PEDESTRIAN BEHAVIOR FOR DEVELOPING STRATEGY IN TOURISM AREA; AGENT-BASED SIMULATION

Agent-based modelling is an approach to develop a design strategy in socio-related studies to understand pedestrian behavior by using simulation through validation using field observation. This study area has a historic city so that having several potential advantages as destination tourists and also having urban issues. Some facilities disseminate prosperous for domestic tourist destinations, transportation hubs (land and water-based transport), and public facilities. The purpose is to develop a design strategy of pedestrian behavior in urban space to be procedure based on computational modelling. By merging the result, it helps designers to depict pedestrian movement flow, permeability, and connectivity patterns, which represent the presumptions of the origins or source of movement, destinations, generators, and attractors of movement. This simulation examines and valuates spatial behavior models allowing to route preferences of each pedestrian in order to be used in the strategy of design process for architect, urban planner, or other designer stakeholders. The result will imply a walkable pedestrian-way design, where this approach of a pedestrian experience might be an effective tool in city planning.

Deep Visible and Thermal Camera-based Optimal Semantic Segmentation using Semantic Forecasting

Visible camera-based semantic segmentation and semantic forecasting are important perception tasks in autonomous driving. In semantic segmentation, the current frame's pixel level labels are estimated using the current visible frame. In semantic forecasting, the future frame's pixel-level labels are predicted using the current and the past visible frames and pixel-level labels. While reporting state-of-the-art accuracy, both of these tasks are limited by the visible camera's susceptibility to varying illumination, adverse weather conditions, sunlight and headlight glare etc. In this work, we propose to address these limitations using the deep sensor fusion of the visible and the thermal camera. The proposed sensor fusion framework performs both semantic forecasting as well as an optimal semantic segmentation within a multi-step iterative framework. In the first or forecasting step, the framework predicts the semantic map for the next frame. The predicted semantic map is updated in the second step, when the next visible and thermal frame is observed. The updated semantic map is considered as the optimal semantic map for the given visible-thermal frame. The semantic map forecasting and updating are iteratively performed over time. The estimated semantic maps contain the pedestrian behavior, the free space and the pedestrian crossing labels. The pedestrian behavior is categorized based on their spatial, motion and dynamic orientation information. The proposed framework is validated using the public KAIST dataset. A detailed comparative analysis and ablation study is performed using pixel-level classification and IOU error metrics. The results show that the proposed framework can not only accurately forecast the semantic segmentation map but also accurately update them.

Impacts of Social Distancing on Pedestrian Behavior and Risk Perception

Since COVID-19 transmission accelerated in the United States in March 2020, guidelines have recommended that individuals wear masks and limit close contact by remaining at least six feet away from others, even while outdoors. Such behavior is important to help slow the spread of the global pandemic; however, it may require pedestrians to make critical decisions about entering a roadway in order to avoid others, potentially creating hazardous situations for both themselves and for drivers. In this survey study, we found that while overall patterns of self-reported pedestrian activity remained largely consistent over time, participants indicated increased willingness to enter active roadways when encountering unmasked pedestrians since the COVID-19 pandemic began. Participants also rated the risks of encountering unmasked pedestrians as greater than those associated with entering a street, though the perceived risk of passing an unmasked pedestrian on the sidewalk decreased over time.

ANALISIS FAKTOR PERILAKU PEJALAN KAKI TERHADAP KESELAMATAN DI JALAN RAYA PADA WILAYAH JABODETABEK

Pedestrian behavior that is not predictable and cannot be controlled effectively results in pedestrians not complying with traffic rules and may be ending up in accidents. This study aims to determine the factors formed and determine the influence of demographics on pedestrian behavior in Indonesia. The questions on the Indonesian PBQ (Pedestrian Behavior Questionnaire) are adapted from an international questionnaire that has been tested in several countries and confirmed by factor analysis. The results of the analysis showed that from the 23 variables of pedestrian behavior, 5 (five) variables are reduced until the remaining 18 variables. The 18 pedestrian behavior variables form 6 factors, namely violations, errors, lapses, aggressive behavior, positive behavior, and traffic light violations. Violations and lapses in Indonesia are mostly committed by young adults, non-private employees, and those who have expenses of less than four million five hundred rupiah per month. Those who in the adult group show positve behavior more oftenly. Those who walk more than once a week show errors and positive behavior more oftenly. Those who have been involved in accidents show violations and aggressive behavior more oftenly.ABSTRAKPerilaku pejalan kaki yang tidak mudah diprediksi dan tidak dapat dikendalikan secara efektif mengakibatkan pejalan kaki tidak mematuhi aturan lalu lintas sehingga terjadi kecelakaan. Penelitian ini bertujuan untuk mengetahui faktor-faktor yang terbentuk serta mengetahui pengaruh demografis terhadap perilaku pejalan kaki di Indonesia. Pertanyaan-pertanyaan pada Indonesia PBQ (Pedestrian Behavior Questionnaire) diadaptasi dari kuesioner Internasional yang telah diuji di beberapa negara dan dikonfirmasi dengan analisis faktor. Hasil analisis menunjukkan dari ke-23 variabel perilaku pejalan kaki tereduksi sebanyak 5 (lima) variabel hingga tersisa 18 variabel. Ke-18 variabel perilaku pejalan kaki tersebut membentuk 6 faktor yaitu violations, errors, lapses, perilaku agresif, perilaku positif, dan pelanggaran lampu lalu lintas. Pelanggaran violations dan lapses di Indonesia paling banyak dilakukan oleh kelompok dewasa muda, pekerjaan bukan pegawai swasta, dan yang memiliki pengeluaran kurang dari empat juta lima ratus rupiah per bulan. Mereka yang termasuk kelompok dewasa madya lebih sering menunjukkan perilaku positif. Mereka yang berjalan kaki lebih dari sekali dalam seminggu lebih sering menunjukkan kesalahan errors dan perilaku positif. Mereka yang pernah terlibat kecelakaan lebih sering menunjukkan perilaku violations dan agresif.

Pedestrian Behavior Recognition Based on Improved Dual-stream Network with Differential Feature in Surveillance Video

In order to improve the pedestrian behavior recognition accuracy of video sequences in complex background, an improved spatial-temporal two-stream network is proposed in this paper. Firstly, the deep differential network is used to replace the temporal-stream network so as to improve the representation ability and extraction efficiency of spatiotemporal features. Then, the improved Softmax loss function based on decision-making level feature fusion mechanism is used to train the model, which can retain the spatiotemporal characteristics of images between different network frames to a greater extent and reflect the action category of pedestrians more realistically. Simulation results show that the proposed improved network achieves 87% recognition accuracy on the self-built infrared dataset, and the computational efficiency is improved by 15.1%.