Near Infrared (NIR) is one of the imaging techniques available when using Digital Multi-Spectral Imagery (DMSI).

The usual deliverables from a sortie gathering NIR data would be a Plant Cell Density (PCD) image and a False colour image. A true colour image is also generated from the data. One of the most powerful tools to be used in NIR data is a Change Detection image which is generated once subsequent data collection passes are made and a later and earlier PCD image are statistically subtracted from each other to show the change in vegetation biomass in the time period between passes.

Digital Multi-Spectral Imagery (DMSI) is image data of the same scene recorded simultaneously through 4 narrow spectral bands. The Digital Multi-Spectral Camera (DMSC) system integrates 4 individual, high quality, pixel digital imaging devices (CCDs) capable of measuring ground reflectance at high resolution (0.5metre – 2metre) and high sensitivity within visible and near-infrared wavelengths.

Advantages of DMSI

  • High pixel resolution for sensitive spatial and spectral characterisation of individual ground features
  • High spectral resolution provides sensitive information for;
  • Discriminating and mapping variations in vegetation type, density, distribution and health
  • And monitoring for changes in vegetation status and condition between successive survey flights
  • Natural Colour and False Colour Infrared images acquired simultaneously
  • No further digitising required
  • GIS ready
  • Allows consistent and rapid interpretation (spectral and textural analysis) across multiple broadscale areas of interest using automated image classification techniques

Digital Multi-Spectral Camera (DMSC) Sensor

The DMSC has been specifically designed for low cost multi-spectral remote sensing of vegetation. From flight planning to image acquisition the system has been developed to take advantage of high solar angles and ground feature illumination without incurring reflectance artefacts usually associated with solar hotspot and the Bidirectional Reflectance Distribution Function

Standard Bandpass Filters for Vegetation Mapping

The DMSI narrow band-pass filters are easily interchanged for specific applications, however the 4 spectral bands utilised for vegetation mapping and monitoring are 20 nanometres wide and centred about the principal reflectance spectra features of vegetation;

DMSI Spectral Band and Vegetation
Reflectance feature

    • Blue – 450nm (leaf pigment absorption)
    • Green – 550nm (relatively higher reflectance and transmission)
    • Red – 675nm (strong chlorophyll absorption)
    • Near Infrared – 780nm (high infrared reflectance “plateau”)

    Data Pre-processing Correction Techniques

    All airborne imagery acquired for the purpose of spectral analysis requires correction for camera distortion and scene brightness effects. Five years of research and development has seen the development of proprietary software that is capable of minimising these effects to allow the consistent and reliable production of high quality seamless mosaics while maintaining the overall spectral integrity of the raw imagery. The software runs in batch mode across entire data sets and requires no user intervention.

    The two major effects and the correction techniques are detailed below;

      Band Mis-registration

      Mechanical alignment of the four DMSC lenses to a point on the ground is only possible to within one to two pixels across the camera field of view. Proprietary image correction software further minimises this mis-registration to within 0.2 of pixel to ensure the integrity of any pixel/band comparison.

      Scene Brightness

      All ground imaging systems with a large field of view are susceptible to differential illumination effect known as the Bi-directional Reflectance Distribution Function (BRDF). This effect means that the same object appearing in more than one image frame will have different brightness and spectral (colour) characteristics in each frame. Using a in-house technique these effects are reduced from typically 20% of the dynamic range of the image to approximately 5% or less. This correction is extremely important for:

    • Maintaining spectral integrity across individual images
    • The generation of radiometrically seamless mosaics
    • The application of band indices and classification procedures to mosaicked image frames
    • Combining multi-temporal images of the same scene for automated change detection purposes

    Image Rectification and Mosaicing

    Off-the-shelf softcopy ortho-rectification software is used to geo-rectify and mosaic frames of raw DMSI. Automated processes are used for moving vast numbers of image frames through the system within short timeframes. A unique system for automatic rectification and mosaicking of images for multi-temporal analysis. This process significantly reduces the time in producing high value Change Detection across survey areas.

    Image Feature Extraction and GIS Layer Creation

    Proprietary automated image classification routines for extracting high detail information layers from DMSI. These routines usually require fine-tuning for an individual client’s mapping requirements, however once finalised will run in batch mode with no user intervention across multiple data sets.