Toward designing efficient application to identify objects for visually impaired

Abstract

There is an increasing interest in developing technologies that attempt to help visually impaired people in their daily lives. However, it is shown that the object identification task is still the major difficulty for visually impaired people. Although there are many applications that can be used for this task, there are still obvious limitations that require more improving. For this reason, this paper provides an analysis and evaluation for the technologies that used in the object identification task. As a result of this evaluation, it is shown that the best technology is the human computation. Moreover, a comparison between the applications that use human computation is provided in order to identify the best features that can be considered in designing efficient application to identify objects for visually impaired.