If we look at the Internet of Things (IoT) from a viewpoint that comprises higher levels of abstraction, we will see that the IoT generated data can actually be transformed into more complex information, which would in turn facilitate the lives of human users. Because sensors have different purposes and measure different phenomena, it is necessary to know them and their different areas and domains of application so we can make a better use of their potential. This paper presents the identification and categorization of the main sensors used these days to build IoT applications, arranged in a taxonomy of application domains and sensor measurement types. To this purpose, we review the literature in order to identify IoT solutions, areas and domains of application and the main sensor types employed in these solutions. We hope this taxonomy can provide IoT designers, developers, and researchers with a snapshot of how sensors are currently used in the IoT application domains. Knowing the source devices is a key strategy to provide publication, discovery, sharing, reuse and integration of data/information within the IoT. We believe identifying and categorizing those sensors could be the first step to creating in the future a common communication model, which could be instantiated from each environmental context on the IoT.