UBFC Logo

Large bande Multi-Spectral data based 400-1100 nm:

Application for Precision Agriculture


Main Image

The multispectral image database of the University of Bourgogne Franche-Comté (UBFC) is an essential resource for researchers and students working in the field of image analysis and optical signal processing. It contains multispectral images captured under different lighting conditions and at different wavelengths, offering a wide range of potential applications from material analysis to remote sensing and medical diagnosis.

Image Capture Information


m77 m82 m85 m91

8-band multispectral mosaic image


Label 613 Label 649 Label 676 Label 724

Labeled image

Download one sample of four image types

Camera Multispectral of project CAVIAR
Resolution 1080 x 1024 pixels
Illuminant Halogenes/ Daylight 5000-7000K
Filter VariSpec liquid crystal tunable filter
Wavelength Range 400 nm - 720 nm
780 nm - 1000 nm
Number of Bands 8 VIS
8 NIR
Aperture f/1.4
Focus Fixed
Format TIFF

NOTE: These multispectral images represent the reflectances of plants and weeds present in the scene. The reflectance spectrum of each pixel was calculated from the multispectral image measured using calibrated estimates of the illumination spectrum and spectral response of the cameras in the visible and near infrared.


Database Organization


The database consists of plant and weed scenes. Each scene has full spectral resolution in 10nm increments (a total of 8 bands) and in the near infrared. Each band is stored as a 16-bit grayscale TIFF image.Each scene also includes a representative color image displayed using sRGB values rendered under neutral daylight illumination (D65).


Contact


The multispectral image database at the University of Bourgogne Franche-Comté (UBFC) is an essential resource for researchers and students working in the field of image analysis and optical signal processing. It contains multispectral images captured under different lighting conditions and at different wavelengths, offering a wide range of potential applications from materials analysis to remote sensing and diagnostics for plant and weed separation.



Email: pgouton@u-bourgogne.fr
Phone: +33-782-194-200