An AI-enabled mobile app to aid visually and hearing impaired to identify the denomination of Indian currency notes
The Reserve Bank of India (RBI) is the central bank of India that manages all major monetary policies of India and handles the economic stability and growth. RBI was set up in 1935 under the Reserve Bank of India Act,1934 and since then RBI has been playing an important part in the Development Strategy of the Government of India. It is also a member bank of the Asian Clearing Union.
- 500,000+ Downloads
- 99.9% Accuracy
- 150,000 Size of Training Dataset
- < 1.5s Detection time
The Reserve Bank of India (RBI) issued new currency notes in 2016. These notes in denomination of INR 10, 20, 50, 100, 200, 500, and 2,000 are identified and distinguished using different currency-embedded features.
Tactile markers and embossments on currency notes help the visually challenged ascertain the denominations of the notes. However, since such marks often fade with use, the visually challenged people often find it difficult to identify the denominations of the notes. RBI planned to eliminate this problem through technology.
The requirement was to develop a mobile application that would help the visually and hearing impaired to identify denominations of currency notes through voice/sonic and vibration patterns.
At the inception, team Daffodil presented a technical proposal, project management plan, software requirement specifications, high-level & low-level design documents, user acceptance test cases to RBI, which were approved by their technical team to proceed for the development process.
To counter the usage of the app in poor internet connectivity areas, the app was required to work in the offline mode as well. In order to scan the banknotes in the offline mode, machine learning capabilities were introduced in the app. Since the app would not be able to learn by itself due to its offline nature, it was important for the Daffodil team to train the app for the live release. Daffodil created a data set of 1,50,000 images of Mahatma Gandhi series banknotes in the denomination of Rs 10, 20, 50, 100, 200, 500, and 2,000 in order to train the app. Whenever a user scans a note, the app compares the scanned image of the note with 1,50,000 images in the data set and gives accurate results in the offline mode or low lighting conditions.
The application has been downloaded by more than 500,000 users since its launch in January 2020. This initiative has also been lauded by several mainstream media agencies throughout the country. The mobile app developed by team Daffodil is enabling the visually and hearing impaired to perform the following actions:
- Identifying the denominations of Mahatma Gandhi Series and Mahatma Gandhi (New) series banknote (Rs 10, 20, 50, 100, 200, 500, and 2,000) by checking front or reverse side/part of the note including half folded notes at various holding angles and a broad range of light conditions (normal light, low light, daylight, etc.).
- Identifying the denomination through an audio notification in Hindi/English and non-sonic mode such as vibration.
- Identifying the banknotes denomination without the internet.
The app works in offline mode and automatically turns on/off the flash light of the smartphone to adjust lights.
- Navigating the mobile application via Siri and Talkback feature for accessing the application features wherever the underlying device & operating system combination supports voice-enabled controls.
Read Related Case Studies
Get in Touch
Sign up for a 30 min no-obligation strategic session with us
Let us understand your business objectives, set up initial milestones, and plan your software project.
At the end of this 30 min session, walk out with:
- Validation of your project idea/ scope of your project
- Actionable insights on which technology would suit your requirements
- Industry specific best practices that can be applied to your project
- Implementation and engagement plan of action
- Ballpark estimate and time-frame for development