Development of an ergonomic and low-cost assistive device based on IoT and computer vision to assist visually impaired individuals in the city of Arequipa.

Visual impairment is a condition that presents significant challenges in the daily lives of those who experience it. In the city of Arequipa, as in many other regions, people with visual disabilities face additional difficulties when trying to identify banknotes and objects in their everyday environment. Therefore, this research aims to develop an ergonomic and low-cost assistive device for banknote recognition based on computer vision and IoT for people with visual impairments in the city of Arequipa.

The methodology used was based on the Design Thinking approach, whose phases detail the development process of the proposed assistive device: (1) empathy, (2) definition, (3) ideation, (4) prototyping, and (5) evaluation. For the evaluation of the device, both functional and non-functional tests were conducted. Functional tests focused on verifying banknote recognition with 44 members of the Asociación Unión de Ciegos de Arequipa under different environmental conditions. On the other hand, non-functional tests were aimed at evaluating aspects such as performance, security, reliability, and usability of the device.

The results demonstrate that the device meets the required specifications and achieves an accuracy level of 99.75% using the Xception computer vision model. In conclusion, it was possible to build an ergonomic and low-cost assistive prototype for banknote recognition for people with visual impairments through the integration of computer vision techniques with IoT technology.

Reference