Robotics for precision agriculture

In order to determine the health status of the crop, the Normalized Difference Vegetation Index (NDVI) is computed in a developed software. Once the aerial imagery is obtained by the drone, it is processed to eliminate image distortions and insert specific metadata needed for generating the orthomosaics with the health information of the plant or soil of interest. Below some images obtained with a camera-aerial system developed at LAB.

Near-infrared reflectance orthomosaic.

Visible reflectance orthomosaic.

Near-infrared reflectance orthomosaic

According to the color map in below, red hues are related to very healthy vegetation. Comparing this image with both orthomosaics in Fig. at left, the red spots are generated due to the shadows of trees giving zero information about vegetation. There are also yellow and green areas that may contain healthy grass and canopy of surrounded trees. Gray hues indicates unhealthy vegetation. It is interesting that some trees appear to be dry but, in the above figure these trees reflect near-infrared radiation. This means that they are actually healthy. This can be verified at left showing trees with high NDVI values (green and yellow).

Drone + multispectral camera

We have developed a quadrotor and a low-cost multispectral camera for conducting the aforementioned experiments.

The drone includes:

Low-cost multispectral camera
GPS
Autopilot
Raspberry Pi 3
Telemetry for communications

Second prototype

This second version in enhanced with two multi-spectral cameras: RedEdge-M and a camera rig fabricated at LAB. Also its payload capabilities have been increased.

Students

MSc. Andrés Montes de Oca Rebolledo. Master thesis: Construction and control of an aerial vehicle for crop monitoring. Centro de Investigaciones en Óptica, A. C; Maestría en Optomecatrónica. July 2018.

Related publications

Low-cost multispectral imaging system for crop monitoring; A. Montes de Oca ; L. Arreola ; A. Flores ; J. Sanchez ; G. Flores; 2018 International Conference on Unmanned Aircraft Systems (ICUAS); Dallas, TX, USA; 12-15 June 2018; pp. 443 - 451.

One journal paper in preparation.

Google Sites

Report abuse