ABSTRACT
The aim of this paper is to give a comprehensive survey in the field of object detection using deep learning techniques. Convolutional Neural Network(CNN) helps to classify more than one objects in the environment with great accuracy. The paper will highlight the features of CNN variations with time that help the researcher to incorporate the CNN in developing Electronic Travel Aid for the Visually Impaired(VI) person. This study is an effort to report the research community about the capabilities of these neural network algorithms and emphasize the improvements over time. Recent advancement in CNN approach such as You Look Only Once(YOLO) and CSP-Net backbone could bring the ETA more compact, accurate and fast that would be easily affordable to VI.
