E-Scooter Rider detection and classification in dense urban environments

Article Multidisciplinary Sciences

E-scooter related injuries: Using natural language processing to rapidly search 36 million medical notes

Kimon L. H. Ioannides et al.

Summary: This study retrospectively reviewed patient data from January 1, 2014 to May 14, 2020 in the greater Los Angeles area to investigate e-scooter injuries and estimate the injury rate. The study found that there were 115 injuries per million e-scooter trips treated in the healthcare system. Additionally, the study revealed that 30% of patients required treatment in multiple clinical settings and 29% required advanced imaging.

PLOS ONE (2022)

Add to Collection

Article Computer Science, Artificial Intelligence

Survey of pedestrian detection with occlusion

Chen Ning et al.

Summary: This paper summarizes the research progress of pedestrian detection technology with occlusion, including types of occlusion, traditional methods, deep learning methods, main ideas, core problems, and outlook on future development.

COMPLEX & INTELLIGENT SYSTEMS (2021)

Add to Collection

Review Engineering, Civil

Overcoming Occlusion in the Automotive Environment-A Review

Shane Gilroy et al.

Summary: Accurate and consistent vulnerable road user detection is challenging for autonomous vehicles, with partial occlusion being a complex and challenging issue. This research reviews solutions for occlusion detection and handling in the automotive environment, including detection, reasoning, and strategies for handling occlusion in pedestrian, vehicle, and object detection applications. Remaining challenges to achieving the required level of object detection performance for safe autonomous driving are also discussed.

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2021)

Add to Collection

Article Computer Science, Artificial Intelligence

TJU-DHD: A Diverse High-Resolution Dataset for Object Detection

Yanwei Pang et al.

Summary: Researchers have built a diverse high-resolution dataset called TJU-DHD, which includes a large number of high-resolution images and labeled objects with rich scale and appearance variations. They conducted experiments on object detection and pedestrian detection using four different detectors, aiming to advance research in these two scenes through the newly constructed dataset.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2021)

Add to Collection

Article Computer Science, Information Systems