Making visually impaired 'see' through technology

Making visually impaired 'see' through technology

Student innovation

Worldwide, 285 million people are estimated to be visually impaired,” says Roopam Sharma, quoting statistics from the World Health Organisation. “And”, he continues, “there hasn’t been an innovation for the community which helps them with literacy.” With this thought, Sharma along with his friends started visiting blind schools across Delhi-NCR, where they conducted researches and surveys which highlighted the three major problems faced by visually impaired people.

“These were the inability to read printed text (non-Braille), navigation outside well-known environments, and accessibility to smartphones,” says Sharma. And it was these insights that led him, along with Akshita Sachdeva and Neeraj Saini (students of Manav Rachna College of Engineering), to develop Manovue.

A combination of two words: ‘mano’ (hand in Spanish) and ‘vue’ (vision in French), Manovue is a multi-utility wearable haptic device which helps a visually impaired person move around in a complex environment, and can also be used for local-sequential text scanning where the user scans the text and hears the recognised words synthesised to audible speech.

The team recently participated in the fourth season of Accenture Innovation Jockeys and won under the ‘Cognitive Computing and Internet of Things’ category.

They say the device measures the distance of objects and provides feedback via vibrations. “The frequency of vibrations is proportional to the distance from the detected object, allowing the user to accurately discern an obstacle’s proximity. It also allows users to get access to print media by just moving their fingers on the printed text and recognise objects or people,” they say.

“Manovue can be used to automatically scan and recognise people as well as objects such as money, packaged goods, CDs, DVDs, medication bottles, as well as landmarks. Point the device video camera at what you wish to ‘see’ and it will pronounce the name quickly in clear and easy to understand speech. It can be taught to recognise all the objects and landmarks you wish to identify,” says Sachdeva, the software leader.

The device, estimated to cost Rs 4,000, is built with a coordinated phone application which makes use of integrated personal assistant technology and gives voice commands. The user just needs to wear the glove and can start using Manovue by turning the switch on.

Once switched on, the user can press the button which starts listening to instructions.

“If you say ‘start reading’ to the glove, it starts the reading function. The user just has to point his finger to any written text and the glove will convert the text into speech. If they say ‘recognise people’, the glove will speak out the name of the person at whom their index finger is pointed. Further, if the ‘start navigation’ command is given, the glove starts the navigation mode. If an object is detected by Manovue while being used in a complex environment, its vibration motor starts operating, giving a haptic feedback to the user so that he can dodge the object and move freely,” Sharma tells Metrolife.

The device comes with a complimentary mobile phone application which opens up the smart phone market for the visually impaired. It is is currently available for Android, Windows and iOS platforms.

Explaining the working principle, Saini says while the device uses the Optical Character Recognition mechanism to read; facial and object recognition are done with the help of image processing.

“For navigation, the device uses the principle of SONAR, which is used by bats and submarines. The device has ultrasonic sensors that are used for detection. The mobile phone application is based on Natural Language Processing and Artificial Intelligence,”
he adds.

The device, created under the mentorship of Dr B S Gill, director, Manav Rachna Innovation and Incubation Centre, has been developed with help from National Association of Blind, New Delhi and Faridabad and Blind Relief Association, New Delhi. It has been tested by more than 200 users at these institutions.