The objective of this research was to develop a method to map weeds in sorghum as the first step as a procedure to control them using site-specific weed management (SSWM). Site-specific weed management is a method to limit the application of herbicides only to areas with weeds.Accurate mapping of weeds is a pre-requisite for applying SSWM. Analysis of hyperspectral remote sensing imagery is recognized as a potentially cost effective technique for discriminating between weeds and crop plants. This research involved: i) collecting hyperspectral reflectance spectra from weeds and sorghum plants, ii) Stepwise Linear Discriminant Analysis (SLDA) to identify the most significant spectral bands, iii) Linear Discrimination Analysis (LDA) to test the accuracy of the SLDA bands for classifying weeds and sorghum, and iv) analysis of customized multispectral imagery to produce maps which detected weeds in the sorghum crop.Hyperspectral signatures of weeds and sorghum were obtained using a FieldSpec®
Handheld2TM spectroradiometer with a spectral range from 325 nm to 1075 nm. Spectra were recorded for different weed species and sorghum plants for three years, 2012 to 2014. Data were collected at four different stages of plant growth each year, from week one to week four after planting.The results show that it is feasible to discriminate spectral profiles of weeds from each other weeds and from sorghum plants. Statistical Analysis Software (SAS) was used to identify the most significant spectral bands (10 nm width) from the hyperspectral reflectance data using SLDA. All weeds and sorghum were correctly classified in 2012 using LDA for week four reflectance data. In 2013, the classification accuracy increased with stage of growth (weeks one to four) from 85% to 90%. In 2014, the classification accuracy also increased with stage of growth (weeks two to four) from 90% to 100%. Combinations of spectral bands were analysed to reduce the number of potential bands identified from the SLDA results. Spectral bands centred on 930, 890, 710, 700, 560and 500 nm were common to the 20 most significant spectral bands identified by SLDA analysis each year. Six spectral bands 850, 720, 710, 680, 560 and 440 nm were subsequently selected for use in multispectral image collection. They were selected based on maximizing the differences and similarities between the 2013 weed and crop reflectance profiles. These bands were used for the band-pass filters in the Tetracam MCA 6 camera used for collecting high spatial resolution terrestrial and aerial imagery.The imagery was analysed using Object-Based Image Analysis (OBIA) and Vegetation Index Analysis (VIA) to classify weeds and sorghum plants. OBIA identified weeds more successfully than VIA. The accuracy of OBIA classification was tested using two methods.Confusion matrices were used to measure the Coefficient of Agreement (Khat), Overall Accuracy iii and Producer's and User Accuracies. Geometry matrices were used to measure under and over segmentation. The Overall accuracy and Khat for all...