Bluetooth Based Product Identification through Voice for Blind People

Abstract

A cam based assistive content perusing system is proposed to help blind persons read content names and item bundling from hand-held protests in their day by day lives. To detach the object from jumbled foundations or other encompassing protests in the cam view, implementation is done in two steps. First training of the object is done by which the images of the objects are stored in the data base and RGB and entropy mean values of those images are calculated. In the second step testing of the objects is performed. Here the images of the products to be tested are taken and RGB values of them are calculated. These are compared with the trained image RGB values if a match then the image name is obtained. Through serial communication it is transferred to the micro controller. It is compared with the code in the controller and the result is transferred through Bluetooth to the android mobile. In mobile by using Android SDK tool an application will be developed such that the name of the object and the brief description of the product will be heard through head phones or in speakers. Thus the system helps the blind people to know the details of the handheld objects.