Object-oriented classification of very high resolution airborne imagery for the extraction of hedgerows and field margin cover in agricultural areas

The recent advancement in technology for the airborne collection and subsequent processing and analysis of digital remotely sensed data has been significant. An investigation into the contributions that recent developments have made to the assessment and extraction of hedgerow and field margins is presented. Research into the active legislation by which hedgerows and agricultural biodiversity in England are bound is brought together in the concept of an ideal classification. Using Leica Geosystems ADS40 data, collected for Berkshire (UK), the degree to which the ideal classification can be reached is investigated, focusing on the extraction of hedgerows and unploughed field margin protective boundary strips as image objects. Elevation data were derived from the ADS40 imagery data to accompany true-colour and colour-infrared spectral information. By buffering the detected hedgerow objects in a GIS, we facilitated an investigation of the presence of measures that satisfy the minimum hedgerow protection requirements stipulated by the UK Department for Environment, Food and Rural Affairs (DEFRA). In our study area, it was estimated that 68% of areas that need to be classified as ‘grassy hedgerow boundary strip’ to satisfy the minimum requirements of the legislation were established with low-lying grassy cover before the introduction of the legislation in July 2005. The implications of this research for the development of automated classification methods for field scale agricultural mapping are discussed.