Generate image & mask tiles for deep learning | DL

GeoTile package: github.com/iamtekson/geotile
Source code here: github.com/iamtekson/deep-lea...
Timestamps:
0:00 Intro
0:42 Environment setup
2:57 Intro to geotile package
4:47 Environment setup cntd...
5:48 Import package and initialize GeoTile
7:38 Generate training tiles and mask tiles
15:08 Generate test imagery tiles and mask tiles
18:23 Preprocessing (remove nan, normalization) and Save it as numpy array
----------------------------------------------------------------------------------------
Here are some playlists that you might be interested in:
----------------------------------------------------------------------------------------
1. Leaflet from basic to advance: • Leaflet from basic to ...
2. GeoServer with leaflet web-GIS: • GeoServer and Leaflet ...
3. GeoDjango project: • Bookmark note on map |...
4. geoserver-rest: • geoserver-rest
5. GIS training in Nepali: • Beginner GIS training ...
6. LULC map production: • LandUse Land Cover Map...
7. Geospatial analysis with python: • GeoSpatial analysis wi...
8. GeoNode from basic to advance: • GeoNode from basic to ...
9. Machine learning from the basic: • Machine Learning for g...
--------------------------------------------------------------------------------------------------------------
Check out my discounted courses at the following link:
--------------------------------------------------------------------------------------------------------------
1. "Geospatial data analysis with python": www.udemy.com/course/geospati...
2. "Web GIS Development 2021": www.udemy.com/course/web-gis-...
3. "Web mapping and Web-GIS from Dev to Deploy 2021: GeoDjango": www.udemy.com/course/web-mapp...
4. "Introduction to Web Mapping and Web GIS 2020: GeoDjango": www.udemy.com/course/introduc...
5. Deep Learning Application for Earth Observation: www.udemy.com/course/deep-lea...
6. Geospatial Data Analysis with Python: www.udemy.com/course/geospati...
---------------------------------------------------------------------------------------------------
Follow me on GitHub: github.com/iamtekson
Follow me on Twitter: / iamtekson
Follow me on Instagram: / iamtekson
Join Facebook group: / 1049524025731229
-----------------------------------------------------------------------------------------------------

[@mohamadroviansah808]

I found this really helpful thank you very much

[@geodev]

Glad to hear it!

[@user-tr7cu7tm5l]

thanks very much!

[@HiepCwalk]

Thank you for very useful lesson

[@geodev]

You are welcome !

[@mandata143]

Thank you

[@geodev]

You are welcome!

[@vimalshrivastava6586]

Thank you for this informative video. I have a query. Is there any way to generate a mask without a shape file?

[@geodev]

If you are planning to use geotile package, you can use any type of vector layers such as geojson, shp etc. But there are other simpler ways too. One thing is, you can use the threshold based on some indices such as NDWI, NDBI MNDWI and create the mask file. But definitely it will also be less accurate label.

[@vimalshrivastava6586]

@@geodev Thank you for the quick response. I am using QGIS software to create mask using semi supervised method. But, as you mentioned it is less accurate. Is it possible to get shape file of specific region for multiclass task?

[@muhammadhammadbashir8989]

when we get data, in geotiff, for a marked area of our choice problem is its resolution are very small, I tried different way, but I want like 50 , 60 mb of file with high pixel values and greater quality

[@geodev]

You mean while downloading the dataset from GEE?

[@muhammadhammadbashir8989]

@@geodev I have trained a model on public datasets like land cover which is detecting the segmenting the greenery, now I want to do this thing for another city let say Pokhara, but when I download images from sentiental 2 as you mentioned images quality are too bad, so thats why I want high size image, and yes its from GEE

[@user-dx9cf6zt2d]

thanks a slot, but i cannot install geotile packages :(

[@geodev]

Have you tried with conda? I suggest to run following commands; 1. conda install -c conda-forge geopandas 2. conda install -c conda-forge rasterio 3. conda install -c conda-forge geotile

[@user-dx9cf6zt2d]

@@geodev yes now iam done, thank u sir so much. iam logging 3 account to subcri your chanel :D

[@user-dx9cf6zt2d]

@@geodev I have one more question, can I apply this method to classify multiple objects?

[@user-dx9cf6zt2d]

etc building, water and forest