Skip to content
English
  • There are no suggestions because the search field is empty.

Weed detection best practices - PIX4Dfields

Weed detection ranks among the most widely adopted drone applications in agriculture. The optimal imagery and resolution depends on the specific weed species present and the timing within the growing season.

Fallow weed detection

When weeds are detected in a fallow field, vegetation indices are generally used to differentiate the green weeds from the bare soil or previous crop residue. These indices can be derived from both RGB and multispectral cameras.

Image resolution/flight height

Orthomosaics with a GSD between 2 and 3 cm are ideal for fallow weed detection.  If weeds occur in larger patches or zones, a lower resolution of GSD between 4 and 8 cm is generally acceptable.

RGB vs. Multispectral in fallow weed detection

At the same flight altitude, multispectral images, even with lower resolution, help to better differentiate weeds. 

  • Multispectral images are superior to RGB for detecting green weeds in fallow fields, especially when using the NDVI and OSAVI indices. 
  • RGB-based methods (VARI, TGI) are less reliable due to "noise," but are much more practical, making multispectral data preferable for accurate weed identification
An example shows how OSAVI detects a 13 cm diameter weed (on the left) while TGI does not detect it (on the right):

GSD MSP: 5 cm/pixel.
GSD RGB: 3 cm/pixel.
2_OSAVIvsTGI


Examples:

Green-on-Green

Recent research indicates that high-resolution RGB cameras are particularly effective for green-on-green weed detection. Integrating RGB imagery with advanced identification tools like the Magic Tool enables reliable detection of weeds growing within crops, pastures, and other vegetative environments.

For accurate detection, one must fly when the weeds are visually distinct from the crop (e.g., by color, height, and diameter).

The Magic Tool provides identification for all weeds that are visually different from the crop, but it does not classify them by type. Satellite imagery from Sentinel-2 is discarded for weed identification due to its low resolution (10 meters per pixel).

Examples:

Recommended camera and resolutions 

  Fallow  Green on green
Crop stalls & residues Plowed field Distinct color difference between crops and weeds. Minimal color difference between crops and weeds.
Camera MSP RGB RGB RGB
Resolution Around 3cm.
The smaller weed diameter should be covered by 3 pixels.
Around 2cm.
The smaller weed diameter should be covered by 3 pixels.
3 to 5 cm. 2 cm or less.