Abstract:

Globally, 1 billion people have a vision impairment that could have been prevented or has yet to be addressed. This 1 billion people includes those with moderate or severe distance vision impairment or blindness due to unaddressed refractive error (123.7 million), cataract (65.2 million), glaucoma (6.9 million), corneal opacities (4.2 million), diabetic retinopathy (3 million), and trachoma (2 million), as well as near vision impairment caused by unaddressed presbyopia (826 million). In terms of regional differences, the prevalence of distance vision impairment in low- and middle-income regions is estimated to be four times higher than in high-income regions. With regards to near vision, rates of unaddressed near vision impairment are estimated to be greater than 80% in western, eastern and central sub-Saharan Africa, while comparative rates in high-income regions of North America, Australasia, Western Europe, and of Asia-Pacific are reported to be lower than 10%. Population growth and ageing are expected to increase the risk that more people acquire vision impairment. In this paper in order to facilitate the blind we develop an artificial vision system in which the blind person can have a device with them which can guide the blind about the surrounding environment and help them lead a safer life as well increase awareness about the surroundings. This is been achieved by using advanced image captioning techniques implementing efficient net algorithms and tokenization methods where the scenes with different captions are learned by the machine. Whenever an image is captured via the camera are been recognized and predicted by the machine. After the prediction, it is been sent as an audio output to the user which can help them identify the scene happening around. Thus, with the help of this paper provide an artificial vision to the blind, which can help them gain confidence while travelling alone.

Keywords: