Skip to content

Using CNN to estimate the intensity of cyclone using INSAT-3D IR Images

Notifications You must be signed in to change notification settings

Chaitanya134/SCIE.ai-Underfitters

Repository files navigation

SCIE.ai

Using CNN to estimate the intensity of cyclone using INSAT-3D IR Images

image

Problem Statement

Development of a deep Convolutional Neural Network (CNN) for Tropical Cyclone intensity estimation using half-hourly INSAT-3D IR Images and development of a web application for visualization of the imagery. INSAT3D/3DR observations are available at every 15- minute interval and these observations are very useful in understanding the instantaneous structural changes during evolution, intensification, and landfall of Tropical Cyclones. Datasets of Cyclones captured by INSAT-3D over the Indian Oceans are available since 2014. These datasets can be used for training and testing of the Model. Traditional methods for Intensity estimation require accurate center determination for intensity estimation. Development of CNN based model for intensity estimation will be very useful during the initial stage of cyclone formation when determination of accurate center becomes difficult.

Our Solution

SCIE.ai (Smart Cyclone Intensity Estimator.ai)

How our solution works

  • We have made a Convolutional Neural Network-based architecture that uses Deep learning to estimate the intensity of cyclones using INSAT-3D satellite images.
  • Our model gives around 90% accuracy as of now and we plan to increase it further in the coming future.
  • We have also made a web application that would take an image as an input and give the prediction in a very user-friendly way.
  • Our Aim is to forecast eddy currents that causes cyclones several hours prior to them happening.

Architechture

image

Flow Chart

image

image

Technologies

  • Python
  • Tensorflow(keras)
  • React
  • Javascript
  • CNN
  • FastAPI

Future Plans

  • Using a better database that has a lot more images so as to increase our accuracy.
  • We also plan to use other factors of the cyclone like hole size and latitudes and longitudes and using historic data, Predict wind speed and thus, cyclone intensity.

Issues that we may face

  • Lack of Proper Dataset
  • Inconsistency in Data
  • Complexity

About

Using CNN to estimate the intensity of cyclone using INSAT-3D IR Images

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages