Esri / raster-deep-learning

ArcGIS built-in python raster functions for deep learning to get you started fast.
Apache License 2.0
186 stars 87 forks source link

OSError: Cannot identify image file #28

Closed daspk04 closed 4 years ago

daspk04 commented 4 years ago

Hello,

Currently, I am trying to train the U-Net Model for the classification of land cover. The input image is a 16-bit unsigned integer Tiff file of 4 bands (Blue, Green, Red and NIR). The training data are in vector format which I converted to unsigned int 8 bit Tiff as well as Exported the Traning data for Deep Learning as "Classified Tiles" per Instructions shown in the link. (Extracting Building Footprint from Drone Data). But I'm stuck in data preparation it gives an error as "Cannot Identify Image File". Could you suggest how can I resolve this issue?

Error

sang4758 commented 4 years ago

Hi, The training works on 8 bit 3 band data. You would need to apply a Extract Band function/tool to the input image to make it 3 bands before exporting and training.

daspk04 commented 4 years ago

Thank you. It did work after I converted them to 8 bit.

abinprajapati commented 1 year ago

I get same error: OSError: Cannot identify image file. I have converted to 8 bit but this couldnt resolve this problem. Can you help me in this regard.