Deep learning Workshop for Satellite Imagery - Data Processing (Part 1/3)

why did after u did patched_image.shape,it showed (2,1,1,256,256,3),i understood the latter but i didn get the 2,1,1 ,can anyone please help?

I am a complete beginner when it comes to all things deep learning. Can you go over some basic steps about how to apply the model to new images? Thanks.

Thanks for clearing my basic on how to process images, mask and associate labels with them. @Prodramp do you also have any tutorial on how to do unsupervised learning on hyperspectral images ? Unsupervised modeling on satellite images will help me greatly because I have some hyperspectral images but they are not labelled.

can i get the shape files so that i can access it in QGIS .

Hello, I am a student from China, and my English level is not very good. So there is a question when listening to your lecture. Why do you need to convert BGR to RGB for masks images, but not for jpg images? This is translated by translation software. If the words or tone are not very good, I hope you will forgive me. Thank you for your selfless sharing!

Hello sir, how to find the height and weight of the road from satellite images
In deep learning?

Thank you for your amazing lessons. I learned a lot!

Instead of using Dubai Dataset can i use Sentinel-2 3 months composit image for classification? can yoou share a video how to do it?

Can you please tell me why did we only convert mask images from B G R to R G B. And also why not we normalize masked images values from 0 to 1

Hey Avkash, Can you please explain why you are patching each image?

Why can't I find you till now? You are amazing. Thank you very much.

Nice. But you spent a lot of time on explaining the file system and how to manipulate that in Python unnecessarily. This confuses and deviates the focus of this lecture.

is this multispectral analysis ? cuz all i saw is RGB

Hello sir thank you so much for this video series, youre doing a great service for the society.
Is it possible for you to make a video on semantic change detection of satellite images using unet where we implement a dissimilarity decoder ?

hello, your video is very useful. I am still a newbie with neural networks but I learnt that for training the model, we need a great number of labelled datasets which I am currently missing. I do have access to Lidar and Hyperspectral imagery covering a specific area and was wondering if its a good idea to use CORINE land cover classes as labels and then make prediction of LCC using Lidar and Hyperspectral and maybe SAR? or the quality of training data (labels) could be too bad? Corine LC2018 is made from Sentinel and Landsat8. Kind regards

Sir how to get into the google colab please inform me

👌🏻 great

Hello Dear Chauhan, i watch your tutorials about Deep learning but i want to know how to create dataset using Landsat imagery or Sentinel-2. Especially how to create mask and label of image Tile. I create tiles using Geotile. Thanks for your help

Great job! Thank you so much for this very helpful content.

I have to work on Tiff file formats, which is very confusing can you please tell me how I can use tiff format for all this workshop?

I am starting my thesis by creating a model that can predict LULC maps from a particular area from time series of the exact location in the past that can be used to train deep learning models that generate future developments in advance. I hope this Workshop will be of great help..thanks in advance

you are one of the few influencer who are really adding values to the community.

thank you so much for sharing this method with !

clear guide worked well with for sharing.

BROOO thankyou so much, this really helped and the tutorial was really easy to use as well :)

god bless u xdd

My training result shows a linier increasing of validation loss on epoch 50 to 100, but the training loss is decreasing through the training epoch, is it a sign of overfitting? and how do I overcome it? Note that I only used 1 dataset from Kagge, I haven't tried using 38GB satellite data (due to lack of storage). Thank you for the tutorial, waiting for the next part!!.

Thanks for creating such an amazing tutorial, waiting for the next part.

you are a Legend

I am really 💪😲🤳 excited for this project! ❤