High-Level Project Summary
We had developed an integrated combination of different technologies in order to collect, analysis and presenting the data associated with landslides. These technologies are satellites database, sensors for more accurate data in local range, Machine Learning/Artificial Intelligence modeling, Geographic Information System mapping, and present the output on a user interface.By getting that light of the land, we achieved better, more accurate, and real-time data within local areas at a low cost.Thus, an accurate prediction within small territories can be calculated to take the necessary actions to avoid the risks of landslides.
Link to Project "Demo"
Link to Final Project
Detailed Project Description
Our prototype is an integrated environment that uses a new technology to combine all types of modern technology simultaneously by using ML, AI, GIS, and web representation.
By combining these technologies, it will be possible to upgrade data usage
Together, these technologies combined form a real-time forecasting system used in local areas, resulting in a more accurate determination of landslide hazards.
Our project is working following these steps:
Collecting data:
1- We get the rainfall and soil characterization data from Nasa satellites.
2- Then we collect data from our sensors and geological maps (local data).
3- getting data from users
Sensors used are:
1- Acoustic (microphone): it detects a cracking sound that happens before the landslides
2- Soil moisture sensor: it measures the moisture of the soil and detect any sudden leakage of water
3- Fiber optic sensor: It detects the pre-earthquakes that happen before landslides using light waves.
ML step:
We then integrate all these data and enter them into our Machine Learning model.
Presenting the output:
Presenting all the parameters in the GIS model and cast the data to our user interface.
Our user interface includes GIS mapping that showing the places where landslides are likely to happen, simulation for predicting landslides by entering the rainfall rate, leakage rate, and speed of the simulation. In addition to image uploading service to make people able to share the cracks and landslides pictures.
The benefits of our project are:
1- Providing inputs used to identify landslide risks worldwide at smaller scales that allow risk mitigation actions to be taken at the local level.
2- Providing prototypes and methodologies for integrating Earth observations with open local data
3- Encouraging participants to include information that the public can contribute by capturing data in their regions to improve the accuracy of the analysis
We hope to achieve an integrated prototype that meets the overall objective of this challenge as it is implemented at low cost by local governments
We used several tools, languages, and Softwares in order to make our project.
Tools & Softwares used were:
1- QGIS: to make our GIS map
2- blender: to make our simulation
3- FFMPEG: processing our video
4- Django: Web server to make our website
5- Azure: For cloud computing
Languages used were:
1- Python: the main language used in our project.
2- Q: for data analysis
Our Solution Website link : https://mysterious588.tech/
Space Agency Data
https://gpm.nasa.gov/data/imerg
IMERG used for getting rain in real time to help predicting. We use this data in GIS as a layer
https://svs.gsfc.nasa.gov/4631
LHAZA is used for landslide suseptibility estimated.
LHASA is the most model that inspires us as it helps predicting the landslides hazards within 30 minutes range on large scale.
Hackathon Journey
Our journey till the present time was tough yet fun. In the beginning, a dozen of new topics had opened beyond us. Although it was very distracting and challenging to work and learn such topics, it gives us a lot of knowledge and experience.
We learned a lot in various aspects like:
· Geology
· Geographic Information System
· Different types of sensors
· Nature hazards that affect societies
· Factors that contributed to these hazards
· Modern technologies that world had reach especially the satellite data
· Dealing with machine learning
· Dealing with free source data.
In addition to different skills like working under pressure, efficient researching using keywords, and management of work by applying suitable workflow for the team.
We were inspired by Egypt’s vision 2030, as they are building on mountains and plateaus. So, hazards like landslides will face them sooner or later.
Therefore, we tended to tackle this challenge in order to provide a tool to help our country with it and preventing any unexpected catastrophes epically after what happened in Mokattam plateau in 2008.
Our strategy solving this challenge was following the next steps:
1- Meetings only for understanding the challenge.
2- Brainstorming in order to get a primary plan.
3- Start researching including both online and offline like asking professors in our university.
4- Determining the final plan to work on.
5- Dividing the tasks among us.
6- Finishing and revising.
We were lucky that we don’t have a lot of obstacles, but for the small setbacks, we were holding an emergency meeting in order to get a solution for it.
References
1- CBC/Radio Canada. (2014, September 29). Landslide early warning system uses fibre-optic sensors | CBC News. CBCnews. Retrieved September 24, 2021, from https://www.cbc.ca/news/science/landslide-prediction-tool-uses-fibre-optic-sensors-1.2781262?fbclid=IwAR2MG-dQ_gceOE4BvTa6FQOBlnXkJ8N4qZ87KQ6IQrge0bQdYSFEwJru5Yk.
2- Chapter 10 - landslide Hazard assessment. (n.d.). Retrieved September 24, 2021, from http://www.oas.org/dsd/publications/unit/oea66e/ch10.htm.
3- Editors, E. (2013, September 26). Developing fiber-optic sensor networks. DigiKey. Retrieved September 24, 2021, from https://www.digikey.com/en/articles/developing-fiber-optic-sensor-networks.
4- Mostafa, S., el-aal, A. el-aziz K. A., & El-Eraki, M. (1970, January 1). Multi scenario seismic hazard assessment for egypt: Semantic scholar. undefined. Retrieved September 24, 2021, from https://www.semanticscholar.org/paper/Multi-scenario-seismic-hazard-assessment-for-Egypt-Mostafa-el-aal/78f9eb3e4e9a59152ddb62632798b5a3e9774ca7.
5- NASA. (n.d.). Giovanni. NASA. Retrieved September 24, 2021, from https://gpm.nasa.gov/data/sources/giovanni.
6- NASA. (n.d.). Landslides @ NASA. NASA. Retrieved September 24, 2021, from https://gpm.nasa.gov/landslides/index.html.
7- NASA. (n.d.). NASA. Retrieved September 24, 2021, from https://giovanni.gsfc.nasa.gov/giovanni/doc/UsersManualworkingdocument.docx.html.
8- Nasa. (n.d.). NASA/Lhasa: Landslide hazard analysis for situational awareness. GitHub. Retrieved September 24, 2021, from https://github.com/nasa/lhasa.
9- Noviyanto, A., Sartohadi, J., & Purwanto, B. H. (2020, September 11). The distribution of soil morphological characteristics for landslide-impacted Sumbing volcano, Central Java - Indonesia. Geoenvironmental Disasters. Retrieved September 24, 2021, from https://geoenvironmental-disasters.springeropen.com/articles/10.1186/s40677-020-00158-8#Sec2.
10- Sartohadi, J., Harlin Jennie Pulungan, N. A., Nurudin, M., & Wahyudi, W. (2018, May 27). The ecological perspective of landslides at soils with high clay content in the Middle Bogowonto watershed, Central Java, Indonesia. Applied and Environmental Soil Science. Retrieved September 24, 2021, from https://www.hindawi.com/journals/aess/2018/2648185/.
11- Staff, P. H. I. V. O. L. C. S. (n.d.). Landslide prepareness. Philippine Institute of Volcanology and Seismology. Retrieved September 12, 2021, from https://www.phivolcs.dost.gov.ph/index.php/landslide/landslide-prepareness.
13,https://gpm.nasa.gov/data/imerg#:~:text=The%20Integrated%20Multi%2DsatellitE%20Retrievals,majority%20of%20the%20Earth's%20surface.
14,https://soilgrids.org/
15,https://svs.gsfc.nasa.gov/4631
Tags
#Landslides #GIS #Machine Learning #Artificial Intelligence #Cloud computing
Global Judging
This project has been submitted for consideration during the Judging process.