Questions about L4A from a newbie

Crop type map (L4A) processor
1

Hi Sen4CAP team,

For this post, I focus on L4A product, especially the predicted crop type. Here are my questions.

  1. Do you have any video showing how to create a site, prepare LUT and shape file with the 3 attributes and run the L4A product?
  2. If I want to know the predicted crop type of a parcel in Thailand, do you think L4A can help me on that?
  3. What is the minimum parcels required in order to get the predicted crop type?
  4. After I enable the site, when will I get the results of L4A product?
  5. If a parcel of 2 sites are duplicated, do we still get the L4A product (focus on predicted crop type) for both sites?

Regards,
Ant

2

Hello Ant,

  1. The online training presentations explain how to do so: http://esa-sen4cap.org/sites/default/files/02_Sen4CAP_TrainingforBeginners_SystemOperation.pdf + the videos of this training will put online soon.
  2. Yes, but you need to update in the system the parcels shp with a crop code for each parcel. The processor was developed to check if the crop code declared in the shp is confirmed by the satellite data. Still, if you don’t have a parcels shp with a crop code, you can use the system for the preprocessing of the satellite data (S1, S2 and L8) and apply a pixel-based classification of your own to these data (unsupervised or supervised classification with training samples). Or you could decide to perform a segmentation to extract the parcels limits, select training samples and work on the classification script to use the training samples for the training of the RF model, and apply it to all the parcels.
  3. See point 2. But in a sense, there is no minimum parcels required, everything is configurable.
  4. I am not sure to totally understand the question but the L4A is run on a specific site, indepently of the other sites.

Best regards,

Philippe

3

@Philippe_Malcorps

Thank you for your reply. I have more questions.

  1. After creating a site with a shape file (a parcel and declared crop code), when will we get the results of L4A?
  2. In CropType.csv result file, do we expect to get the “predicted crop type” in the file?

Regards,
Ant

4

@Philippe_Malcorps

Let’s say I have a newly installed Sen4CAP and I create a site with a shape file (with a parcel and declared crop code). If I choose to run L4A processor, will I get the result of Predicted Crop Type?

What is “pa-min” in default configuration of L4A?

Regards,
Ant

5

Hello @chananartc,

The L4A crop type processor was developped to check the reliability of the crop type declarations. So the user imports all the parcels that she/he wants to check. Then, the processor will use part of the parcels to calibrate the classification model using the preprocessed EO data (by default 25% of the parcels for the main crop types) and will apply the classification model to all the parcels to predict the crop type for each parcel based only on the EO data. So it was developped to ingest a large number of parcels with crop declarations and check the reliability of these crop declarations (CAP context). Still, you can use it as you want, for example to extract markers by parcel and then make your own classification process. Or to build a classification model based on the parcels with crop declarations that you have and then apply this classification model to more parcels (but this will need some work from you on the script). So if you upload only one parcel, the L4A processor cannot be used.

For the “pa-min” parameter: it is the minimum number of parcels by crop type for the crop type to be included in the classification (after selection of parcels with min S1 and S2 pix and min S2 pix for the calibration selection). It was defined to avoid to classify very rare crop types that are so few compared to the other crop types.

Best regards,

Philippe

6

@Philippe_Malcorps

Thank you for the detailed reply. I see.

Regards,
Ant