Analysing on finer resolution-based data

[Bondopadhyay]

Hey, Killian. Firslty, a superb package that you built.

Two queries + one issue:

1. Is it suppose if I want to discretely work on images with finer resolution, such from the WorldView2/3, Quickbird etc sensors, could that be possible? I won't be able to fetch from cloud as these are paid ones but I could have these on my local system. I have downloaded a few (not expecting any good time-series analysis) and I want to use those for CoastSat.

2. Is ti possible to use CoastSat on Google Colab or likewise cloud-based platforms?

3. For the date, I am willing to use a range with a certian number(s) of years as interval. While the function is not causing any issues, the inputs that are being being passed to the SDS_download.retrieve_images function or how it is handling the dates internally. The loop is not being taken after 2 years. Suppose if it is from 2005, the loop is being off after 2008.

def generate_date_range(start_year, end_year, interval_years):
    start_date = datetime(start_year, 6, 1)  
    end_date = datetime(end_year, 8, 15)  
    date_range = []

    while start_date <= end_date:
        print("Current start_date:", start_date)
        date_range.append(start_date.strftime('%Y-%m-%d'))
        
        start_date = datetime(start_date.year + interval_years, 6, 1)

    return date_range

If I am being mistaken somewhere, shed light.

Thank you.

[kvos]

hey @Bondopadhyay , the main problem you will face when working with the higher res is that the bands change (no more short-wave infrared). As CoastSat uses the short-wave infrared for delineating the shoreline, you will need to detect it differently. We had a go at this with Yarran Doherty on Maxar images about 5 years ago, just as a test, you could fork and continue there: https://github.com/kvos/CoastSat.Maxar.
All the best