You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Is your feature request related to a problem? Please describe.
I want to be able to get the screen grab of visible pixels in the layer and pass it for let's say Machine Learning Inference. It would be nice if there a was direct toDataUrl giving all the visible pixels.
Currently, it looks very difficult to obtain, slice and stitch values from getActiveTiles.
Describe the solution you'd like
A toDataUrl or getPixelArray kind of function to give me the values. I'm relatively new to JS, but seems if pixels are visible on Canvas, they might as well be extracted.
Describe alternatives you've considered
Solution I am trying currently is to use getActiveTiles, slice to canvas extent and concatenate all the tiles.
An alternative is to use getValues, but that too would need some stitching.
Another alternative, not very relevant here, is I'm planning to use rasterio based python service in the backend, which would bring unacceptable latency.
Additional context
The text was updated successfully, but these errors were encountered:
Is your feature request related to a problem? Please describe.
I want to be able to get the screen grab of visible pixels in the layer and pass it for let's say Machine Learning Inference. It would be nice if there a was direct toDataUrl giving all the visible pixels.
Currently, it looks very difficult to obtain, slice and stitch values from getActiveTiles.
Describe the solution you'd like
A toDataUrl or getPixelArray kind of function to give me the values. I'm relatively new to JS, but seems if pixels are visible on Canvas, they might as well be extracted.
Describe alternatives you've considered
Solution I am trying currently is to use getActiveTiles, slice to canvas extent and concatenate all the tiles.
An alternative is to use getValues, but that too would need some stitching.
Another alternative, not very relevant here, is I'm planning to use rasterio based python service in the backend, which would bring unacceptable latency.
Additional context
The text was updated successfully, but these errors were encountered: