EHLM: Empirical Design of Novel Road Curve and Lane Identification Scheme using Effective Hybrid Learning Methodology DOI

S. Sudha Mercy,

G. Ramya,

Ravi Sankar

et al.

2022 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems (ICSES), Journal Year: 2023, Volume and Issue: unknown, P. 1 - 8

Published: Dec. 14, 2023

In this study, we present a novel Road Curve and Lane Identification Scheme that harnesses the power of an Effective Hybrid Learning Methodology (EHLM). This advanced approach combines Convolutional Neural Networks (CNN), Mask R-CNN, ResNet, creating formidable framework for road curve detection lane identification in complex driving scenarios. The EHLM offers versatile solution excels detecting curves accurately identifying lanes, crucial components autonomous systems driver assistance. It leverages strengths each architecture, from CNN's feature extraction capabilities to R-CNN's precise instance segmentation ResNet's deep learning prowess. study provides comprehensive overview approach, showcasing its efficacy real-world Through extensive experimentation evaluation, demonstrate superiority our methodology, achieving identification. Our research contributes development safer more efficient vehicles, ultimately enhancing safety transportation systems.In have considered several models, including CNN, DCNN, MRCNN, CNN-LSTM, ANN, Proposed Model. Among these contenders, Model stands out prominently terms accuracy, impressive 97.23%. indicates remarkable ability correctly classify recognize target elements within dataset.

Language: Английский

AI-Powered Automated Wheelchair with Lane Detection DOI

B Abiraj,

K. R. Sethuraman,

J Bethanney Janney

et al.

2021 International Conference on Emerging Smart Computing and Informatics (ESCI), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 5

Published: March 5, 2024

A wheelchair is a revolutionary assistive mobility aid that improves the flexibility and standard of life for those with limitations. There are many guidance systems impaired people to navigate against obstacles other dangers confronted rapidly safely. With developments modern techniques, there distinct kinds devices available help mobility. The key objective proposed study create technologically intelligent can recognize voice, gesture, facial, lane, gaze in natural setting individuals who have impairments. In this paper, an automatic based on lane detection shown, which made real-time using hardware through use deep learning algorithms image manipulation. device also promotes overall well-being health-tracking capabilities. It offers communication support unable speak. Empower Wheel Chair redefines accessibility, providing users newfound freedom, convenience, connection their daily lives.

Language: Английский

Citations

0
Towards Safer Driving: ADAS Enhancements in Lane Keeping and Sign Detection DOI
Mona Helmy,

Abdalla Ibrahim,

Amr Helaly

et al.

2022 International Telecommunications Conference (ITC-Egypt), Journal Year: 2024, Volume and Issue: unknown, P. 78 - 84

Published: July 22, 2024

Language: Английский

Citations

0
Improve Lane Line Detection Performance Through Infrared and Visible Image Fusion DOI
Jinxing Zhao,

Haolan Zheng,

Jinfeng Bo

et al.

Published: Jan. 1, 2024

Language: Английский

Citations

0
Transfer Learning-based Lane Line Detection System for Visual Path Following Control DOI
Ana Almeida, Tiago T. Ribeiro, André G. S. Conceição

et al.

2022 IEEE 18th International Conference on Automation Science and Engineering (CASE), Journal Year: 2024, Volume and Issue: unknown, P. 782 - 787

Published: Aug. 28, 2024

Language: Английский

Citations

0
Vehicle road detection based on pyramid network and self-attention mechanism DOI
Junting Zhang

Multiagent and Grid Systems, Journal Year: 2024, Volume and Issue: 20(3-4), P. 203 - 217

Published: Nov. 1, 2024

Aiming at the imbalance between accuracy and real-time performance of lane detection, this paper proposes a novel vehicle road detection based on pyramid network self-attention mechanism (namely PFSA). In method, two-sided multi-scale fusion is used to realise information exchange shallow features deep features, obtain contextual semantics. A new asymmetric convolution module proposed, which fuses into cavity layers with different expansion rates improve feature extraction capability reduce computation. two-stage training method was train public data set compared other advanced methods. The experimental results show that 98.3% speed 18.391 frames per second (fps), much better than

Language: Английский

Citations

0
Adaptive Multilane Detection for Autonomous Driving in Different Visibility Conditions Using Deep Learning DOI

Rajvi Jasani,

Gaurav Singh, Shashank Mouli Satapathy

et al.

Published: Jan. 1, 2023

Presently, road or lane detection to monitor and perform navigation is among the most critical obstacles in Autonomous Driving Assistance Systems. Lane used accomplish various tasks, such as avoidance of crashes, vehicle navigation, departure warning systems. Numerous works literature utilize both traditional deep learning techniques detect roads lanes. Traditional methods include those based on computer vision curve modeling techniques. These are less computationally intensive compared ones, but do not well. Thus, we have a solution that well mixed scenarios structured unstructured roads, bad weather traffic conditions, etc. Even models expensive thus cannot work an embedded system for real-world applications. This paper tries fill this gap by developing adaptive, lightweight model using learning. Our approach uses image segmentation ability You Only Look Once version 7 model. It provides us with speed result its single-shot detection. includes two trained different conditions incorporated into our Mobile Net 2 classifier. done increase robustness can be easily expanded adding more individual specific conditions. The been tested data obtained from adverse ensure better results real world.

Language: Английский

Citations

0
A Comprehensive Review on Algorithms of Image Processing for Autonomous and Electric Vehicles DOI

Ismayel Gollapudi,

Kallol Bhaumik, Digvijay Pandey

et al.

Advances in computational intelligence and robotics book series, Journal Year: 2023, Volume and Issue: unknown, P. 39 - 55

Published: June 30, 2023

The development of autonomous electric vehicles has gained significant attention due to their potential reduce carbon emissions and improve road safety. Image processing become an important tool in the these vehicles, enabling them detect respond objects obstacles environment. In this review paper, we explore use image driverless cars, with a focus on various techniques proposed by authors. comparison performance effectiveness different approaches, including deep learning, computer vision, sensor fusion, detecting recognizing Our highlights advantages limitations each technique for future field vehicles. Overall, shown be promising solution safe efficient

Language: Английский

Citations

0
Detecting Danger: AI-Enabled Road Crack Detection for Autonomous Vehicles DOI Creative Commons

Raza Alisha,

Khatua Debnarayan,

Dutta Rachaita

et al.

E3S Web of Conferences, Journal Year: 2023, Volume and Issue: 430, P. 01160 - 01160

Published: Jan. 1, 2023

The present article proposes the deep learning concept termed ―Faster-Region Convolutional Neural Network‖ (Faster-RCNN) technique to detect cracks on road for autonomous cars. Feature extraction, preprocessing, and classification techniques have been used in this study. Several types of image datasets, such as camera images, faster-RCNN laser real-time considered. With help GPU (graphics processing unit), input is processed. Thus, density measured information regarding acquired. This model aims determine crack precisely compared existing techniques.

Language: Английский

Citations

0
A Sliding Mode Approach-Based Adaptive Steering Control Algorithm for Path Tracking of Autonomous Mobility with Weighted Injection DOI Creative Commons
Sehwan Kim, Kwangseok Oh

Machines, Journal Year: 2023, Volume and Issue: 11(10), P. 972 - 972

Published: Oct. 18, 2023

The increasing complexity of mathematical models developed as part the recent advancements in autonomous mobility platforms has led to an escalation uncertainty. Despite intricate nature such models, detection, decision, and control methods for path tracking remain critical. This study aims achieve based on pixel-based errors without parameters model. proposed approach entails deriving from a multi-particle filter camera, estimating error dynamics coefficients through recursive least squares (RLS) approach, using sliding mode weighted injection formulate cost function that leverages estimated errors. resultant adaptive steering expedites convergence towards zero by determining magnitude variable finite-time condition. efficacy is evaluated S-curved elliptical equipped with single driving module. results demonstrate capability reasonably track target paths facilitated lidar-based obstacle detection system.

Language: Английский

Citations

0
AR-NET: lane detection model with feature balance concerns for autonomous driving DOI
Guoxiang Tong,

Chuanye Zu

Neural Computing and Applications, Journal Year: 2023, Volume and Issue: 36(8), P. 3997 - 4012

Published: Dec. 8, 2023

Language: Английский

Citations

0