An Accident Detection and Classification System Using Internet of Things and Machine Learning towards Smart City

Daily traffic accidents increase annually, causing a significant number of death and disability cases. Most of fatalities occur because of the late response to these emergency cases. The time after the traumatic injury is called the golden hour, where providing essential medical and surgical aid at that time increases the probability of saving human lives by one-third an average. Thus, the focus of this paper was to develop a system based on IoT for accident detection and classification. The system detects and classifies vehicle accidents based on severity level and reports the essential information about the accident to emergency services providers. The system consists of a microcontroller, GPS, and a group of sensors to determine different physical parameters related to vehicle motion. In addition, different types of machine learning classifiers were examined with the developed system to determine the most accurate classifier for the system. The classifiers are the Gaussian Mixture Model (GMM), Naive-Bayes Tree (NB), Decision Tree (DT), and Classification and Regression Trees (CART). The implementation of the system showed that GMM and CART models were better in terms of precision and recall. It was also shown that the severity of accidents depends mainly on the g-force value and fire occurrence.

Download Full-text

Multiple vehicles detection and tracking for intelligent transport systems using machine learning approaches

Transport and Communication Science Journal ◽

10.25073/tcsj.70.3.7 ◽

2019 ◽

Vol 70 (3) ◽

pp. 214-224

Author(s):

Bui Ngoc Dung ◽

Manh Dzung Lai ◽

Tran Vu Hieu ◽

Nguyen Binh T. H.

Keyword(s):

Machine Learning ◽

Gaussian Mixture ◽

Research Field ◽

Transport Systems ◽

Learning Approaches ◽

Subtraction Method ◽

Intelligent Transport Systems ◽

Intelligent Transport ◽

Detection And Tracking ◽

Multiple Vehicles

Video surveillance is emerging research field of intelligent transport systems. This paper presents some techniques which use machine learning and computer vision in vehicles detection and tracking. Firstly the machine learning approaches using Haar-like features and Ada-Boost algorithm for vehicle detection are presented. Secondly approaches to detect vehicles using the background subtraction method based on Gaussian Mixture Model and to track vehicles using optical flow and multiple Kalman filters were given. The method takes advantages of distinguish and tracking multiple vehicles individually. The experimental results demonstrate high accurately of the method.

Download Full-text

A Comparative Survey of Feature Extraction and Machine Learning Methods in Diverse Acoustic Environments

Sensors ◽

10.3390/s21041274 ◽

2021 ◽

Vol 21 (4) ◽

pp. 1274

Author(s):

Daniel Bonet-Solà ◽

Rosa Ma Alsina-Pagès

Keyword(s):

Machine Learning ◽

Feature Extraction ◽

Best Practice ◽

Nearest Neighbor ◽

Gaussian Mixture ◽

Machine Learning Algorithms ◽

Multimedia Retrieval ◽

Natural Environments ◽

K Nearest Neighbor ◽

Acoustic Environments

Acoustic event detection and analysis has been widely developed in the last few years for its valuable application in monitoring elderly or dependant people, for surveillance issues, for multimedia retrieval, or even for biodiversity metrics in natural environments. For this purpose, sound source identification is a key issue to give a smart technological answer to all the aforementioned applications. Diverse types of sounds and variate environments, together with a number of challenges in terms of application, widen the choice of artificial intelligence algorithm proposal. This paper presents a comparative study on combining several feature extraction algorithms (Mel Frequency Cepstrum Coefficients (MFCC), Gammatone Cepstrum Coefficients (GTCC), and Narrow Band (NB)) with a group of machine learning algorithms (k-Nearest Neighbor (kNN), Neural Networks (NN), and Gaussian Mixture Model (GMM)), tested over five different acoustic environments. This work has the goal of detailing a best practice method and evaluate the reliability of this general-purpose algorithm for all the classes. Preliminary results show that most of the combinations of feature extraction and machine learning present acceptable results in most of the described corpora. Nevertheless, there is a combination that outperforms the others: the use of GTCC together with kNN, and its results are further analyzed for all the corpora.

Download Full-text

Machine Learning in Medical Emergencies: a Systematic Review and Analysis

Journal of Medical Systems ◽

10.1007/s10916-021-01762-3 ◽

2021 ◽

Vol 45 (10) ◽

Author(s):

Inés Robles Mendo ◽

Gonçalo Marques ◽

Isabel de la Torre Díez ◽

Miguel López-Coronado ◽

Francisco Martín-Rodríguez

Keyword(s):

Machine Learning ◽

Systematic Review ◽

Language Processing ◽

Emergency Services ◽

Clinical Decision ◽

Quality Of Healthcare ◽

The Other ◽

Other Hand ◽

Increasing Demand ◽

Commercial Applications

AbstractDespite the increasing demand for artificial intelligence research in medicine, the functionalities of his methods in health emergency remain unclear. Therefore, the authors have conducted this systematic review and a global overview study which aims to identify, analyse, and evaluate the research available on different platforms, and its implementations in healthcare emergencies. The methodology applied for the identification and selection of the scientific studies and the different applications consist of two methods. On the one hand, the PRISMA methodology was carried out in Google Scholar, IEEE Xplore, PubMed ScienceDirect, and Scopus. On the other hand, a review of commercial applications found in the best-known commercial platforms (Android and iOS). A total of 20 studies were included in this review. Most of the included studies were of clinical decisions (n = 4, 20%) or medical services or emergency services (n = 4, 20%). Only 2 were focused on m-health (n = 2, 10%). On the other hand, 12 apps were chosen for full testing on different devices. These apps dealt with pre-hospital medical care (n = 3, 25%) or clinical decision support (n = 3, 25%). In total, half of these apps are based on machine learning based on natural language processing. Machine learning is increasingly applicable to healthcare and offers solutions to improve the efficiency and quality of healthcare. With the emergence of mobile health devices and applications that can use data and assess a patient's real-time health, machine learning is a growing trend in the healthcare industry.

Download Full-text

Automobile tire life prediction based on image processing and machine learning technology

Advances in Mechanical Engineering ◽

10.1177/16878140211002727 ◽

2021 ◽

Vol 13 (3) ◽

pp. 168781402110027

Author(s):

Jianchen Zhu ◽

Kaixin Han ◽

Shenlong Wang

Keyword(s):

Machine Learning ◽

Image Processing ◽

Life Prediction ◽

Traffic Accidents ◽

Confusion Matrix ◽

Small Sample ◽

Image Feature ◽

Engineering Practice ◽

Learning Technology ◽

K Nearest Neighbor

With economic growth, automobiles have become an irreplaceable means of transportation and travel. Tires are important parts of automobiles, and their wear causes a large number of traffic accidents. Therefore, predicting tire life has become one of the key factors determining vehicle safety. This paper presents a tire life prediction method based on image processing and machine learning. We first build an original image database as the initial sample. Since there are usually only a few sample image libraries in engineering practice, we propose a new image feature extraction and expression method that shows excellent performance for a small sample database. We extract the texture features of the tire image by using the gray-gradient co-occurrence matrix (GGCM) and the Gauss-Markov random field (GMRF), and classify the extracted features by using the K-nearest neighbor (KNN) classifier. We then conduct experiments and predict the wear life of automobile tires. The experimental results are estimated by using the mean average precision (MAP) and confusion matrix as evaluation criteria. Finally, we verify the effectiveness and accuracy of the proposed method for predicting tire life. The obtained results are expected to be used for real-time prediction of tire life, thereby reducing tire-related traffic accidents.

Download Full-text

Design and Development of Lubricating Material Database and Research on Performance Prediction Method of Machine Learning

Scientific Reports ◽

10.1038/s41598-019-56776-2 ◽

2019 ◽

Vol 9 (1) ◽

Cited By ~ 3

Author(s):

Dan Jia ◽

Haitao Duan ◽

Shengpeng Zhan ◽

Yongliang Jin ◽

Bingxue Cheng ◽

...

Keyword(s):

Machine Learning ◽

High Performance ◽

Prediction Method ◽

Lubricating Oil ◽

Physical Parameters ◽

First Principle ◽

Vast Number ◽

First Principle Calculation ◽

Lubricating Material ◽

Machine Learning Model

AbstractLong developing period and cumbersome evaluation for the lubricating materials performance seriously jeopardize the successful development and application of any database system in tribological field. Such major setback can be solved effectively by implementing approaches with high throughput calculation. However, it often involves with vast number of output files, which are computed on the basis of first principle computation, having different data format from that of their experimental counterparts. Commonly, the input, storage and management of first principle calculation files and their individually test counterparts, implementing fast query and display in the database, adding to the use of physical parameters, as predicted with the performance estimated by first principle approach, may solve such setbacks. Investigation is thus performed for establishing database website specifically for lubricating materials, which satisfies both data: (i) as calculated on the basis of first principles and (ii) as obtained by practical experiment. It further explores preliminarily the likely relationship between calculated physical parameters of lubricating oil and its respectively tribological and anti-oxidative performance as predicted by lubricant machine learning model. Success of the method facilitates in instructing the obtainment of optimal design, preparation and application for any new lubricating material so that accomplishment of high performance is possible.

Download Full-text

Mining Accident Detection Using Machine Learning Methods

IFAC Proceedings Volumes ◽

10.3182/20130825-4-us-2038.00051 ◽

2013 ◽

Vol 46 (16) ◽

pp. 31-33

Author(s):

Francisco Santibáñez ◽

Carlos Flores ◽

Franco Basso ◽

Abelino Jiménez ◽

Francisco Bravo ◽

...

Keyword(s):

Machine Learning ◽

Learning Methods ◽

Machine Learning Methods ◽

Accident Detection

Download Full-text

Predicting carbon nanotube forest attributes and mechanical properties using simulated images and deep learning

npj Computational Materials ◽

10.1038/s41524-021-00603-8 ◽

2021 ◽

Vol 7 (1) ◽

Cited By ~ 1

Author(s):

Taher Hajilounezhad ◽

Rina Bao ◽

Kannappan Palaniappan ◽

Filiz Bunyak ◽

Prasad Calyam ◽

...

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Carbon Nanotube ◽

High Throughput ◽

Self Assembly ◽

Mechanical Performance ◽

Physical Parameters ◽

Structure Property ◽

Processing Parameter ◽

Physics Based Simulation

AbstractUnderstanding and controlling the self-assembly of vertically oriented carbon nanotube (CNT) forests is essential for realizing their potential in myriad applications. The governing process–structure–property mechanisms are poorly understood, and the processing parameter space is far too vast to exhaustively explore experimentally. We overcome these limitations by using a physics-based simulation as a high-throughput virtual laboratory and image-based machine learning to relate CNT forest synthesis attributes to their mechanical performance. Using CNTNet, our image-based deep learning classifier module trained with synthetic imagery, combinations of CNT diameter, density, and population growth rate classes were labeled with an accuracy of >91%. The CNTNet regression module predicted CNT forest stiffness and buckling load properties with a lower root-mean-square error than that of a regression predictor based on CNT physical parameters. These results demonstrate that image-based machine learning trained using only simulated imagery can distinguish subtle CNT forest morphological features to predict physical material properties with high accuracy. CNTNet paves the way to incorporate scanning electron microscope imagery for high-throughput material discovery.

Download Full-text