Plant Count for Production Field

This analytic automatically detects and counts plants and gaps.

1. Inputs and parameters

  • Scouting maps (NDVI by default) * (Scouting Maps)‍ 
  • Field boundaries * (How to create a Field Boundaries‍)
  • Row vectorization (if not available in input, it will be calculated automatically)
  • Theoretical Inter plants spacing *
  • Row spacing *
  • Deliverables suffix

*Mandatory

Delete
  • It is mandatory to see bare ground in between two plants on the map, otherwise only visible gaps will be detected.
  • For row crops, with 6cm of GSD and a multispectral camera, optimal corn stage will be 4 leaves, and starbud for sunflower.
  • With enhanced resolution, earlier stages can be targeted. The basic rule is : size of GSD <= half canopy size (from above)
  • For orchards and vineyards plants must be planted in rows. If canopy is closed, only gaps will be detected. Height of the plant must be > 50cm.
  • For palm trees, plant must be > 50 cm as well.

Sensor required:

  • Multispectral
  • RGB

2. Deliverables

  • Gaps + suffix.geojson
  • Rows with plant count + suffix.geojson
  • Plant count + suffix.geojson
  • Gaps + suffix.csv
  • Rows with plant count + suffix.csv
  • Plant count trial + suffix.csv
Delete
The deliverables can be exported in another file extension with the export functionality

3. Attributes

The geojson and the corresponding csv file have the same attributes.

“Gaps” output - only for plants in rows with the following attributes:

  • row_index: row id
  • gap_id: id of the gap
  • gap_length: Length of the gap in meter

Plant count” output with the following attributes per field:

  • gap_length: cumulated gaps length in the field in meter
  • plant_length: cumulated plant length in the field in meter
  • plant_count: number of plants in the field
  • row_count: number of row in the field

Rows with plant count” output with the following attributes per rows:

  • row_length: row length in meter
  • plant_length: cumulated plant length in the row in meter
  • plant_count: number of plants in the row in meter
  • gap_length: cumulated gaps length in the row in meter
  • veg_ratio: plant_length/row_length, between 0 and 1

4. Workflow

Step 1:

  • Select the vegetation indice that will be used to process the Plant and Gap counting (by default : NDVI is selected)
    Note: for RGB data sets, select VARI (the VARI indice has to be previously generated)
  • Select field boundaries*
  • Select the rows (optional / if already vectorized)

*field boundaries needs to be a vector file. You can obtain this vector in drawing an annotation which represents the field boundaries and convert it into a vector.

Step 2 :

  • Fill the inter plant spacing (distance between 2 plants within the same row)
  • Fill in the row spacing (distance between 2 rows)
  • Click on Launch

5. How to use it

From layers panel, select Inventory layer from SURVEY DATA section and select between plant count or gaps layers to display :

  • If you display the Gap layer, it is better to also display the RGB map to locate where is the gaps. Gaps are represented by small dots and depending on the gap length, a specific color is displayed and associated to the legend on the right panel. The gap length can also be displayed if you click on the dot.

  • If you display the Plant count layer, the vectorized field will appear. You can click on the field on the map to display its plant number.


Learn more: Vector Layers Styling‍