FarmArk : Image Recognition to Detect Weed in Crops

[prathimacode-hub]

Hey Hacktoberfest participants, good to see you.

This issue's goal is to get a dataset with sufficient parameters of different kinds of crops, then apply image recognition using ML algorithms to detect weed in crops

How to get it approved?

• Add the details of how you had done and how much accuracy you're attaining then wait for one of the mentors to get back to you.
• Once it's approved, go ahead and make pr.

Please read CONTRIBUTING.md at least once before contributing code.

[bansal247]

Hi, I would like to work on this issue. My approach will be:

1. Indian weeds' names are given at https://www.fao.org/agriculture/crops/core-themes/theme/biodiversity/weeds/db-countries/i/en/#india so I will first scrap images from google of all the weeds
2. I will then merge databases from other websites like

• https://data.mendeley.com/datasets/nj4vtk4tt6/1
• https://www.kaggle.com/datasets/ravirajsinh45/crop-and-weed-detection-data-with-bounding-boxes
• and if found, from other websites also

3. after creating the database I will clean the database and increase the database through various transformation techniques.
4. I will create then try different methods and models to classify the weeds
5. After selecting the best model I will start development in a modular fashion with proper logging and develop an API where an Image will be posted and annotated image with classification will return.

[prathimacode-hub]

You can proceed with issue. Issue assigned to @bansal247

[prathimacode-hub]

Update please? @bansal247

[prathimacode-hub]

Update please? @bansal247