The urban environment is highly complex and heterogeneous and is characterised by rapid changes in its configuration and characteristics, which scholars have referred to as urban growth. However, urban growth is not synonymous with urban development. However, urban growth is not synonymous with urban development. For development to accompany growth, territorial ordinances must be adopted, which highlights the need for planning. The high frequency and broad scope of geographic alterations in the urban environment require quick and inexpensive methods to produce and update spatial information, such as those methods that depend on remote-sensing tools. The advent of remote-sensing satellite imagery with high spatial resolution introduced a new perspective from which to analyse and study urban areas, particularly with respect to the impact of transportation systems and human activities that operate in the midst of a global context that is looking for ways to promote a sustainable urban growth and development model. In this context, the present paper proposes a methodology for identifying useful urban features for transportation planning, particularly with respect to areas with higher concentrations of trip generators that are identified from satellite images, using object-based classification techniques. The proposed methodology for classifying images minimises costs and prioritises field activities related to research on trip generators, as well as origin/destination studies. The methodology was used in the city of João Pessoa, Paraíba State, Brazil with satisfactory and promising results.