Filtering Imagery¶
To extract the best shorelines, it's essential to use only high-quality imagery, which means filtering out any poor-quality images. You should filter out bad imagery in data/roi_id/jpg_files/preprocessed/RGB
by moving any undesirable images to a designated subdirectory named 'bad'.
Filtering the RGB images does not delete any TIFF files but helps to improve the efficiency of shoreline extraction and enhance the quality of the extracted shorelines. For more information on the shoreline extraction process, check out the wiki page about the shoreline extraction process.
Step-by-Step Guide¶
Before You Begin¶
Download data with the SDS_coastsat_classifier.ipynb
and ensure it is saved to your data
directory within CoastSeg.
1. Open the data
Directory¶
- Navigate to the
data
directory within CoastSeg. - Click on the ROI directory you want to sort
In this example, you can see that ROI 3 is located within the CoastSeg/data
folder. The jpg_files > preprocessed > RGB
directory contains all the RGB imagery for all the satellites.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 |
|
2. Navigate to the RGB Directory¶
- Go to
jpg_files > preprocessed > RGB
within the ROI directory. - Example on Windows:
CoastSeg\data\ID_yvk1_datetime06-05-23__06_57_26\jpg_files\preprocessed\RGB
3. Move Bad Imagery¶
- Create a subdirectory named 'bad' within the
data/roi_id/jpg_files/preprocessed/RGB
directory if it does not already exist. - Identify the images you want to remove.
- Move these images to the 'bad' subdirectory.
4. Test Shoreline Extraction¶
Load your ROIs in CoastSeg and run extract shorelines
again. None of the images that were sorted into the 'bad' directory will have their shorelines extracted. For more details, refer to the guide How to Extract Shorelines.