A Methodology for Mapping Perceived Spatial Qualities

Abstract

This manuscript proposes a five-step methodology that enables the mapping of perceived spatial qualities. To achieve such a goal, crowdsourcing and neural network methods were used. Crowdsourcing enables gathering data from people, and neural networks facilitate extending that knowledge to perform automatic classifications. The proposed method is then applied in the implementation of two use cases: perceived safety and perceived atmosphere of urban spaces. The use cases were conducted in the frame of the project Streetwise with a project partner that had the goal of creating the first maps of perceived spatial quality in Switzerland. The results obtained from the use cases showed that the application of the proposed methodology grants capturing the perceptions of a collective accurately.