Söderström, Mats
- Institutionen för mark och miljö, Sveriges lantbruksuniversitet
Proximal sensing refers to measurements using sensors in close proximity to the object of interest (Adamchuk et al. 2018). Proximal crop sensors are used to collect information about a growing crop and can be mounted on the ground, handheld or borne by vehicles such as tractors or robots. Remote sensing involves the measurement of crop properties often with similar equipment as used in proximal sensing, but from a greater distance, using satellites, airplanes or unmanned aerial vehicles (UAVs, drones). The latter may also be used at short distances. At the other end of the spatial scale, there are sensors that can be used very close to, or in contact with, plant parts. Hence, there is a wide range of different types of sensors and scales on which they are used. Different sensors can also be used in combination, e.g. a proximal sensor can be used to calibrate data collected by remote sensing (Fig. 1).In most cases, crop sensors collect inferential data, i.e. they do not directly measure crop properties of interest but rather produce a metric that can be used to estimate these properties. For example, light in different wavelengths reflected by a crop canopy can be recorded by a crop sensor and translated into useful information for agricultural management using established empirical relationships with e.g. the protein content of the crop.Proximal crop sensors are used to assess and predict a range of different crop conditions, such as nutrient status, incidence of weeds and diseases, and drought stress in plants. The ripening stage of fruits and even number of spikes in a wheat stand can also be determined from digital images. Rapid technological development and access to artificial intelligence and machine learning methods have enabled new applications that were not possible just a few years ago, while still employing sensor techniques that have been used for a number of years.
Titel: Precision agriculture for sustainability
Utgivare: Burleigh Dodds
Jordbruksvetenskap
Jordobservationsteknik
https://res.slu.se/id/publ/144604