Object Recognition Development for Android Mobile Devices with Text-to-Speech Function Created for Visually Impaired People

The aim of this work is presenting how to create an artificial intelligence, a computer brain or a talking smart robot with the usage of a mobile device. The system is asked to recognize an object, after it recognizes the objects and says loud what the best guess of the object is. The object was previously not seen by the system. The system recognizes objects based on image comparison with the pictures present in the database. All images from the database have attached a string which represents the object shown in the image. When finding a match of the newly taken image with an image present in the database the new image takes the string of the matching image. This way the neural network is trained. The more images are saved in the database, the more accurate the system can be. The system has added voice command and text-to-speech support in order to be a real robot to be asked what the object is and to say loud what the object is after it was recognized by the system.