2 of 18

GO MUSEUM

Team:

Karim Yasser, Ahmed Amr

Amr Salama & Loay Yehia

Supervised By:

Dr. Fatma Helmy

Supervising Assistant:

Eng. Samira Raafat

Date: 4/1/2022

3 of 18

INTRODUCTION

PROBLEM STATEMENT

SYSTEM OVERVIEW�

USE-CASE DIGRMA

TABLE OF CONTENTS

NON-FUNCTIONAL REQUIRMENT

CLASS DIAGRAM

DATA AUGMENTATION

GAN

DATASET

FUNCTIONAL REQUIRMENT

DEMO

4 of 18

OBJECTIVE

facilitate the tourist tour by guiding them to have the best experience

The goal is to create a user-friendly mobile application that will allow tourists to visit historical sites without having to hire a tour guide

Augmented reality layer that will add an explanation about the recognized statues

It will also have the ability to out-paint an occluded Monuments.

5 of 18

PROBLEM STATMENT

The main problem focused on is monuments detection. Firstly, the aim of Go Museum is the recognition of statues. However, the main problem Go Museum is trying to solve is addressing occluded, disoriented, and distorted monuments

6 of 18

SYSTEM OVERVIEW

7 of 18

CONTEXT DIAGRAM

Cropped image

8 of 18

USE CASE DIAGRAM

9 of 18

FUNCTIONAL REQUIRMENT�

The User shall be able to Scan the monuments.

The user should be able to view the history of Monuments he Scanned

After the user Scan the Monument its history will be displayed as speech.

The Admin Can manage the Application pages

The System shall be able to preform Advanced Deep learning Technique Called 'GAN' on the Scanned images by the user to help on Scanning any occluded monument

The system shall show augmented reality monument to the user that tells the story of the monument

10 of 18

NON - FUNCTIONAL REQUIRMENT�

The system’s security must be reinforced by hashing or encrypting all passwords.,

The system must prevent the entry of erroneous data types or the deletion of critical data without providing confirmation warnings.

It should be simple to update and expand the system. Using a design pattern model like the MVVM model will make changing the system much easier.

The system must be able to handle an increasing volume of data and perform well

The system will be available on the all device, regardless of operating system

The system shall be available at all times, so that the user can use it at anytime

Security

Maintainability

Availability

Scalability

Portability

11 of 18

CLASS DIAGRAM

12 of 18

Generative adversarial network�

We intend to apply GAN neural network in out-painting for monument detection since this approach is guaranteed to detect visible and occluded monuments However, there are many hyper parameters to be selected carefully for better design, such as the number of epochs.

13 of 18

DATA AUGMENTATION(Experiment)

Data augmentations have been widely studied to improve the accuracy and robustness of classifiers. We provide insights and guidelines on how to augment images for both vanilla GANs and GANs with regularizations, improving the fidelity of the generated images substantially When this GAN training is combined with other augmentation-based regularization techniques, the augmentations further improve the quality of generated images . we provide both consistency loss and contrastive loss as additional regularizations.

14 of 18

DATASET

The original dataset had around 284 images(PNG/JPG) collected in 22 Class. After augmentation, the total is 1,724 images. Several augmentation methods were used such as zooming, or brightness, and re-scaling. Our finalized dataset set is divided into 22 classes that are ready to be applied to our proposed techniques.

Original

Augmented

15 of 18

RFID Tags Vs Go Museum

An RFID tag is an

An object that can be attached to or incorporated into a

product, animal, or person for the purpose of

identification using radio waves.

1 of 18