• ALETI NAGITHA REDDY
• DR M RAMA BAI

Visual Impedance has been treated as an obstruction to normal functioning of vision system in human beings. The purpose of proposing this system is to aid visually impaired people because they require steady help in nearly all circumstances, particularly in the activities such as basic groceries shopping, self-navigating in indoor situations, reading magazines, newspapers, receipts, letters, bank documents etc. in order to know what is happening around them. Some of the main obstacles include the inability to move from one location to another without assistance of someone, “reading traffic signals and street signs can be extremely challenging. In order to overcome these challenges, visually impaired person might use walking cane, guide dog, and sighted guide. This paper demonstrates a generative model based on a deep neural network. The concept is accomplished using an Android mobile app that combines many technology elements. development such as voice assistant, caption generation, face detection, object detection into a single multipurpose app which helps in coordinating the visually impaired persons. The proposed app auto captures the input picture when the camera is centered through gadget and predicts results of that picture in the form of UI display as well as the audio output with confidence percentile of detection because, the algorithm is tuned to increase the probability of seeing the target depiction sentence in the training image. Using Convolutional Neural Network algorithm in the TensorFlow framework which has the necessary implementation modules of this work incorporated in it and are deployable into the application.

TTS, Image caption generation, Object detection, Face recognition, Audio tutorial.

"SMART HELPER FOR VISUALLY IMPAIRED PERSONS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 5, page no.h437-h441, May-2022, Available :http://www.jetir.org/papers/JETIR2205856.pdf

"SMART HELPER FOR VISUALLY IMPAIRED PERSONS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 5, page no. pph437-h441, May-2022, Available at : http://www.jetir.org/papers/JETIR2205856.pdf