do Amazonas (UFAM)
José L. S. Pio, Associate Professor at Universidade Federal do Amazonas (UFAM)