Maize field images
========================
We provide an image dataset composed of maize fiel images and their corresponding labelled images which were made by inspection and carefully hand painted. Images were captured with a single camera mounted on board a tractor, which is part of the fleet in the RHEA project: http://www.rhea-project.eu/

If you find  these images helpful, please cite it as:

Yerania Campos, Erik Rodner, Joachim Denzler, Humberto Sossa and Gonzalo Pajares. Vegetation segmentation in cornfield images using bag of words. Advanced Concepts for Intelligent Vision Systems (Acivs 2016). Oct. 24-27, 2016.  Lecce, Italy.


Additional Information
========================
Images were captured with a monocular colour-based camera mounted on-board a tractor during April/May/June 2012 to 2014. All the acquisitions were spaced by five/six days; and they were obtained under different illumination (high/low, shades) conditions, different growth states for plants. 
The machine vision system was arranged as specified in Emmi et al. (2014). It was located at 2m height from the ground and inclined 45º with respect to the ground. The camera-based sensor is the SVS4050CFLGEA model from SVS-VISTEK, built with the CCD Kodak KAI 04050M/C sensor with a GR Bayer color filter; its resolution is 2336 (Horizontal) by 1752 (Vertical) pixels with a 5.5 by 5.5 ?m pixel size.  This camera is Gigabit Ethernet compliant. The digital images were captured under perspective projection and stored as 24-bit raw color images in the RGB color space.

Emmi, L., Gonzalez-de-Soto, M., Pajares, G., & Gonzalez-de-Santos, P. (2014). Integrating sensory/actuation systems in agricultural vehicles. Sensors, 14(3), 4014-4049.

For any questions and comments, please email: yeraniac@ucm.es


Acknowledgments
========================
This work was supported in part by:
The European Unions Seventh Frame- work Programme (FP7/2007-2013) under Grant Agreement No. 245986. 
CONACyT under call: Frontiers of Science (grant number 65) for the economic support. 
CONACyT for the doctoral grant number 210282 to undertake doctoral studies. 

Last update: August 19, 2016.