Developed by researchers Suriya Singh, Kumar Vishal and C.V. Jawahar
from the International Institute of Information Technology, Hyderabad,
and Shushman Choudhury from the Indian Institute of Technology,
Kharagpur, the currency recognition software relays the information as
an audio clip.
It can be used on low-end smart phones. In our current approach we have been able to report a recognition accuracy of 96.7 percent on 2,584 images. We have worked with the Indian Rupee.
However, it can be extended to other currencies," told Singh. The app will be available in English, Hindi, Bengali, Telugu, Kannada. It takes 1.09 seconds for the app to tell the currency denomination, running on a typical smart phone with specifics of 1.2 Ghz CPU and IGB RAM.
It can be used on low-end smart phones. In our current approach we have been able to report a recognition accuracy of 96.7 percent on 2,584 images. We have worked with the Indian Rupee.
However, it can be extended to other currencies," told Singh. The app will be available in English, Hindi, Bengali, Telugu, Kannada. It takes 1.09 seconds for the app to tell the currency denomination, running on a typical smart phone with specifics of 1.2 Ghz CPU and IGB RAM.
Source: Latest News from Latest Tech News