Comparative Analysis of Intelligent Driving and Safety Assistance Systems Using YOLO and SSD Model of Deep Learning

2021 ◽  
Vol 12 (1) ◽  
pp. 131-146
Author(s):  
Nidhi Sindhwani ◽  
Shekhar Verma ◽  
Tushar Bajaj ◽  
Rohit Anand

Bad road conditions are one of the main causes of road accidents around the world. These kinds of accidents prove to be fatal as many lives are lost in these accidents that are mainly caused by potholes or distress on surface of roads. This paper suggests a system that will not only help in reducing the chances of these accidents by making the driver aware of the upcoming distress/potholes on the road but also saving the location of these potholes which can be sent to respective authorities so that they can be repaired. The authors have used technologies like image processing, computer vision, deep learning, and internet of things (IoT) to make this happen. It uses a camera mounted in front near windshield that will capture the images which will be further be processed to get the location of the potholes and distress on road. These detected potholes can be projected on a heads-up display (HUD) placed near windshield which will notify the driver of the potholes.

Author(s):  
Shubham Naphade ◽  
Sumit Jare ◽  
Rohan Chavan ◽  
Nayan Nawale ◽  
Dr. Shwetambari Chiwhane

In our day to day life there are several civic issues which are being faced by each& every person in the world.& if we take countries just like India or any other country where population is too much, in such countries there are several civic issues which are faced by each citizen in such countries. Larger the country larger the issues, because there are several complaints regarding a single issue. If we take just an example out of those like street damages, garbage management problems (garbage bin over owing), Electricity problem, Water problem etc. For that there is a system also available in such countries but problem is just like that it’s time consuming. If any one of us go to register complaint into the municipality then they will register our complaint, after registration there are several days required to solve their civic issues of each individual person who are residing at that place. So, from this project we are just reducing the time of the process from both sides, i.e. if any one of the persons sees that garbage or can say that potholes on the road but he / she neglect that& move on, just because of they think that when will we go to the municipality & when they will register complaint against it. Nobody has a time in today’s 21stcentury. There is another kind of people also present in respective countries, they are very sciatic who really wants to act regarding such as garbage wasting or any other issues. They register their complaint but they don’t know the status of their complaint, just because of that they leave from that topic & easily move on. So, this is just an effort for finding remedy for all above mentioned issues.


2019 ◽  
Vol 1 (2) ◽  
pp. 11-18
Author(s):  
V Sowndharya ◽  
Hebziba Jeba Rani S ◽  
Susmitha D

In today’s era one of the most common problems which the world is facing is an exponential increase in population. This has indirectly increased a lot of other issues; one of them being the quantity of vehicles on the road. The increased number of vehicles results in shortage of parking areas. This project aims to present an intelligent parking system for vehicles that identifies the parking slot automatically through sensors and displays it without making the drivers to circle around the parking area. The availability of parking slots will be displayed to the drivers at the  entrance.  It also captures the number plate of vehicles by using camera and recognises the number using image processing and stores it in the server at the entrance and also at the exit of parking area for ease of payment purposes. All the information’s will be simultaneously updated in the IoT server and can be used for future use. It is found that the system decreases the manual work and provides high efficiency and high accuracy.


2021 ◽  
Author(s):  
Bianka Tallita Passos ◽  
Wemerson Delcio Parreira ◽  
Anita Maria da Rocha Fernandes ◽  
Eros Comunello

The road infrastructure conditions are directly related to the safetyand operational cost of transportation. Potholes are defects in thepaving that affect safety on the road. Therefore, identifying potholesis an important step in defining road maintenance and interventionstrategies. Among the approaches used to detect defects in roadsare vibration techniques, laser scanning and 3D reconstruction, andfinally methods that are vision-based. These vision-based methodsutilize image processing, considered low cost and that can be performedby common two-dimensional cameras. This research aimsto combine digital image processing and deep learning concepts facilitatingthe recognition of pothole-like defects in road images withasphalt paving. In order to carry out these experiments, differentnetwork architectures were used.


2015 ◽  
Vol 77 (29) ◽  
Author(s):  
Nassiriah Shaari ◽  
Aeni Zuhana Saidin ◽  
Asmidah Alwi

Road safety campaigns and programs have been extensively introduced and implemented in Malaysia. However, their effectiveness is still being debated. Children especially will become the unfortunate victims of road accidents if they are unaware of the danger and precaution actions to be safe on the road. In response to that, this paper introduces an application as an alternative that inculcates road safety awareness to further support existing related programs and campaigns. Particularly, an interactive web application incorporating interactive multimedia elements has been designed and evaluated. Results on the usability test indicate a promising success and highlight aspects and issues that can be further focused for improvement and enhancement. 


2015 ◽  
Vol 5 (1) ◽  
pp. 133-149
Author(s):  
Juhan Värk

AbstractOn 15 November 2012, at the plenary session of the Central Committee of the Communist Party of China, Xi Jinping was elected the Party' general secretary, whereas he also became the chairman of the influential Central Military Commission. Too eager to wait to be inaugurated as President of the People' Republic of China in March 2013, the new national leader announced that in the following decade he is guided by the main objective of his predecessor Hu Jintao to double the prosperity of the people by the year 2020 and to keep the country' economy stable and growing fast. Unfortunately, it will be difficult for the new leader of China to implement his intentions, since, presumably, the country' new leadership will be from the older generation, hardliners, and, most importantly, politically conservative. But the difficulties lie in carrying out economic reforms because of rampant corruption and shadow banking in the central apparatus and in regions.According to the World Bank' analysis, China has become the world' largest economy. But the large expenditures to military reform and environmental protection are not sufficient for Xi Jinping to accomplish the goal to raise significantly the poor living standards of Chinese people. China has also faced difficulties in complying with the basic principles of its foreign policy, especially after the annexation of Crimea and its incorporation into the Russian Federation by China' strategic partner, Russia. In the past, China has carried out campaigns against “Americanization”, although with no tangible results. Thus, the leaders of today' China are faced with a number of dilemmas.


2018 ◽  
Vol 7 (2.7) ◽  
pp. 614 ◽  
Author(s):  
M Manoj krishna ◽  
M Neelima ◽  
M Harshali ◽  
M Venu Gopala Rao

The image classification is a classical problem of image processing, computer vision and machine learning fields. In this paper we study the image classification using deep learning. We use AlexNet architecture with convolutional neural networks for this purpose. Four test images are selected from the ImageNet database for the classification purpose. We cropped the images for various portion areas and conducted experiments. The results show the effectiveness of deep learning based image classification using AlexNet.  


2021 ◽  
Author(s):  
Henk-Jan Dekker

In an effort to fight climate change, many cities try to boost their cycling levels. They often look towards the Dutch for guidance. However, historians have only begun to uncover how and why the Netherlands became the premier cycling country of the world. Why were Dutch cyclists so successful in their fight for a place on the road? Cycling Pathways: The Politics and Governance of Dutch Cycling Infrastructure, 1920-2020 explores the long political struggle that culminated in today’s high cycling levels. Delving into the archives, it uncovers the important role of social movements and shows in detail how these interacted with national, provincial, and urban engineers and policymakers to govern the distribution of road space and construction of cycling infrastructure. It discusses a wide range of topics, ranging from activists to engineering committees, from urban commuters to recreational cyclists and from the early 1900s to today in order to uncover the long and all-but-forgotten history of Dutch cycling governance.


2021 ◽  
Vol 46 (2) ◽  
pp. 11-25
Author(s):  
Jolanta Brzykcy

The article is an analysis of the poetry of Gisella Lachman (1895–1969), poet of the “first wave” of Russian emigration, from the perspective of the poetics of space. The poet expressed her emigration experience (multiple changes of residence: Russia, Germany, Switzerland, USA) in her poems in spatial relations. They appear on different levels of the works’ morphology: in the construction of the lyrical “I”, in the organisation of the presented world, in the repertoire of motifs and the selection of poetic lexis and genre forms. Space plays a literal role in Lachman’s poetry; it is a representation of extra-literary reality, seen subjectively. It is also subject to metaphorisation, becoming a tool for expressing philosophical content. The poet creates not only a spatial model of the world, but also a spatial model of human life, which she perceives as a transit on the road to eternity.


Author(s):  
Mr. Aniket Ashok Bhamani ◽  
Mr. Sanyam Sanjay Mehta

There are a lot of road accidents that occur due to drowsy driving. Drowsy driving is when the driver of a vehicle is found to be sleepy and probable to get into a car crash because of the same. Being drowsy might cause the driver to lose concentration from the road, and also reduce the reaction time. Statistics suggest how thousands of deaths and crashes happen every year due to it. Major victims of such crashes tend to be the commercial drivers who need to drive long distances overnight. Our project intends to propose a solution to this problem by providing an Internet of Things based approach. This approach monitors the driver’s face while he or she is driving the vehicle and in case if the driver is to be found falling asleep, an instant voice call is made to the driver’s registered phone number. Additionally, a text message is also sent to the driver’s emergency contact which will get him/her notified and provide the driver with quick assistance if needed. This approach is unique and different in its own way as it provides cross platform support and remote monitoring of the driver. Additionally, it also makes drowsy-detection ‘device independent’. It offers a simplified mechanism to derive real time accurate results and readings with reduced complexities. This project does have a lot of scope, especially considering that there is a lack of methodologies currently being implemented to prevent road accidents due to drowsy driving. KEYWORDS- Drowsy Driving, Monitoring, Machine Learning, Internet of Things, Remote, Algorithm, Eye Aspect Ratio, Python.


Sign in / Sign up

Export Citation Format

Share Document