Cost analysis of the application of various photointerpretation techniques to land use classification

When facing land use classification by photointerpretation, an initial decision-taking process happens regarding the technique to apply to get the best results from the source of information. Some factors are taken into account like adaptation to the photograph properties and flight parameters, as well as ground relief and coverage. Not less important have to be the economic variables involved regarding the equipment, personnel and other resources involved. A comparative study was carried out in a study area to measure the effectiveness of each of the available techniques using a cadastral parcel map-based approach, and with the goal of building a spatial database. Once the stages to be followed for each of each procedure have been defined, a test area with a mixed inland-coastal character within the whole study area was selected. The three approaches applied were map shading-tablet digitising-tesselation coding, map coding-digital parcel coding, photo scanning-georreferencing-digital parcel overlayingland use extraction. When having the original photographs on which digital parcels were drawn the most efficient working approach is the third. In other circumstances the second has demonstrated a higher efficiency and an alternative. Then this method was extended to the whole study area and measured the work times for the several subareas, since difficulties arise when having varied topography, mixed land uses and non-updated cadastral map. Results show varied costs with the same technique in non-homogenous areas in the range of 1.2-7.5 hours.