In the European Landscape Convention, landscape is seen as a key element in individual and social well-being and its protection, management and planning entail rights and responsibilities for everyone. Furthermore, the Convention states that landscape is an important public interest and constitutes an important part of the quality of life for people everywhere, contributing to the formation of local cultures and Europe’s natural and cultural heritage and identity. Consequently, it may be valuable to include people’s perception of landscapes into landscape planning. Therefore, it is important to know how people observe landscapes and how different landscape features are perceived. Most landscape perception studies use landscape photographs as visual stimuli, usually in combination with a questionnaire. It is, however, difficult to objectively measure how people observe landscapes. An innovative tool for measuring people’s landscape perception is provided by eye tracking. This technology enables to measure the speed and direction of eye movements (saccades) and fixations while observing images. Consequently, the entire scan path, made by an observer on an image, can be reconstructed and visualised. The aim of this study is to examine the difference between built and natural environments, represented by a set of landscape photographs ranging from urban to rural landscapes in Flanders (Belgium). Additionally, the observation patterns of different groups of respondents are compared (experts versus non-experts). The methodology consists of three steps. First, landscape photographs were taken in different landscape types, using a fixed focal length and a tripod. Second, the experiment was executed using an iView X RED eye tracking system. In this study, around 40 observers (20 experts and 20 non-experts) participated to the test. The experts consisted of graduate geographers and master students of geography, while undergraduate geography students participated as non-experts. During the experiment, the respondents were instructed to observe the landscape photographs for a few seconds, without executing specific search tasks. After the eye tracking measurements, the respondents were asked to rank the pictures from rural to urban landscapes in an additional questionnaire. Finally, the output of the experiment was statistically analysed to identify significant differences in perception between the landscape types ranging from urban to rural environments on the one hand and between the experts and non-experts on the other hand. The results were then visualised in fixation and saccade maps. Based on these maps, two-coloured heat maps were created to represent on the photographs which elements in the landscape drew most attention. The technique of eye tracking and the results of this study may be helpful in transdisciplinary landscape planning and design as it offers insights into the observation patterns of rural landscapes and built environments.