C-IMAGE: city cognitive mapping through geo-tagged photos

Abstract

Traditional research categorizes people's perceptions towards the city environment with Kevin Lynch's five elements: node, path, edge, district, and landmark. His method has been a keystone in guiding both urban design and urban study for decades. However, enabled by the proliferation of crowd sourcing technology, this thesis tries another angle to detect, measure, and analyze people's perceptions through geo-tagged photos.Using Python scripting language, the project downloads photos of 26 cities (an average of 100,000 photos per city) with geographic information, in North America, Europe, and Asia, from Panoramio and partially from Flickr. The process of image analysis is built on a computer vision technique -scene understanding, which categorizes the photos into 102 scenes by content. This paper aims to introduce photos collected at the massive scale as an additional method of city image. The project name of C-IMAGE reflects the interaction between city computation and city cognition.This thesis implies different roles of C-IMAGE in three applications for urban planning: a monitor for city forms, an evaluator for planning strategies, and a reference for urban functions. Important discoveries through these applications include that 1) C-IMAGE can partially confirm Kevin Lynch's city image efficiently; 2) C-IMAGE can disclose both agent-led and agent-less urban changes; 3) There are mainly four prototypes among the tested 26 cities, based on a seven-category C-IMAGE patterns; 4) C-IMAGE can evaluate planning suitability because its indicators are associated with the real activity; 5) Part of land use is statistically predictable in the case of Manhattan.Three impacts on urban planning are highlighted: First C-IMAGE is a method to pull data from social media to describe vividly the collective perception from the public. Second it computationally extracts information from photos at a deeper level for the study of urban space and activities. Third, C-IMAGE benefits from the multidisciplinary cooperation between planning and computer science to open up a new channel and provide analytical tools for further research on the cognitive mapping of urban spaces. 3/180

Crowdsourced Data Mining for Urban Activity: Review of Data Sources, Applications, and Methods

“…In recent years, the application of social media in urban dynamics has expanded to urban social problems by integrating with census data to assign socioeconomic/demographic features to users (Li et al 2013;Shelton et al 2015) or by extracting demographic features (gender, age, and ethnicity) from profile information through methods such as face detection and name analysis (Bocconi et al 2015;Luo et al 2016). Moreover, social media sources such as Instagram and Flickr contain massive image content examined for urban perception (Liu et al 2016b) and patterns of tourists (Li et al 2018).…”

Perceiving Beijing’s “City Image” Across Different Groups Based on Geotagged Social Media Data

“…For example, Salesses et al used thousands of geotagged photos to compare the differences in security, social hierarchy and uniqueness between New York, Boston, Linz and Salzburg [19]. Liu et al classified the Flickr photos using deep learning, after which they statistically analyzed the image in seven typical cities around the world to explore the relevance and diversity of city image [20]. Long and Zhou investigated the image characteristics and similarities in 24 Chinese cities through analyzing photo locations, tags and contents on Flickr [21].…”

Measuring accessibility of bus system based on multi-source traffic data

“…Points of Interest (POI) refer to some geographic entities closely related to people's daily life, which plays an essential role as an indicator in traffic planning (Liu et al 2016;Xing and Meng 2018). POI indicate the potential opportunities of residents, so the number of POI is used to quantify attractiveness in study region.…”