Now showing 1 - 3 of 3
  • Publication
    A Comparison of Two Ranging Approaches in an Active, Optical Plant Canopy Sensor
    (Institute of Electrical and Electronics Engineers (IEEE), 2014) ; ;
    Active optical sensors that contain their own modulated light sources are becoming popular for 'sensing' photosynthetically-active biomass in crops and pastures. These sensors detect optical reflectance to derive spectral vegetation indices, such as the normalised difference vegetation index (NDVI), and are subsequently calibrated to measure plant parameters e.g. biomass. However, research has demonstrated the accuracy of the derived measurements can often be improved by including both a spectral index and a corresponding measure of plant height. This paper describes an active, optical sensor that integrates modulated reflectance sensing with the ability to measure (range) the distance between the source and a target surface. Two ranging techniques are evaluated; one based on the inverse square law (ISL) of reflected radiation and another based on a position-sensitive detector (PSD). Both ranging methods proved capable of reliably delineating target distances out to 4.0 m from the source. Over this range, the PSD detector exhibited a distance-invariant RMSE of ± 2.6 cm whilst the ISL method exhibited an almost linear increase in error of ± 25 % of the measured distance to a spectralon target. Application to a vegetative target (Kikuyu grass), demonstrated the ISL ranging method to yield an average RMSE of ± 3.0 cm in the range of 0.60-1.40 m, while the average RMSE of the PSD over a range of 0.50-1.10 m was observed to be ± 10.0 cm. Despite superior accuracy, target reflectance variations may prove problematic in the use of a PSD ranging sensor and requires further investigation.
  • Publication
    Automatic detection of evaporation reducing monolayers
    (Australian Society for Engineering in Agriculture, 2011)
    Coop, Paul A
    ;
    ; ;
    Mono-molecular layers, or monolayers, on the surface of water can reduce evaporation by up to 40% on small storages. Effective reduction requires the maintenance of coverage by dispensers, which must be informed by sensors capable of detecting the presence/absence of the monolayer. However, no system has yet been developed that can reliably automatically detect the presence of monolayer. Numerous detection techniques for potential field deployment, including optical and infrared reflectance, measurement of surface tension and wave damping methods were investigated. A wind assisted surface probe (WASP), based on the temperature differential between the sub-surface and volume of a small sample of periodically-entrapped water was subsequently developed. In small scale testing it has shown to be a reliable indicator of monolayer in all but very high humidities.
  • Publication
    A comparison of two ranging approaches to an integrated active, optical, reflectance and ranging sensor
    (Australian Society for Engineering in Agriculture, 2011) ; ;
    Active optical sensors that contain their own modulated light sources are becoming popular for 'sensing' photosynthetically-active biomass in crops. Primarily confined to on-ground deployment, these sensors rely on detecting optical reflectance in two or more wavebands (for example red and near infrared). The derived spectral vegetation indices, such as the widely-used normalised difference vegetation index (NDVI) are subsequently calibrated to a measure of biomass, tiller number, leaf area index or the like. However research has demonstrated the accuracy of the conversion process can often be improved by including both spectral index and a corresponding measure of plant geometry such as height. This paper describes an active, optical sensor that integrates the modulated reflectance sensing with the ability to measure (range) the distance between the source and a target surface. Two ranging techniques are evaluated; one based on the inverse square law (ISQ) of reflected radiation from a target and another based on position-sensitive detector (PSD). Both ranging methods proved capable of reliably delineating target distances out to 4.0m from the source. Over this range, the PSD detector exhibited a distance-invariant uncertainty of ±2.6cm whilst the ISQ method exhibited an almost linear increase in error of ±25% of the measured distance.