Public Discourse
Data science with Satellite Imagery

The observation of the world through satellite imagery has greatly improved in the last 20 years. Public domain imagery from organisations like NASA and ESA has become better, and the private sector is actively putting up satellites creating images which can be purchased and put into commercial applications. This is a global frontier of observation of the world, and it is particularly valuable in data-poor situations.

Working with satellite imagery requires a combination of high-performance computation and statistical thinking. The datasets are extremely large and require special effort in computer engineering and data science. In addition, traditional statistics knowledge needs to be adapted to fit this space.

In order to understand the possibilities of this world, and to get people going with hands-on capabilities in working with this data, Chennai Mathematical Institute and XKDR Forum will organise a two-day workshop. Kawa Space is the knowledge partner for this event, and Spacefed is the ecosystem partner.

19 - 20 August 2022
Chennai Mathematical Institute H1-Sipcot IT Park, Siruseri, Kelambakkam, Chennai - 603103 Tamil Nadu
09:30 - 10:30

Satellite data in the emerging world of alternative data

Hemen Sampat, JP Morgan

10:30 - 11:30

Night lights as direct observation of economic activity

Susan Thomas, XKDR Forum

11:00 - 11:15

Coffee break

11:15 - 11:45

Geospatial variation in prosperity: some evidence from satellite images

Carlos Mendes, Nagoya University

11:45 - 12:30

Satellite imagery as an input to credit access for low income households in agriculture

Debabrata Ghorai, Dvara E-registry

12:30 - 13:50


13:50 - 14:30

The Suomi NPP and Sentinel satellites

Ayush Patnaik, XKDR Forum

14:30 - 15:15

Machine learning and satellite imagery

Sudhir Silwal, Aquaconnect

15:15 - 16:05

Real time application of satellite data

Devleena Bhattacharjee, Numer8

16:05 - 17:00

Summing up

Ajay Shah, XKDR Forum

09:30 - 17:30

Hands on with satellite imagery

09:30 - 10:30

Introduction to Julia

10:30 - 11:30

Satellite image data

11:30 - 12:00

Coffee break

12:00 - 12:30

Coordinate reference systems

12:30 - 13:00

Problems with raw nighttime lights and data cleaning

13:00 - 14:00


14:00 - 16:00

Zonal statistics

16:00 - 15:30

Q & A