Compared to pixel based where only the spectral information available for that individual picture element is considered for mapping purposes, OBIA (Object Based Image Analysis) is based on a localized group of pixels, considering the spatial properties of each pixel as they relate to each other. This method is comparable to the human visual interpretation of digital images that groups individual pixels into larger objects by colour, shape, size, and texture. The first step in OBIA is the creation of object primitives, which are created after image segmentation. Object primitives are delimited based on the pixel values and represent a group of pixels in a definite space within a scene with a wide range of information.

Object primitives after image segmentation
In order to evaluate the suitability of OBIA for mapping aquatic reed beds, two classification methodologies were implemented and compared in terms of accuracy. In this study was addressed:

• How accurate can aquatic reed beds be mapped by processing official CIR imagery with OBIA?
• How accurate can density of aquatic reed beds be mapped by implementing manual delineation by visual interpretation of CIR orthophoto imagery?

Comparisson between manual (top) and OBIA (bottom) image analysis