This paper describes a vision based pedestrian detection and tracking system which is able to count people in very crowded situations like escalator entrances in underground stations. The proposed system uses motion to compute regions of interest and prediction of movements, extracts shape information from the video frames to detect individuals, and applies texture features to recognize people. A search strategy creates trajectories and new pedestrian hypotheses and then filters and combines those into accurate counting events. We show that counting accuracies up to 98 % can be achieved.
Origin–destination flow maps are a popular option to visualize connections between different spatial locations, where specific routes between the origin and destination are unknown or irrelevant. Visualizing origin–destination flows is challenging mainly due to visual clutter which appears quickly as data sets grow. Clutter reduction techniques are intensively explored in the information visualization and cartography domains. However, current automatic techniques for origin–destination flow visualization, such as edge bundling, are not available in geographic information systems which are widely used to visualize spatial data, such as origin–destination flows. In this article, we explore the applicability of edge bundling to spatial data sets and necessary adaptations under the constraints inherent to platform-independent geographic information system scripting environments. We propose (1) a new clustering technique for origin–destination flows that provides within-cluster consistency to speed up computations, (2) an edge bundling approach based on force-directed edge bundling employing matrix computations, (3) a new technique to determine the local strength of a bundle leveraging spatial indexes, and (4) a geographic information system–based technique to spatially offset bundles describing different flow directions. Finally, we evaluate our method by applying it to origin–destination flow data sets with a wide variety of different data characteristics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.