Article ID Journal Published Year Pages File Type
6458612 Computers and Electronics in Agriculture 2017 10 Pages PDF
Abstract

•Further improvement of a novel spectral-reflectance-based plant discrimination sensor for use in selective herbicide spot spraying systems is described.•To maintain high discrimination accuracy over the practical temperature range 16-34 °C a temperature-stabilized laser-based three-waveband plant discrimination sensor has been developed and tested indoor on Canola and Wild Radish leaves.•Experimental results show that by controlling the temperature of the laser diodes, canola crops can be discriminated from wild radish weeds, which is a dominant weed in the canola crop field, with accuracy as high as 90% and traveling speed of about 7.5 km/h.

We propose the use of temperature-stabilized lasers to improve the accuracy of a spectral-reflectance-based plant discrimination sensor for use in selective herbicide spraying systems. The discrimination of Canola from Wild-radish is based on Normalized Difference Vegetation Index (NDVI) measurements at two different laser wavelengths. Indoor experimental results show that the relative discrimination accuracy for a temperature non-stabilized sensor drops to 12.5% when the laser temperature varies between 16 °C and 34 °C. Experimental results also show that by controlling the temperature of the laser diodes, canola crops can be discriminated from wild radish weeds with accuracy as high as 90%.

Related Topics
Physical Sciences and Engineering Computer Science Computer Science Applications
Authors
, , , , ,