scholarly journals Deep Learning Architectures for Skateboarder–Pedestrian Surrogate Safety Measures

2021 ◽  
Vol 1 (2) ◽  
pp. 387-413
Author(s):  
Chowdhury Erfan Shourov ◽  
Mahasweta Sarkar ◽  
Arash Jahangiri ◽  
Christopher Paolini
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
State Of The Art ◽  
Pedestrian Safety ◽  
Time Signal ◽  
Single Shot ◽  
Safety Measures ◽  
Surrogate Safety Measures ◽  
Learning Architectures

Skateboarding as a method of transportation has become prevalent, which has increased the occurrence and likelihood of pedestrian–skateboarder collisions and near-collision scenarios in shared-use roadway areas. Collisions between pedestrians and skateboarders can result in significant injury. New approaches are needed to evaluate shared-use areas prone to hazardous pedestrian–skateboarder interactions, and perform real-time, in situ (e.g., on-device) predictions of pedestrian–skateboarder collisions as road conditions vary due to changes in land usage and construction. A mechanism called the Surrogate Safety Measures for skateboarder–pedestrian interaction can be computed to evaluate high-risk conditions on roads and sidewalks using deep learning object detection models. In this paper, we present the first ever skateboarder–pedestrian safety study leveraging deep learning architectures. We view and analyze state of the art deep learning architectures, namely the Faster R-CNN and two variants of the Single Shot Multi-box Detector (SSD) model to select the correct model that best suits two different tasks: automated calculation of Post Encroachment Time (PET) and finding hazardous conflict zones in real-time. We also contribute a new annotated data set that contains skateboarder–pedestrian interactions that has been collected for this study. Both our selected models can detect and classify pedestrians and skateboarders correctly and efficiently. However, due to differences in their architectures and based on the advantages and disadvantages of each model, both models were individually used to perform two different set of tasks. Due to improved accuracy, the Faster R-CNN model was used to automate the calculation of post encroachment time, whereas to determine hazardous regions in real-time, due to its extremely fast inference rate, the Single Shot Multibox MobileNet V1 model was used. An outcome of this work is a model that can be deployed on low-cost, small-footprint mobile and IoT devices at traffic intersections with existing cameras to perform on-device inferencing for in situ Surrogate Safety Measurement (SSM), such as Time-To-Collision (TTC) and Post Encroachment Time (PET). SSM values that exceed a hazard threshold can be published to an Message Queuing Telemetry Transport (MQTT) broker, where messages are received by an intersection traffic signal controller for real-time signal adjustment, thus contributing to state-of-the-art vehicle and pedestrian safety at hazard-prone intersections.

scholarly journals Evaluating the Single-Shot MultiBox Detector and YOLO Deep Learning Models for the Detection of Tomatoes in a Greenhouse

Sensors ◽  
2021 ◽  
Vol 21 (10) ◽  
pp. 3569
Author(s):  
Sandro Augusto Magalhães ◽  
Luís Castro ◽  
Germano Moreira ◽  
Filipe Neves dos Santos ◽  
Mário Cunha ◽  
...  
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
Life Cycle Stage ◽  
Single Shot ◽  
Learning Models ◽  
Perception System ◽  
Ripe Tomato ◽  
Harvesting Process ◽  
Tomato Detection

The development of robotic solutions for agriculture requires advanced perception capabilities that can work reliably in any crop stage. For example, to automatise the tomato harvesting process in greenhouses, the visual perception system needs to detect the tomato in any life cycle stage (flower to the ripe tomato). The state-of-the-art for visual tomato detection focuses mainly on ripe tomato, which has a distinctive colour from the background. This paper contributes with an annotated visual dataset of green and reddish tomatoes. This kind of dataset is uncommon and not available for research purposes. This will enable further developments in edge artificial intelligence for in situ and in real-time visual tomato detection required for the development of harvesting robots. Considering this dataset, five deep learning models were selected, trained and benchmarked to detect green and reddish tomatoes grown in greenhouses. Considering our robotic platform specifications, only the Single-Shot MultiBox Detector (SSD) and YOLO architectures were considered. The results proved that the system can detect green and reddish tomatoes, even those occluded by leaves. SSD MobileNet v2 had the best performance when compared against SSD Inception v2, SSD ResNet 50, SSD ResNet 101 and YOLOv4 Tiny, reaching an F1-score of 66.15, an mAP of 51.46 and an inference time of 16.44ms with the NVIDIA Turing Architecture platform, an NVIDIA Tesla T4, with 12 GB. YOLOv4 Tiny also had impressive results, mainly concerning inferring times of about 5ms.

scholarly journals A Generalization Performance Study Using Deep Learning Networks in Embedded Systems

Sensors ◽  
2021 ◽  
Vol 21 (4) ◽  
pp. 1031
Author(s):  
Joseba Gorospe ◽  
Rubén Mulero ◽  
Olatz Arbelaitz ◽  
Javier Muguerza ◽  
Miguel Ángel Antón
Keyword(s):  
Deep Learning ◽  
Embedded Systems ◽  
Embedded System ◽  
General Purpose ◽  
Learning Networks ◽  
Performance Study ◽  
Learning Techniques ◽  
Wide Range ◽  
Learning Architectures

Deep learning techniques are being increasingly used in the scientific community as a consequence of the high computational capacity of current systems and the increase in the amount of data available as a result of the digitalisation of society in general and the industrial world in particular. In addition, the immersion of the field of edge computing, which focuses on integrating artificial intelligence as close as possible to the client, makes it possible to implement systems that act in real time without the need to transfer all of the data to centralised servers. The combination of these two concepts can lead to systems with the capacity to make correct decisions and act based on them immediately and in situ. Despite this, the low capacity of embedded systems greatly hinders this integration, so the possibility of being able to integrate them into a wide range of micro-controllers can be a great advantage. This paper contributes with the generation of an environment based on Mbed OS and TensorFlow Lite to be embedded in any general purpose embedded system, allowing the introduction of deep learning architectures. The experiments herein prove that the proposed system is competitive if compared to other commercial systems.

scholarly journals Comparing Deep-Learning Architectures and Traditional Machine-Learning Approaches for Satire Identification in Spanish Tweets

Mathematics ◽  
2020 ◽  
Vol 8 (11) ◽  
pp. 2075
Author(s):  
Óscar Apolinario-Arzube ◽  
José Antonio García-Díaz ◽  
José Medina-Moreira ◽  
Harry Luna-Aveiga ◽  
Rafael Valencia-García
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
User Interfaces ◽  
State Of The Art ◽  
Learning Approaches ◽  
Word Embeddings ◽  
Linguistic Features ◽  
Intended Meaning ◽  
Language User ◽  
Learning Architectures

Automatic satire identification can help to identify texts in which the intended meaning differs from the literal meaning, improving tasks such as sentiment analysis, fake news detection or natural-language user interfaces. Typically, satire identification is performed by training a supervised classifier for finding linguistic clues that can determine whether a text is satirical or not. For this, the state-of-the-art relies on neural networks fed with word embeddings that are capable of learning interesting characteristics regarding the way humans communicate. However, as far as our knowledge goes, there are no comprehensive studies that evaluate these techniques in Spanish in the satire identification domain. Consequently, in this work we evaluate several deep-learning architectures with Spanish pre-trained word-embeddings and compare the results with strong baselines based on term-counting features. This evaluation is performed with two datasets that contain satirical and non-satirical tweets written in two Spanish variants: European Spanish and Mexican Spanish. Our experimentation revealed that term-counting features achieved similar results to deep-learning approaches based on word-embeddings, both outperforming previous results based on linguistic features. Our results suggest that term-counting features and traditional machine learning models provide competitive results regarding automatic satire identification, slightly outperforming state-of-the-art models.

scholarly journals Investigation of optimal configurations of a convolutional neural network for the identification of objects in real-time

2019 ◽  
pp. 417-423
Author(s):  
M A Isayev ◽  
D A Savelyev
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Real Time ◽  
State Of The Art ◽  
Average Precision ◽  
The Core ◽  
Particular Solution ◽  
Optimal Configurations

The comparison of different convolutional neural networks which are the core of the most actual solutions in the computer vision area is considers in hhe paper. The study includes benchmarks of this state-of-the-art solutions by some criteria, such as mAP (mean average precision), FPS (frames per seconds), for the possibility of real-time usability. It is concluded on the best convolutional neural network model and deep learning methods that were used at particular solution.

scholarly journals Examining Deep Learning Architectures for Crime Classification and Prediction

Forecasting ◽  
2021 ◽  
Vol 3 (4) ◽  
pp. 741-762
Author(s):  
Panagiotis Stalidis ◽  
Theodoros Semertzidis ◽  
Petros Daras
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
Open Data ◽  
Training Data ◽  
Crime Prediction ◽  
Crime Types ◽  
Improved Performance ◽  
Learning Architectures ◽  
And Training ◽  
Crime Classification

In this paper, a detailed study on crime classification and prediction using deep learning architectures is presented. We examine the effectiveness of deep learning algorithms in this domain and provide recommendations for designing and training deep learning systems for predicting crime areas, using open data from police reports. Having time-series of crime types per location as training data, a comparative study of 10 state-of-the-art methods against 3 different deep learning configurations is conducted. In our experiments with 5 publicly available datasets, we demonstrate that the deep learning-based methods consistently outperform the existing best-performing methods. Moreover, we evaluate the effectiveness of different parameters in the deep learning architectures and give insights for configuring them to achieve improved performance in crime classification and finally crime prediction.

scholarly journals Sistem Pendeteksi Pejalan Kaki di Lingkungan Terbatas Berbasis SSD Mobilenetv1 Menggunakan Gambar 360° Ternormalisasi

2021 ◽  
Vol 3 ◽  
pp. 111-122
Author(s):  
Ni Nyoman Ayu Marlina ◽  
Denden Mohammad Ariffin ◽  
Arief Suryadi Satyawan ◽  
Mohammed Ikrom Asysyakuur ◽  
Muhammad Farhan Utamajaya ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Single Shot

Seiring dengan perkembangan zaman, setiap produsen mobil selalu menciptakan produkterbarunya lebih canggih. Ide ini kemudian melahirkan konsep kendaraan listrik otonom (KLO). Hal ini dimaksudkan untuk selalu menghadirkan kendaraan yang dapat memenuhi selera konsumen yang terus berkembang, disamping juga ramah lingkungan Kehadiran kendaraan listrik otonom pastinya akan dialami oleh Indonesia yang masyarakatnya sudah mulai bergantung pada alat transportasi mobil. Oleh sebab itu situasi ini mengharuskan kita bersiap menghadapi era Mobility in Society 5.0, dimana kita harus dapat menguasai teknologi pendukungnya. Kendaraan litrik otonom dapat terealisasi jika sistemnya mampu mendeteksi objek dengan baik. Oleh sebab itu pada penelitian ini dilakukan pengembangan sistem pendeteksi pejalan kaki berbasis deep learning dan memanfaatkan gambar 360°. Sistem software deteksi objek yang dibangun menggunakan Single Shot Multibox Detector (SSD) MobilenetV1, sedangkan hardware yang digunakan untuk pengembangan ini adalah Jetson AGX Xavier. Proses pengembangan yang dilakukan dimulai dari pengambilan gambar 360° ternormalisasi berisi informasi pejalan kaki di area kampus Universitas Nurtanio yang dipergunakan sebagai dataset dan data pengujian, melatih SSD MobileNetV1 dengan dataset tersebut (19.038), dan menguji model software terlatih secara real-time maupun offline.Hasil pengujian offline terhadap 735 gambar 360° pada kondisi siang hari menunjukan bahwa55,5% gambar dapat terdeteksi sempurna, sedangkan dari 595 gambar 360° pada kondisi sore hari, 51,2% gambar dapat terdeteksi sempurna. Pada pengujian secara real-time diperoleh kepastian bahwa 98% pejalan kaki pada siang hari terdeteksi, sedangkan pada sore hari hanya 95%. Waktu proses rata-rata pada sebuah gambar kondisi siang hari adalah 32,81283 ms jika menggunakan CPU, sedangkanjika menggunakan GPU adalah 32,79766 ms. Untuk sebuah gambar dengan informasi yang sama pada kondisi sore hari diperoleh waktu proses 37,42598 ms jika menggunakan CPU, sedangkan jika menggunakan GPU adalah 37,45174 ms.

scholarly journals Real-Time Monitoring of COVID-19 SOP in Public Gathering Using Deep Learning Technique

2021 ◽  
Vol 5 ◽  
pp. 182-196
Author(s):  
Muhammad Haris Kaka Khel ◽  
Kushsairy Kadir ◽  
Waleed Albattah ◽  
Sheroz Khan ◽  
MNMM Noor ◽  
...  
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Social Distance ◽  
Virus Transmission ◽  
World Health ◽  
Single Shot ◽  
Accuracy Score ◽  
Computationally Efficient ◽  
Safe Distance ◽  
Crowd Management

Crowd management has attracted serious attention under the prevailing pandemic conditions of COVID-19, emphasizing that sick persons do not become a source of virus transmission. World Health Organization (WHO) guidelines include maintaining a safe distance and wearing a mask in gatherings as part of standard operating procedures (SOP), considered thus far the most effective preventive measures to protect against COVID-19. Several methods and strategies have been used to construct various face detection and social distance detection models. In this paper, a deep learning model is presented to detect people without masks and those not keeping a safe distance to contain the virus. It also counts individuals who violate the SOP. The proposed model employs the Single Shot Multi-box Detector as a feature extractor, followed by Spatial Pyramid Pooling (SPP) to integrate the extracted features to improve the model's detecting capabilities. The MobilenetV2 architecture as a framework for the classifier makes the model highly light, fast, and computationally efficient, allowing it to be employed in embedded devices to do real-time mask and social distance detection, which is the sole objective of this research. This paper's technique yields an accuracy score of 99% and reduces the loss to 0.04%. Doi: 10.28991/esj-2021-SPER-14 Full Text: PDF

scholarly journals PEDESTRIAN SAFETY EVALUATION OF SIGNALIZED INTERSECTIONS USING SURROGATE SAFETY MEASURES

Transport ◽  
2020 ◽  
Vol 35 (1) ◽  
pp. 48-56
Author(s):  
Sankaran Marisamynathan ◽  
Perumal Vedagiri
Keyword(s):  
Safety Evaluation ◽  
Pedestrian Safety ◽  
Signalized Intersections ◽  
Signalized Intersection ◽  
Mixed Traffic ◽  
Model Framework ◽  
Safety Measures ◽  
Severity Level ◽  
Traffic Conditions ◽  
Surrogate Safety Measures

The large proportions of pedestrian fatalities led researchers to make the improvements of pedestrian safety at intersections. Thus, this paper proposes a methodology to evaluate crosswalk safety at signalized intersections using Surrogate Safety Measures (SSM) under mixed traffic conditions. The required pedestrian, traffic, and geometric data were extracted based on the videographic survey conducted at signalized intersections in Mumbai (India). Post Encroachment Time (PET) for each pedestrian were segregated into three categories for estimating pedestrian–vehicle interactions and Cumulative Frequency Distribution (CDF) was plotted to calculate the threshold values for each interaction severity level. The Cumulative Logistic Regression (CLR) model was developed to predict the pedestrian mean PET values in the cross-walk at signalized intersections. The proposed model was validated with a new signalized intersection and the results were shown that the proposed PET ranges and model appropriate for Indian mixed traffic conditions. To assess the suitability of model framework, model transferability was carried out with data collected at signalized intersection in Kolkata (India). Finally, this study can be helpful to rank the severity level of pedestrian safety in the crosswalk and improve the existing facilities at signalized intersections.

scholarly journals Road Object Detection: A Comparative Study of Deep Learning-Based Algorithms

Electronics ◽  
2021 ◽  
Vol 10 (16) ◽  
pp. 1932
Author(s):  
Malik Haris ◽  
Adam Glowacz
Keyword(s):  
Image Processing ◽  
Deep Learning ◽  
Object Detection ◽  
Real Time ◽  
Large Scale ◽  
Single Shot ◽  
Automated Driving ◽  
Convolutional Network ◽  
Image Processing Algorithms ◽  
Processing Algorithms

Automated driving and vehicle safety systems need object detection. It is important that object detection be accurate overall and robust to weather and environmental conditions and run in real-time. As a consequence of this approach, they require image processing algorithms to inspect the contents of images. This article compares the accuracy of five major image processing algorithms: Region-based Fully Convolutional Network (R-FCN), Mask Region-based Convolutional Neural Networks (Mask R-CNN), Single Shot Multi-Box Detector (SSD), RetinaNet, and You Only Look Once v4 (YOLOv4). In this comparative analysis, we used a large-scale Berkeley Deep Drive (BDD100K) dataset. Their strengths and limitations are analyzed based on parameters such as accuracy (with/without occlusion and truncation), computation time, precision-recall curve. The comparison is given in this article helpful in understanding the pros and cons of standard deep learning-based algorithms while operating under real-time deployment restrictions. We conclude that the YOLOv4 outperforms accurately in detecting difficult road target objects under complex road scenarios and weather conditions in an identical testing environment.

scholarly journals Ship Detection Based on YOLOv2 for SAR Imagery

2019 ◽  
Vol 11 (7) ◽  
pp. 786 ◽  
Author(s):  
Yang-Lang Chang ◽  
Amare Anagaw ◽  
Lena Chang ◽  
Yi Wang ◽  
Chih-Yu Hsiao ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Real Time ◽  
Experimental Results ◽  
Detection Methods ◽  
Computational Time ◽  
Detection Accuracy ◽  
Single Shot ◽  
Ship Detection ◽  
Sar Imagery

Synthetic aperture radar (SAR) imagery has been used as a promising data source for monitoring maritime activities, and its application for oil and ship detection has been the focus of many previous research studies. Many object detection methods ranging from traditional to deep learning approaches have been proposed. However, majority of them are computationally intensive and have accuracy problems. The huge volume of the remote sensing data also brings a challenge for real time object detection. To mitigate this problem a high performance computing (HPC) method has been proposed to accelerate SAR imagery analysis, utilizing the GPU based computing methods. In this paper, we propose an enhanced GPU based deep learning method to detect ship from the SAR images. The You Only Look Once version 2 (YOLOv2) deep learning framework is proposed to model the architecture and training the model. YOLOv2 is a state-of-the-art real-time object detection system, which outperforms Faster Region-Based Convolutional Network (Faster R-CNN) and Single Shot Multibox Detector (SSD) methods. Additionally, in order to reduce computational time with relatively competitive detection accuracy, we develop a new architecture with less number of layers called YOLOv2-reduced. In the experiment, we use two types of datasets: A SAR ship detection dataset (SSDD) dataset and a Diversified SAR Ship Detection Dataset (DSSDD). These two datasets were used for training and testing purposes. YOLOv2 test results showed an increase in accuracy of ship detection as well as a noticeable reduction in computational time compared to Faster R-CNN. From the experimental results, the proposed YOLOv2 architecture achieves an accuracy of 90.05% and 89.13% on the SSDD and DSSDD datasets respectively. The proposed YOLOv2-reduced architecture has a similarly competent detection performance as YOLOv2, but with less computational time on a NVIDIA TITAN X GPU. The experimental results shows that the deep learning can make a big leap forward in improving the performance of SAR image ship detection.

Sign in / Sign up

Export Citation Format

Share Document