Mobile visual recognition and applications
Tan, Amanda Qian Yi
Date of Issue2017-05-15
School of Electrical and Electronic Engineering
With the increasing popularity and usage of smart phones, people are getting increasingly dependent on their smart phones to do the work for them. This includes making use of the mobile visual search system, which has been an emerging trend in different mobile applications. Mobile visual search involves using images, rather than text, to search for information. This makes for a fast and convenient method of obtaining information related to the image scanned. This project aims to develop a mobile application making use of mobile visual search system. The name of this mobile application is “Textbook Companion”. It is targeted towards both primary and secondary school students to aid in their learning. This project involves two sides – the front end and the back end. The front end is the mobile application, where the user can take a picture of an image from their science textbook. The image will then be transmitted to the server for further processing. To match the image captured to a prebuilt database of images, the Bag-of-Words (Bow) model is used. For every image queried, SIFT descriptors are extracted and compared against local image descriptors. The highest scored images are then returned as candidates and geometric verification is used to find the best match among the candidates generated. The back end is the web application, where the administration side can manage the database of textbook images. The enhanced web application contains different functionalities like uploading to the database, updating the database and viewing the existing records in the database. Users are also able to choose between uploading/updating a single record, or multiple records. After integrating the mobile application and the web application, the “Textbook Companion” should serve as a useful mobile application with high accuracy for students to aid in their learning. At the same time, the administration side should be able to manage the database of images efficiently and conveniently.
Final Year Project (FYP)
Nanyang Technological University