Detecting tourist attractions using geo-tagged photo clustering

Abstract

Millions of geo-tagged photos are becoming available due to the wide spread of photo-sharing websites, which provide valuable information to mine spatial patterns from human activities. In this study, we present a simple and fast density-based spatial clustering algorithm to detect popular scenic spots using geo-tagged photos collected from Flickr. In this algorithm, Gaussian kernel is applied to estimate local density of data points, and a decision graph is used to obtain cluster centers easily. More than 289,000 geo-tagged photos located in five typical cities of China are downloaded as case studies, and data pre-processing such as duplicate removing is performed to improve the quality of clustering result. Finally, popular tourist attractions of each sample city are successfully detected with this algorithm, and our result is useful for recommending some interesting destinations which might not be on the list of tourist website or mobile guide applications. The proposed solution is robust with respect to different distributions of photos, and it is efficient by comparing with other popular clustering approaches.

3D Visibility Analysis for Evaluating the Attractiveness of Tourism Routes Computed from Social Media Photos

“…We use a cell size of 250 × 250 m (e.g., [42,43]), assuming this size represents the common grid structure of city streets, and the fact that it is rare to find adjacent attractive locations within this distance. Since we analyze walking routes that are at least several hundreds of meters, the grid cell approach ensures that each cell will have a POI, overcoming alternative clustering processes that use non-adaptive Kernel values (e.g., density-based) that can lead to numerous dispersed photos clustered to a single POI-instead of several ones (e.g., [44]). A smaller grid cell size allows the retrieval of local attractive places to compute more tuned routes; still, these routes can be less natural in terms of walkability, producing more walking deviations and detours due to the concertation of nearby numerous POIs that show lesser popularity.…”

Enriching Walking Routes With Tourism Attractions Retrieved From Crowdsourced User Generated Data

“…Faraznyar and Cerocone (2015) and Subramaniyaswamy et al (2015), for example, use the Mean Shift algorithm for arranging big data of geotagged photos to find group centroids. Other studies propose the use of DBSCAN and its variants (e.g., Sun et al, 2015, Korakakis et al, 2016, and Zhang et al, 2018. Becker et al (2015) and Ali et al (2013) choose to divide the geographic framework to equal cell sizes, searching for "popular" (cluster) cells, while Doytsher et al, (2017) developed a partitioned space to 3D equal size grid that copes also with the temporal dimension.…”

Learning Patterns of Tourist Movement and Photography from Geotagged Photos at Archaeological Heritage Sites in Cuzco, Peru

“…Multi-modal analysis of large collections offers new ways to visualize tourist trends (Crandall et al 2009). As many tourist photos are street scenes or self-portraits irrelevant to the sites themselves, automatic detection of iconic images can help capture key parts of a site, with applications in photo summarization and browsing (Zhang et al 2018) and landmark segmentation (Simon and Seitz 2008). When geotags are not available, vision methods can estimate geolocation (latitude and longitude) from the images themselves (Hays and Efros 2008, Kalogerakis et al 2009, Chen and Grauman 2011.…”