Skip to content

Perform Thresholding for LandSat-8 Images #14

@Chintan2108

Description

@Chintan2108
  • Right now, a simple band math formula has been applied to the input image.

  • This outputs a grayscale image - the pixels having higher probability of having the feature of interest have a high value (0-255) and hence appear towards the brighter side.

  • Perform thresholding, decide a threshold value, in order to convert the output grayscale image into a binary feature map

  • Use data available as a part of PR added Landsat 8 NDWI indices notebook #9

  • For eg: If the threshold is decided as 200, then modify the output image such that all pixels having values greater than or equal to 200 will now have a value of 1, and other pixels will have a value of 0. This will result in a binary image, in which the bright pixels would represent features.

Metadata

Metadata

Assignees

Labels

SLoP-75SLoP Level 3 IssueenhancementNew feature or request

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions