A Systematic Approach to Urban Block: Defining Automatic Tool for Urban Form

The unprecedented rapid change in the urban form has prompted a growing number of research to analyze and understand the phenomenon in recent years. In a never-ending cycle of change and re-elaboration, the broad diversity of urban forms that we see today serves as the baseline for future and new forms. At the same time, the growing accessibility of geographic data and mapping tools have boosted urban morphology studies. The burgeoning development of automatic tools enables machines to get a human-like understanding of urban form hinged on images. In this new context, a comprehensive, systematic method of evaluation and comparison of forms needed to be defined. This study aims to present the manual definition of urban form features to create systematic input for automatic tools. Particularly, as a constitutional element of urban form, urban block is analyzed within the scope of classification approaches. The preliminary step is to present organized knowledge of urban block to understand how it is constructed. The methodological process is encompassed detection and classification of urban block by in-depth analysis of relative literature. The second step of the study is defined by using this structural classification to detect the urban block with automatic tools such as a deep convolutional neural network. The preliminary outcome of this study is the representation of urban block by providing a classification tree of the urban block based on comparative literature