Millions of people in this world can't understand environment because they are blind or visually impaired. They also have navigation difficulties which leads to social awkwardness. They can use some other way to deal with their life and daily routines. It is very difficult for them to find something in unknown environment. Blind and visually impaired people face many difficulties in conversation because they can't decide whether the person is talking to them or someone else. Computer vision-based technologies have increased so much in this domain. Deep convolutional neural network has developed very fast in recent years. It is very helpful to use computer vision-based techniques to help the visually impaired. In this chapter, hearing is used to understand the world. Both sight sense and hearing have the same similarity: both visual object and audio can be localized. Many people don't realise that we are capable of identifying location of the source of sound by just hearing it.
TopIntroduction
Concurring factual investigation of WHO (World Health Organization) (Global Data on Visual Impairments, 2010), roughly 285 million individuals of the globe are visually impaired and 246 million have genuine vision issues. Outwardly debilitated individuals generally face challenges in development just as distinguishing individuals and staying away from obstructions in their everyday exercises. The regular answers for these circumstances are frequently observed to be use of guide sticks to recognize snags before them or hand-off on vocal speculating for distinguishing proof of people. As a result, outwardly disabled individuals can't foresee the careful condition includes about what kinds of items lies before them or whom they are confronting nearness.
Not just for the helping of outwardly disabled individuals, this idea is in usage in numerous areas, for example, security and modern assembling. The productivity and exactness contrasts by the calculation and handling capacities. Distinctive programming framework models are planned all things considered that it initially takes the info pictures of the database and actualize the profound learning procedure to group and afterward explicitly distinguish the required outcome, for example, objects, facial character or appearance, fabrication and so forth for the constant peripheries, the caught information pictures contain a few elements and needs increasingly productive program to extricate and determine the predefined classification and some of the time for continuous investigation, numerous recognition are recognized and it is a test to recognize effectively.
For the growing competitive world around, it is quite difficult for a visually impaired person to move around independently and identify surrounding objectives correctly with ease. With the advancement of technology, there are several solutions but most of them have demerits such as low acceptance, high cost, difficult to usage etc. (Gori, M., Cappagli, G., Tonelli, A., Baud-Bovy, G., & Finocchietti, S., 2016). Based on the demand, devises supporting the visually impaired people has been introduced for a time now. Again, keeping with pace, the more advanced algorithms and processing devises are introduced and progressing to higher accuracy and efficiency. This has inspired us to combine the concepts of implementing a processing devise for serving the blind individuals with a higher efficient methodology.