Micro-MCA

Recommendations for

Band Pass Filter Selection

 
ADC ADC Lite ADC Micro ADC Snap -MCA RGB+3 Single Band 
 
 

 

The cameras in Micro-MCA systems each contain a camera lens, a changeable narrow-band filter and a 1.3 mega-pixel CMOS sensor.  Users select filters to insert into each camera between the lens and the sensor in order to restrict the radiation that contacts the sensor to a narrow band of wavelengths.

The combination of filters selected by the user enables the system to sense a unique spectral signature.  This can be used to identify one or more plants, plant conditions or other compounds.

 
 

 

TECHNICAL PARAMETERS AFFECTING MICRO-MCA BAND-PASS FILTER SELECTION

Electric current is developed in an image sensor in response to the electromagnetic radiation that contacts it.  The amount of current generated depends upon the brightness and wavelength of the radiation that hits the sensor.  For a given brightness, Micro-MCA output is greatest in response to wavelengths in the area of 800nm. The output drops in a smooth curve to an approximate 20% peak output at 450 nm in the visible spectrum and 1050 nm in the near-infrared at the limits of its range.  A graph of the sensitivity of the camera's image sensor to impinging radiation is shown below. 

The spectroscopic filters used in Micro-MCA systems are supplied by Andover Corporation. These are placed between each Micro-MCA camera's lens and sensor.  The filters restrict the radiation that is able to contact the sensor to a narrow band of wavelengths.   Andover filter designations indicate the filter's center wavelength, band width and the size of the aperture they fit.  All Micro-MCA systems use filters with a 25 mm diameter.

The amount of light that is allowed to hit the sensor increases as filter bandwidth increases.  As a rule, 10 nm bandwidths provide optimum multi-spectral imagery in Micro-MCA systems.  Wider bandwidths tend to over saturate the sensor.  The exceptions to this occur at the limits of the sensor's sensitivity range.  Here, the greater amount of light that broader bandwidth filters pass is offset by the sensor's decreased sensitivity.

ORDERING AND REPLACING Micro-MCA BAND-PASS FILTERS

Micro-MCA filters are customer-designated at the time of order.  The filters are field-replaceable using standard 25mm spectrometer filters so these systems may be reconfigured in the field to sense entirely new spectral combinations.  Field replacement of filters is limited by the following considerations:

Since wavelength impacts focus, Micro-MCAs are initially set up in the factory for optimum focus based on the wavelengths of the filters initially selected by the user.  Changing filters in the field, changes the camera's focal point.  To insure the Micro-MCAs cameras provide optimum focus, users should only replace a Micro-MCA filter that is near the wavelength of the filter originally inserted into that camera. 

In the visible spectrum, replacing a filter with a new one that is within plus or minus 100 nm of the original is acceptable.  Beyond 700 nm, replacement filters should be within plus or minus 50 nm of the original.  Replacement of filters beyond this range requires factory re-focusing of the unit.

SELECTING OPTIMUM MICRO-MCA BAND-PASS FILTERS FOR YOUR APPLICATION

Both Andover's standard and custom filters may be used for Micro-MCA applications.  The filters we select to include in our standard filter sets are chosen based on their ability to expose conditions able to be derived from vegetation spectral reflectance curves such as the one shown below.

 The selection criteria we use are based on the following factors:

We select filters for our standard filter set from Andover's list of Standard Filters.  This is available on the Andover Corporation website.  Customers may pick any filter from Andover's Standard List to substitute for a filter in our Standard Filter Set with no change in the product's price.  The price of custom filters are quoted at the time of their order. 

Below is a list of filters in the Standard Filter Set for each Micro-MCA model with a description of each filter's uses

Micro-MCA STANDARD FILTERS AND THEIR USES

Micro-MCA

Model Number

Filter Designation

 Reason for Inclusion in Standard Set (Filter Uses or Significance)

Micro-MCA4    
- 490FS10-25 Blue - Ten nanometer slice of Landsat 5 TM Band 1,  This provides increased penetration of water bodies and also is capable of differentiating soil and rock surfaces from vegetation and for detecting cultural features.1  This is the point of crop to soil reflectance ratio Microma for blue and green bands. The band is sensitive to loss of chlorophyll, browning, ripening, senescing, and soil background effects). It is also sensitive to senescing rates and is generally an excellent predictor of grain yield.2
- 550FS10-25 Green - Ten nanometer slice of Landsat 5 TM Band 2, it is sensitive to water turbidity differences.1  Positive change in reflectance per unit change in wavelength of this visible spectrum is maximum around this band. The so called "green hump" in vegetation spectral reflectance curves is useful in predicting chlorophyll content.
- 680FS10-25 Red - Ten nanometer slice of Landsat 5 TM Band 3,  this may be used in the derivation of the Normalized Difference Vegetation Index where NDVI=(TM4-TM3)/(TM4+TM3).  Cited for NDVI measurement by ENVI with 800 nm 9  Absorption in the red band (600 to 700 nm) varies significantly due to changes in factors such as biomass, LAI soil background, cultivar types, canopy structure, nitrogen, moisture, and stress in plants 2   Note:  Users may substitute a 660 nm filter with designation of 660FS10-25 for the 680 nm designated 680FS10-25.  The 660 nm filter is a common choice among Tetracam customers.  This is the chlorophyll absorption pre-maxima (or reflectance Microma). Absorption in the red band (600 to 700 nm) varies significantly due to changes in factors such as biomass, LAI soil background, cultivar types, canopy structure, nitrogen, moisture, and stress in plants
- 800FS10-25 NIR - Ten nanometer slice of Landsat 5 TM Band 4, this may be used in derivation of Normalized Difference Vegetation Index where NDVI=(TM4-TM3)/(TM4+TM3).  Cited for NDVI measurement by ENVI with 680 nm 9 This may also be used in derivation of an NDVI variant called Re-normalized Difference Vegetation Index where RDVI=(R800 - R671))/(R800 + R671)1/2 .   RDVI is based on the contrast between the maximum absorption in the red due to chlorophyll pigments and the maximum reflection in the infrared caused by leaf cellular structure.  This filter also plays a role in a Modified Soil Adjusted Vegetation Index (MSAVI) developed to cancel soil reflectance and Soil and Atmospherically Resistant Vegetation Index (SARVI), which Micromizes both canopy background and atmospheric effects, 7  This filter is also useful in calculations of various plant pigment ratios such as Pigment Specific Simple ratio Chl (PSSR), Pigment Specific Normalized Difference (PSND) and Structure-Insensitive Pigment Index (SIPI).  Ratios between plant pigments often change in response to specific vegetation stress conditions such as changes in the carotenoid/chlorophyll ratio in apple trees in response to mite attacks. 6
   
Micro-MCA6    
- 490FS10-25 - See Micro-MCA4
- 550FS10-25 - See Micro-MCA4
- 680FS10-25 - See Micro-MCA4
- 720FS10-25 - Red Edge - The red edge describes the steeply sloped region of the vegetation spectral reflectance curve between 690 nm and 740 nm that is caused by the transition from chlorophyll absorption of red wavelengths and near-infrared reflection due to the mesophyll cells in leaves which in healthy plants act like a mirror to NIR.  This band is sensitive to temporal variations in crop growth and vegetation stress and provides additional information about chlorophyll and nitrogen status of plants. 2.
- 800FS20-25 - See Micro-MCA4
- 900FS20-25 -NIR- Peak or maximum reflectance region of the NIR spectrum for certain types and/or growth stages of vegetation or crops.  For crops such as cotton and corn or when crops are under stress or senescing there is significant change in reflectance along the "NIR shoulder."  Useful for computing crop moisture sensitive index. 2.
   

Micro-MCA12

   

- 490FS10-25

- See Micro-MCA4
- 520FS10-25 -Green - Positive change in reflectance per unit change in wavelength of this visible spectrum is maximum around this "green" waveband. First order derivative plot of crop spectra will show this.  Green band peak (or the point maximal reflectance) in the visible spectrum. 2
- 550FS10-25 - See Micro-MCA4
- 570FS10-25 - Green -Negative change in reflectance per unit change in wavelength of the visible spectrum is maximum around this wavelength.2   Factor in photochemical reflectance index (PRI) for estimating photosynthetic light use efficiency. Also compares the reflectance in the red and blue regions of the spectrum.  Used as factor in calculating PRI for xanthophyll cycle pigment change detection, carotenoid/chlorophyll ratio monitoring and water stress detection.
- 671FS10-25 -Red - This may be used in the derivation of an NDVI variant called Re-normalized Difference Vegetation Index where RDVI=(R800 - R671) / (R800 + R671)1/2 .  RDVI is based on the contrast between the maximum absorption in the red due to chlorophyll pigments and the maximum reflection in the infrared caused by leaf cellular structure. This filter also plays a role in a Modified Soil Adjusted Vegetation Index (MSAVI) developed to cancel soil reflectance in NDVI calculations and Soil and Atmospherically Resistant Vegetation Index (SARVI), which Micromizes both canopy background and atmospheric effects, 7 
- 680FS10-25 - See Micro-MCA4
- 700FS10-25 - Red Edge - This is a second red edge monitoring wavelength.  The red edge slope decreases as NIR reflectance drops in insect-damaged foliage. The wavelength and point of maximum slope also shift during senescence or stress-induced chlorosis. 8   So,Red Edge Position (REP) is an excellent indicator of plant stress and growth. 2. Also a factor in Chlorophyll Absorption in Reflectance Index (CARI) used in estimating chlorophyll content in crops. 6
- 720FS10-25 - See Micro-MCA6
- 800FS10-25 - See Micro-MCA4

- 840FS10-25

NIR - This filter's maximum transmissivity actually occurs at 845 nm.  This operates in the best spectral region to distinguish vegetation varieties and conditions.   This is the center of the "NIR shoulder."  For many crops, a broad-band or a narrow-band will provide the same result due to near uniform reflectance throughout the NIR shoulder (740 to 940 nm). In such instances, other bands along the NIR shoulder will be redundant.
- 900FS20-25 - See Micro-MCA6

- 950FS40-25

 

-NIR - Center of the moisture sensitive "trough" portion of NIR. The "trough" portion varies from 940 to 1040 nm and typically has Micromum reflectance around 975 nm (or point of maximum "dip" in the trough portion). Plant moisture sensitive band.This band is used in calculation of the Water Band Index where WBI = P900 / P970 9
   

   1   Band Combinations (James Quinn)  University of Northern Iowa

    2   Evaluation of Narrowband and Broadband Vegetation Indices for DeterMicrong Optimal Hyperspectral Wavebands for Ag Crop Characterization

    3   Relationships between leaf pigment content and spectral reflectance across a wide range of species, leaf structures and developmental stages

    4   Assessing canopy PRI for water stress detection with diurnal airborne imagery

    5   Vegetation indices as indicators of damage by the sunn pest (Hemiptera: Scutelleridae) to field grown wheat

    6   Soil Backgrounds Impact Analysis on Chlorophyll Indices Using Field, Airborne and Satellite Hyperspectral Data

    7   Hyperspectral vegetation indices and novel algorithms for predicting green LAI of crop canopies

    8   Spectral reflectance characteristics of eucalypt foliage damaged by insects

    9  Reference ENVI Users Guide