An Imaging Vegetation Status Monitoring System Mini Senior Design Project Submitted by Hector Erives August 31, 2005

1
An Imaging Vegetation Status Monitoring
SystemMini Senior Design ProjectSubmitted
byHector ErivesAugust 31, 2005
2
ObjectivesUse of a small computer-camera with
color filters arrangement to take images of
vegetation.Development of a system that will be
unattended for weeks. The system could be
powered by a battery or PV cells.The product
will be ratio images (2D maps) of a field, of the
NIR to Red bands.
3
Tools and MethodsA monochrome CCD camera with
Red and IR filters (or a wheel filter).
Standard CCD cameras 1024x1024 pixels, 12-bit
resolution (4096 DNs). Standard CCD camera
lenses. Red filter (650 nm CW), NIR filter (850
nm CW).A PC, Laptop, or a Single Board Computer
(SBC) with enough memory to save the images,
i.e. Image(1024x1024 pixels)x12 bits/pixel/8
bits/Byte1.5 MB. Required 2 images (Red,NIR)
of the scene x 3 images/day. Could save only
the Ratio to save memory. i.e. RatioNIR/Red
4
Tools and MethodsUse MATLAB to manipulate the
images, i.e. compute the ratioin a
pixel-by-pixel basis, i.e. The product will be
ratio images of 1024x1024 pixels.Can also write
a program in C, Visual Fortran, or IDL to
manipulate the images.Save the ratio images to
a file with a name so that it differentiates from
the rest of the images (from other times and
dates), i.e.ratio_940_9105, ratio_130_9105,
Could use a standard Nickel-Cadmium battery
(economic solution) 12 V _at_ 100 mA 4A. Can also
use Photo-Voltaic (PV) cells but have low
efficiency (lt 15).
5
Why use Red and NIR wavelengths?
Healthy vegetation
High DNs (Reflectivity 1)
Vegetation with lack of Nutrients (water,
fertilizer)
DNs or Reflectivity
Soil
Low DNs (Reflectivity 0)
UV(400nm) Red (650nm) NIR(850
nm)
Wavelengths (nm)