Physical planning frequently requires land-use information. Unfortunately, in most developing countries, the available land-use information sometimes cannot satisfy the need for particular planning tasks, when newness, accuracy, compatibility and relevance are taken into consideration. On the other hand, the advent of satellite remote sensing technology offers a wide range of data availability with various spatial, spectral and temporal resolutions. In addition, geographic information systems offer capability and flexibility in combining data coming from different sources and in different formats, so that land-use information can easily be integrated with other spatial data to support planning processes. With respect to this background, a new perspective on the development of land-use information has been developed in the form of a versatile land-use classification scheme. In this classification scheme, the land-use information is broken down into five elements, namely spectral-related cover, spatial, temporal, ecological and socio-economic dimensions. Each dimension is presented in a separate layer, and is generated using different spatial processing methods. The categorisation of each dimension is also designed for multi-resolution mapping. An example of Semarang area, Central Java, Indonesia is given using Landsat ETM+ and Quickbird imagery as the data source.