Name:- Sarbojit Ghosh

Department:- Computer Science and Engineering

University Roll No.:-16500121061
Registration No.:-211650100110030(2021-22)
Year:- 3rd
Project:- Research Methodology (PROJ- CS601)
Topic:- Flood Visualizer
 1

INDEX:-
TOPIC PAGE No.

1.Introduction ___________________________ 2

2.Declaration ____________________________ 3

3.Certification ____________________________ 4

4.Software Requirements Specifications ________ 5-9

5.Method ________________________________ 10-14

6.Implementation(& Code) ___________________ 15-23

7.Summary ________________________________ 24

8.Conclusion & Reference ___________________ 25,26

 2

❖​ Introduction:

The Flood Visualizer project aims to develop a tool utilising

Google Earth Engine to visualise and analyse flood patterns.
Floods are one of the most devastating natural disasters,
affecting millions of people worldwide annually. By leveraging
satellite imagery and remote sensing data, this tool provides
insights into flood occurrences, extents, and impacts, aiding in
disaster management and mitigation efforts.

We are working here with the devastating flood that occurred in

Ernakulam district,Kerala on 16th August,2018.The Indian
government had declared it a Level 3 Calamity, or "calamity of
a severe nature".Probably the worst flood of the entire century.

Furthermore,the Flood Visualizer API represents a pivotal

element within our project ecosystem, serving as the backbone
for data management, retrieval, and dissemination. Through
this API, users gain access to critical flood-related information,
empowering them to make informed decisions and bolster their
preparedness in the face of potential flooding events.

As we embark on this journey, our goal is to not only provide

users with a reliable and efficient platform for accessing
flood-related information but also to foster a sense of
community resilience and empowerment. By harnessing the
power of technology and data-driven insights, we aim to
mitigate the impacts of flooding and promote sustainable
strategies for disaster preparedness and response.
 3

❖​ Declaration:
This is to declare that the project report entitled submitted by
Srinjoy Ganguly, Sarbojit Ghosh, Biki Saha, Disha Majumder,
Janvi Kumari to the Calcutta Institute of Engineering and
Management, Kolkata in partial fulfilment for the award of the
degree of B. Tech in Computer Science and Engineering is a
bona fide record of the project work carried out by them under
my supervision. The contents of this report, in full or in parts,
have not been submitted to any other institution or University
for the award of any degree or diploma.
Mrs.Subhasree Maity

Signature of students:

●​ Biki Saha

●​ Disha Majumder

●​ Janvi Kumari

●​ Sarbojit Ghosh

●​ Srinjoy Ganguly
 4

❖​ Certification:
This is to certify that the project report entitled submitted
by Srinjoy Ganguly, Sarbojit Ghosh, Biki Saha, Disha
Majumder, Janvi Kumari to the Calcutta Institute of
Engineering and Management, Kolkata in partial fulfilment
for the award of the degree of B. Tech in Computer Science
and Engineering is a bona fide record of the project work
carried out by them under my supervision. The contents of
this report, in full or in parts, have not been submitted to
any other institution or University for the award of any
degree or diploma.
Mrs. Subhasree Maity.

_______________________ _______________________
Mr. Debjyoti Bagchi Mrs. Subhasree Maity
Head of the Department
Computer Science and Engineering

Date:
 5

❖​ Software Requirements Specifications(SRS):

1.Introduction:
1.1)Purpose
We are working here with the devastating flood that
occurred in Ernakulam district,Kerala on 16th
August,2018.The Indian government had declared it a
Level 3 Calamity, or "calamity of a severe
nature".Probably the worst flood of the entire century.
The Flood Visualizer project aims to develop a tool
utilising Google Earth Engine to visualise and analyse
flood patterns. Floods are one of the most devastating
natural disasters, affecting millions of people worldwide
annually. By leveraging satellite imagery and remote
sensing data, this tool provides insights into flood
occurrences, extents, and impacts, aiding in disaster
management and mitigation efforts.The purpose of this
document is to outline the requirements for the
development of a Flood Visualizer using Google Earth
Engine (GEE). This software will enable users to
visualize and analyze flood events using satellite
imagery and geospatial data provided by Google Earth
Engine.

1.2)Scope
The Flood Visualizer will provide users with the ability
to:
●​ Visualise historical and near real-time satellite
imagery related to flood events
 6

●​ Analyze flood extent,severity,and dynamics using

various geospatial analysis techniques.
●​ Generate customizable visualisations and reports
for further analysis and decision making.

1.3)Definition,Acronyms and Abbreviations:

SRS:Software Requirements Specification
GEE:Google Earth Engine
API:Application Programming Interface

2.System Overview:
The Flood Visualizer will be a web-based application
built on top of the Google Earth Engine platform. It will
utilize GEE's extensive catalog of satellite imagery and
geospatial datasets to provide users with flood-related
data and analysis capabilities.

3.Functional Requirements:
3.1)User Authentication:
●​ The system shall provide user authentication
functionality to ensure secure access to the
application.
●​ Users shall be able to log in using their Google
account credentials.

3.2)Data Visualization:
 7

●​ The system shall allow users to visualize historical

and near real-time satellite imagery of flood events.
●​ Users shall be able to overlay flood extent and
severity layers on the satellite imagery.

3.3)Geospatial Analysis:
●​ The system shall provide geospatial analysis tools to
analyze flood extent, severity, and dynamics.
●​ Users shall be able to delineate flood boundaries,
calculate flood extent, and identify flood-affected
areas.
●​ The system shall support basic geospatial operations
such as buffering, clipping, and spatial querying.

3.4)Customization:
●​ The system shall allow users to customize
visualizations by adjusting layer transparency, color
schemes, and symbology.
●​ Users shall be able to create and save custom
visualization presets for future use.

3.5)Reporting
●​ The system shall enable users to generate
customizable reports summarizing flood-related data
and analysis results.
●​ Reports shall include maps, charts, and tables
depicting flood extent, severity, and other relevant
information.
 8

4.Non-Functional Requirements:
4.1)Performance
●​ The system shall provide responsive performance,
allowing users to interact with the application
smoothly and efficiently.
●​ Data visualization and analysis tasks shall be
completed in a timely manner, even for large
datasets.

4.2)Reliability
●​ The system shall be reliable and available for use at
all times, with minimal downtime for maintenance
and updates.
●​ Data integrity shall be maintained to ensure the
accuracy and consistency of flood-related
information.

4.3)Security
●​ The system shall implement robust security
measures to protect user data and prevent
unauthorized access.
●​ User authentication and authorization mechanisms
shall comply with industry standards for data privacy
and security.

4.4)Usability
 9

●​ The system shall be user-friendly and intuitive, with

clear navigation and informative user interfaces.
●​ Help documentation and tutorials shall be provided
to assist users in learning how to use the application
effectively.

5.System Constrains
5.1)System Constraints
●​ The system shall be developed using web
technologies compatible with Google Earth Engine,
such as JavaScript and HTML5.
●​ Integration with Google Earth Engine APIs and
services shall be utilized for data access and
analysis.

5.2)Data constraints
●​ The availability and quality of satellite imagery and
geospatial datasets may vary depending on data
sources and acquisition parameters.
●​ Users may encounter limitations in data coverage
and resolution, particularly for remote or sparsely
populated regions.
 10

❖​ Method:

1.Theory:
The Flood Visualizer project utilizes a combination
of remote sensing, geographic information systems
(GIS), and image processing techniques to analyze
satellite imagery and detect flood extents. The
theoretical framework involves several key
components:

1.Remote Sensing Data: Satellite imagery,

particularly from sensors such as Landsat and
Sentinel, serves as the primary source of data for
flood detection. These images provide
multispectral data at various wavelengths, enabling
the identification of water bodies and land cover
changes associated with floods.

2.Image Processing Algorithms: The project

employs image processing algorithms to
preprocess satellite imagery, enhance features,
and extract relevant information related to flood
extents. Techniques such as image classification,
thresholding, and spectral analysis are utilized to
differentiate water bodies from other land cover
types.

3.Google Earth Engine (GEE): GEE is a

cloud-based platform that facilitates the analysis of
 11

geospatial data at scale. By leveraging GEE's vast

archive of satellite imagery and powerful computing
capabilities, the Flood Visualizer project can
efficiently process large volumes of data and
generate flood maps in near real-time.

4.Flood Detection Algorithm: The project utilizes a

flood detection algorithm that combines spectral
indices, such as Normalized Difference Water
Index (NDWI), with thresholding techniques to
identify flooded areas within satellite imagery. This
algorithm considers factors such as water
reflectance properties, vegetation indices, and
topographic features to improve the accuracy of
flood detection.

2.Entity Relationship Diagram (ERD):

The ERD illustrates the relationships between
various data entities involved in the Flood
Visualizer project:

●​ Satellite Imagery: Represents the raw satellite

images acquired from sources such as Landsat
and Sentinel.
●​ Preprocessed Imagery: Refers to the satellite
images after preprocessing, including radiometric
and geometric corrections.
●​ Flood Extent Data: Indicates the output data
generated by the flood detection algorithm,
 12

representing the extent of flooded areas within

the satellite imagery.
●​ User Inputs: Encompasses any user-provided
inputs, such as location coordinates, timeframes,
and analysis parameters.
●​ System Outputs: Denotes the final outputs
produced by the Flood Visualizer tool, including
flood maps, statistical summaries, and
visualizations.

3.Data Flow Diagram (DFD):

The DFD illustrates the flow of data within the
Flood Visualizer system:

●​ Input Data: User inputs, including geographic

coordinates, timeframes, and analysis preferences,
are provided to the system.
●​ Data Preprocessing: Raw satellite imagery
undergoes preprocessing steps, such as
radiometric and geometric corrections, to enhance
data quality and consistency.
●​ Flood Detection Algorithm: The preprocessed
imagery is analyzed using the flood detection
algorithm, which identifies flooded areas based on
predefined criteria and thresholds.
●​ Output Generation: The detected flood extents are
mapped onto the satellite imagery, generating flood
maps and visualizations for user interpretation.
 13

●​ User Interface: The final outputs are presented to

the user through an intuitive graphical interface,
allowing for interactive exploration and analysis of
flood patterns and extents.

Level-0 DFD diagram:

Level-1 DFD diagram:

 14

Overall, the methodology employed in the Flood

Visualizer project integrates advanced remote sensing

techniques, image processing algorithms, and cloud

computing infrastructure to deliver a powerful tool for

flood monitoring and analysis.

 15

❖​ Code and Implementation of Flood

Visualizer:

1)Load and Filtering Sentinel-1 Data:

Output:
 16

2)Visualising RGB composite:

Output:
 17

3)Applying Speckle Filter:

18
 19

Output:
Before Flood:

Before Flood (Filtered):

 20

After Flood:

After Flood(Filtered):

4)Applying A Threshold:
We will calculate only more than 8-pixels of water as a
legitimate flooded area and all exceptions are filtered
out.
 21

Output:
Before Threshold(Purple patches):

After Threshold(Orange Patches):

5)Applying Musk:
•We need to discriminate Permanent water bodies
from the actual flooded areas so that we don’t
confuse between them.Importing GSW(Global
Surface Water Mapping Layers),it helps recognise us
 22

the any river(that generally fills up after rain) or any

type of water source then it will let us identify it.
•Apart from that we can also apply a filter for Steep
areas.We used elevation data HYDROSHADES
void-filled DEM,another dataset for steep areas.
Importing two datasets:

Output:
 23

6)Calculating Total Area Affected Due To Flood:

Output:
 24

❖​ Summary:

The Flood Visualizer project harnesses the power of

Google Earth Engine to provide a comprehensive
solution for monitoring and analyzing flood patterns.
By amalgamating remote sensing data and image
processing techniques, the tool offers users an
intuitive interface to visualize flood extents, assess
impacts, and aid in disaster management efforts.
Through the utilization of satellite imagery and GIS
technologies, the Flood Visualizer empowers users
with valuable insights into flood occurrences,
facilitating informed decision-making and proactive
response strategies. While the current implementation
showcases the potential of the tool, ongoing
development efforts are focused on refining
algorithms for improved accuracy, enhancing user
experience, and incorporating real-time data feeds.
The Flood Visualizer represents a significant step
forward in leveraging advanced technologies to
address the challenges posed by natural disasters,
contributing to enhanced resilience and preparedness
in flood-prone regions worldwide.
 25

❖​ Conclusion:

The Flood Visualizer project harnesses the power of

Google Earth Engine and remote sensing
technologies to address the critical need for effective
flood monitoring and analysis. Through its intuitive
interface and robust analytical capabilities, the tool
offers valuable insights into flood occurrences,
extents, and impacts, empowering stakeholders with
the knowledge needed to make informed decisions
in disaster management and mitigation efforts.

While the current implementation marks a significant

step forward in flood visualization, there remains
ample room for future enhancements and
advancements. By continually refining algorithms,
integrating real-time data sources, and improving
user accessibility, the Flood Visualizer can evolve
into an indispensable tool for emergency responders,
policymakers, and researchers alike.

As we look ahead, the Flood Visualizer project

stands poised to make a lasting impact in enhancing
flood resilience and preparedness worldwide,
demonstrating the transformative potential of
technology in addressing complex environmental
challenges.
 26

❖​Reference:
References for the Flood Visualizer API mini
project may include:
1.Gorelick, Noel et al. "Google Earth Engine:
Planetary-scale geospatial analysis for everyone."
Remote Sensing of Environment, vol. 202, 2017.
2.Brakenridge, G. Robert et al. "Development of
the Dartmouth Flood Observatory archive."
Natural Hazards and Earth System Sciences, vol.
3, no. 6, 2003.
3.Yan, Kai et al. "A Review of Satellite Remote
Sensing Applications for Surface Water
Monitoring." Remote Sensing, vol. 13, no. 10,
2021.
4.Dottori, Francesco et al. "Development and
validation of a global database of historical flood
events." Natural Hazards and Earth System
Sciences.
5.Open Geospatial Consortium. "OGC Web
Services Common Specification." OGC Standard,
Version 2.0, 2018.
https://docs.ogc.org/cs/webservicescommon/18-0
95.html
6.Google Earth Engine Developers. "Google Earth
Engine Documentation."
https://developers.google.com/earth-engine/