Urban swarming transportation (UST) is a type of road transportation where multiple types of vehicles such as cars, buses, trucks, motorcycles, and bicycles, as well as pedestrians are allowed and mixed together on the roads. Predicting the traffic jam speed under UST is very different and difficult from the single road network traffic prediction which has been commonly studied in the intelligent traffic system (ITS) research. In this research, the road network wide (RNW) traffic prediction which predicts traffic jam speeds of multiple roads at once by utilizing citizens’ mobile GPS sensor records is proposed to better predict traffic jam under UST. In order to conduct the RNW traffic prediction, a specific data preprocessing is needed to convert traffic data into an image representing spatial-temporal relationships among RNW. In addition, a revised capsule network (CapsNet), named OCapsNet, which utilizes nonlinearity functions in the first two convolution layers and the modified dynamic routing to optimize the performance of CapsNet, is proposed. The experiments were conducted using real-world urban road traffic data of Jakarta to evaluate the performance. The results show that OCapsNet has better performance than Convolution Neural Network (CNN) and original CapsNet with better accuracy and precision.
23 November 2022 in IEEE Transactions on Network Science and Engineering
A cloud-based Artificial Intelligence (AI) service has recently empowered the Internet of Medical Things (IoMT) in many applications on the remote Human Interaction Recognition of Pervasive Healthcare Monitoring (HIR-PHM). In this work, a framework of the skeleton-based HIR-PHM under secure Edge-Fog-Cloud computing (EFCC) was proposed to manage the computation and storage resources, latency, cyber-attack, and privacy-preserving simultaneously. At the Edge, IoMT with a camera, record the human interaction as videos and sends them to the Fog, which installed a human pose estimation model, PoseNet, to convert the videos into human skeleton data. At the Cloud, a skeleton-based Spatial-Temporal Graph Convolution Network with Pairwise Adjacency Matrix (STGCN-PAM) was employed to recognize the human interaction. A Hybrid of One Time Password (OTP), Hashed Messages Authentication Code (HMAC), and symmetric cipher were employed to secure the skeleton data. The UT-Interaction dataset was used to evaluate the proposed framework. Besides, the computation performance and latency under EFCC were compared with Edge-Fog and Edge-Cloud deployments. Experimental results confirmed two major contributions: 1) the proposed framework has promising performance compared with other methods on the dataset. 2) The skeleton-based HIR-PHM with secure EFCC shows the advantages of better computation performance in terms of frame per second.
In metropolitan areas, common traffic issues include traffic congestion, traffic accidents, air pollution, and energy consumption occur. To resolve this issues, Intelligent Transportation Systems (ITS) have been evolved by many researchers. One of the important sub-systems in the development of ITS is a Traffic Management System (TMS) which attempts to reduce a traffic congestion. In fact, TMS itself relies on the estimation of traffic flow, therefore providing such an accurate traffic flow prediction is needed. For this reason, we aim to provide an accurate traffic flow prediction to facilitate this system. In this works, a Supervised Deep Learning Based Traffic Flow Prediction (SDLTFP) was proposed which is a type of fully-connected deep neural network (FC-DNN). Timely prediction is also a major issue in guaranteeing reliable traffic flow prediction. However, training a deep network could be time-consuming, and overfitting is might be happening, especially when feeding small data into the deep architecture. The network is learned perfectly during the training, but in testing with the new data, it could fail to generalize the model. We adopt the Batch Normalization (BN) and Dropout techniques to help the network training. SGD and momentum are carried out to update the weight. We then take advantage of open data as historical traffic data which are then used to predict future traffic flow with the proposed method and model above. Experiments show that the Mean Absolute Percentage Error (MAPE) for our traffic flow prediction is within 5 % using sample data and between 15% to 20% using out of the sample data. Training a deep network faster with BN and Dropout reduces the overfitting.
Weather change such as raining is a crucial factor to cause traffic congestion, especially in metropolises with the limited sewer system infrastructures. Identifying the roads which are sensitive to weather changes, defined as weather-sensitive roads (WSR), can facilitate the infrastructure development. In the literature, little research focused on studying weather factors of developing countries that might have deficient infrastructures. In this research, to fill the gap, the real-world data associating with Jakarta, Indonesia, was studied to identify WSR based on smartphone sensor data, real-time weather information, and road characteristics datasets. A spatial-temporal congestion speed matrix (STC) was proposed to illustrate traffic speed changes over time. Under the proposed STC, a sequential clustering and classification framework was applied to identify the WSR in terms of traffic speed. In this work, the causes of WSR were evaluated based on the variables’ importance of the classification method. The experimental results show that the proposed method can cluster the roads according to the pattern changes in the traffic speed caused by weather change. Based on the results, we found that the distances to shopping malls, mosques, schools, and the roads’ altitude, length, width, and the number of lanes are highly correlated to WSR in Jakarta.
1 June 2021 in Journal of Physics: Conference Series
An image resulting from a low-resolution (LR) camera on the mobile phone has lower quality than a high-resolution(HR) camera on a DSLR. Meanwhile, the HR camera is pricing if compared with the LR camera. How to achieve a single-image quality on LR camera likewise on HR camera becomes essential research in the past years. Addressing this issue can be done by upscaling a single LR image. Recently, the super-resolution generative adversarial network (SRGAN) model is one of the state-of-the-art super-resolution(SR)models employed on single-image SR. However, implementing a deep learning model like SRGAN on a mobile device is challenging in computation power and resources. This study aims to develop a smaller and lower resources model while preserving single-image SR quality on mobile devices. To meet these objectives, we convert, quantize, and compress the SRGAN model on Snapdragon Neural Processing Engine (SNPE) as an example. We then validate the SRGAN on the DIV2K dataset on which improves the model performances. Besides, we conduct experiments on GPU, DSP environment. The experimental result confirmed that SNPE-SRGAN capable of achieves not only HR images’ quality but also low latency by 0.06 second and smaller model by 1.7 Mb size running on DSP. Also, the SRGAN-DLC-Quantized running on GPU has a smaller size by 1.7 Mb and lower latency by 1.151 seconds compared with Non-quantized SRGAN-TensorFlow by 9.1 Mb and 1.608 seconds latency.