How to determine the acres devoted for crop type

s1tbx
#1

I did crop classification, corn and soybeans, from sentinel 1 data for study case and I would like to determine the acres devoted to corn and what is for soybeans. Could I do this using SNAP or QGIS?

#2

Hi,

If you have a vector layer ( shapefile) for corn and soybeans areas, then its very easy to calculate the area of those shapefile in QGIS

Have a look at this video: https://www.youtube.com/watch?v=dKykspe3uDc

#3

Hi Ghazal,

The work you’re doing sounds very interesting. May I ask which country is this classification being done?

I’m trying to do a similar classification for India (Issues such as very small farms, mixed cropping, multi-cropping, etc.) using SNAPPy and Sentinel-1 and would like to know more about the methods of classification you’re using.

Thank you.

Regards,
Adhithya

#4

I used Random forest for the classification of my case study in USA, Kossuth county. Furthermore since you are from India, I would like to help me to have polygon for Kerala in order to use it as study area to extract data from sentinel 1 if possible

#5

bu the output that I have from SNAP, the classification product, is raster and I want to do calculation of the area for every crop type.

#6

convert your raster to polygons and calculate the size of the polygons in a new attribute column.

#7

In order to calculate the area of some parts on your raster, you can follow two ways:

  • Vectorize your raster as mentioned by ABraun
  • Digitize those parts of the raster you want to calculate the area from. Once you have the digitized polygons, you can calculate the area.
#8

IDRISI has a tool to directly assess the size of pixel aggregates of the same value but I don’t see this in any free GIS

#9

Hi Ghazal!

Thank you.

You could download shapefiles of various states and districts in India from this website: http://projects.datameet.org/maps/

1 Like
#10

Hi adhithya

I would like to do crop classification over India and I want to know if possible how to get vector training for any area in India.

Thanks in advance.

#11

It’s extremely hard doing crop classification in India. Especially South India. The reasons being:

  1. Mixed cropping
  2. Multi cropping
  3. Small farm sizes (Most marginal farmers have holdings of less than 30 m x 30 m)
  4. Finally, It’s nearly impossible to find training data online (I haven’t come across any till now). Without actually being on a field trip to collect the ground data points I was never able to do classification.

Hope this answers your question. Apologies for being overly negative.

1 Like
#12

have you manged to do the crop classification despite the heterogeneity of pixels? how did you manage to do it. I’m interested because i’m doing a project on the same area facing challenges of mixed cropping and small farm sizes