Smart VR Based Urban Flood Disaster Management App Using Deep Learning & Drone Technologies

Project Title

Smart VR Based Urban Flood Disaster Management App Using Deep Learning & Drone Technologies

Project Area of Specialization Artificial IntelligenceProject Summary

Urban floods are considered to be one of the natural disasters. It affects all most every area of life all over the world. Besides heavy rainfall, environmental changes and low drainage of water the improper utilization of resources, and inappropriate planning of urban development are some of the factors that rises the effects of such floods. According to Asian Development Bank (ADB), Climate Risk Country report 2021 Pakistan’s on average suffer $1 billion every year because of such floods. In Pakistan, more than 20,000 people get affected every year because of them. Due to them, major metropolitan cities get choked and people get trapped in their houses for days. In literature, it has been found that many experts have presented mathematical and statistical-based models to predict the after-effects of such floods. These models are very complex and time-consuming. Moreover, the accuracy of such models is also very low. To overcome the above-mentioned gaps a drone-based smart VR-based urban flooding disaster management app is proposed. The basic objective of this application is to provide a realistic view to the metropolitan planners about the road conditions after the rain.  This application captures real-time data through drone and spot the dangerous areas in 3D VR view. The basic modules of the applications will be Image Processing, Flood Prediction Module, 3D Mapping, and Mobile Application Manager. For the successful implementation of the proposed project various tools will be used (Python, PHP, and Android Studio (Unity Package), and Makesense.ai), techniques (DL, and photogrammetry), and technologies (Smart Phones, VR glasses and Drone) android platform with the combination of python and C-Sharp (C#) for app development will be used. 70% of the data sets will be used for training and 30% will be used for testing. This project will not only facilitate the metropolitan department but also help the researchers in their further researches.

Project Objectives

The objectives of the proposed idea are
•    An empirical investigation of the literature will be done in this project which helps in the identification of suitable tools, techniques, and methodologies needed for the successful completion of the project.
•    An android based mobile application will be created which captures the realistic images of flood-affected areas through drone and display them in a 3D VR view.
•    In the Image Processing module, image acquisition is done through the drone. These images are further processed to train the proposed predicted model development by using Deep learning techniques (e.g. maskRCNN). 
•    Flood Prediction module acquires drone-based realist images from the mobile application module. The basic purpose of this module will be to run the predicted model and classifies the flood-affected areas (i.e. normal, moderate, and catastrophic). The captured images are further forwarded to the 3D mapping module for the construction of the 3D model. 
•    3D Mapping module will develop a 3D map of the flood-affected areas with the help of the photogrammetry technique.  For a realistic view of the 3D maps, it will be further forwarded to the mobile application user interface where users can see the affected areas in the VR environment by using VR glasses. 
•    More than 1000 data sets will be taken from a realistic environment out of which 70% will be used for training and the remaining 30% will be used for testing.
•    The effectiveness of the final experimental results will be tested and verified by the mobile application connected with the database.

Project Implementation Method

The proposed Smart VR-based urban flood disaster management app is an effective and efficient way the detection of flood-affected areas.  The block diagram of the project shown in Fig. 1 elaborates the working of the four basic modules (i.e. Image Processing, Flood Prediction Module, 3D Mapping, and Mobile Application Manager). Each module is further decomposed into sub-sections. The whole project is logically divided into three main sections. In the first section, the training on the proposed model will be done. In the second section, the realistic data will use the trained model and identify the damaged areas.  In the third section, the 3D VR view will be provided to the user for assessing the intensity and spectrum of the flood-affected areas.

          Figure 1: Block Diagram of the Proposed Project

Benefits of the Project

The basic benefits provided by the proposed project are:
•    It will facilitate numerous authorities such as land development authorities, metropolitan planners, and civil engineering departments for better decision making while constructing roads, manholes, pedestrians, poles, and underground lines(e.g. gas, internet, etc.).
•    It will beneficial to the citizens by providing them post flooded areas detail. The information facilitates them to choose alternative paths of traveling which will reduce traffic congestion hectic, save money, time and energy. 
•    It will be an effective tool for assessing the damage to the affected infrastructure.
•    The empirical investigation done in this project facilitates the researchers and practitioners to further investigate this area in different scenarios and contexts. 
•    The datasets generated by the proposed software using deep learning techniques will be used for further research.

Technical Details of Final Deliverable

In the end, we will deliver the code and the documentation in the form of a final project report containing all the proposed algorithms pseudo and implementation code.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries Others , Security Core Technology Augmented & Virtual RealityOther Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and Infrastructure, Sustainable Cities and Communities, Life on LandRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Drone	Equipment	1	50000	50000
VR Glasses	Equipment	1	6000	6000
Mobile Phone	Equipment	1	14000	14000
Documentation	Miscellaneous	1	10000	10000