Object Recognition Using Perspective Glass For Blind/Visually Impaired

S. Naveen Kumar, K. Varun, J Mohamed Rahman


Perspective glass is wearable pair of glass which is designed for blind people which helps them in resolving a major difficulty they face such as identifying the objects or obstacles present before them during walking. The Perspective glass consists of a raspberry pi board, 5mpcamera, ultrasonic sensors, buzzer, headphone, power source. This glass is controlled by a Power button which when pushed ON, will take pictures of the surroundings with respect to the position of the person wearing it. The clarity and the resolution of the picture purely reside on the camera used. The captured information is transferred to the application software which is manually built with the help of Fire Base application. The transferred data is then analyzed with the help of library files that are interfaced to the application via Neural Network “Tensor Flow”. Finally the captured image is recognized with the help of Tensor Flow, and the information regarding the object is given out as a voice output to the person via speaker/headphones. All the components are interfaced to the Raspberry pi board which act as the central processing unit. It has the overall control on this setup. In addition to that an Ultrasonic Sensor is connected to the setup which when sensing an obstacle within a particular distancze of 3 meter, will produce a buzzer sound instantly giving a caution alert to the blind person. The overall setup is powered through an external power source (power bank). These Smart Glasses for Blind people is a portable device, easy to use, light weight, and user friendly. These glasses could easily guide the blind people and help them in better handling of obstacles.                    


Full Text:



Bourne RRA, Flaxman SR, Braithwaite T, Cicinelli MV, Das A, Jonas JB, et al.; Vision Loss Expert Group. Magnitude, temporal trends, and projections of the global prevalence of blindness and distance and near vision impairment: a systematic review and meta-analysis. Lancet Glob Health. 2017 Sep; 5(9):e888–97.

Md Sheikh Sadi, Saifudin Mahmud, Md Mostafa Kamal, Abu Ibne Bayazid “Automated walk-in assistant for blinds”, Electrical Engineering and Information and Communication Technology (ICEEICT), 2014 International Conference, 2014.

Feng Lan, Guangtao Zhai, Wei Lin “Lightweight smart glass system with audio aid for visually impaired people”, TENCON, IEEE, Region 10 Conference, 2015.

W.C.S.S. Simoes, V.F.de Lucena “Blind user wearable audio assistance for indoor navigation based on visual markers and ultrasonic obstacle detection”, Consumer Electronics (ICCE), 2016 IEEE International Conference, 2016.

Chinese Author “intelligent auxiliary system blind glasses”, CN106937909A,11thJuly,2017

CI2S Lab of Buenos Aires, UAI, AUSTRAL and UADER, BuenosAires,Argentinamdldl@ci2s.com.ar.

Chen Wang Future Media & Convergence Institute, Xinhuanet Beijing, China.wangchen@news.cn.


  • There are currently no refbacks.