nanoll extt
Please use this identifier to cite or link to this item: http://lrcdrs.bennett.edu.in:80/handle/123456789/4528
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSharma, Shraddha-
dc.contributor.authorVerma, Naaz-
dc.date.accessioned2024-05-30T11:37:50Z-
dc.date.available2024-05-30T11:37:50Z-
dc.date.issued2023-
dc.identifier.issn978-93-5053-920-0-
dc.identifier.urihttp://lrcdrs.bennett.edu.in:80/handle/123456789/4528-
dc.description.abstractThe considerable research and development efforts have been mostly focused on self-driving automobiles, owing to their potential to profoundly transform transportation networks on a worldwide scale. A crucial obstacle in the development of autonomous vehicles is guaranteeing their safety and dependability, which is intricately linked to precise and resilient object recognition under diverse environmental circum- stances. This research introduces an innovative method to enhance object identification for self-driving cars that operate in challenging settings such as low-light, foggy, or wet surroundings. Our strategy integrates three state-of-the-art techniques: deep learning-based object detection, YOLO algorithms, and satellite-based GAN models._x000D_ We validate the efficacy of our method by conducting thorough test- ing and evaluation on a publicly available dataset. We used YOLO techniques in the object detection process to enhance precision and re- silience by considering measurement error and noise. We have assessed our methodology by utilizing openly accessible datasets and replicat- ing diverse extreme scenarios, such as low-light, foggy, and wet circum- stances. Our findings demonstrate that the suggested methodology sub- stantially enhances the accuracy of object detection in challenging circumstances when compared to current techniques. The integration of YOLO-based deep learning object detection methods with GAN net- work yields an average precision of 92.5% in low-light conditions, 88.7% in foggy_x000D_ Object Detection for Self-Driving Cars in Foggy Conditions_x000D_ 19_x000D_ conditions, and 86.2% in rainy conditions. This demonstrates a respective improvement of 12.5%, 8.7%, and 6.2% over the baseline. The results of our research indicate that the combination of several method- ologies can enhance object identification for self-driving automobiles. This has the potential to greatly enhance the safety and dependability of these vehicles in the future. The suggested methodology can be expanded to encompass other use cases that necessitate precise and resilient object recognition in challenging circumstances, such as surveillance systems, robots, and_x000D_ security systems.en_US
dc.publisherCyber Tech Publicationsen_US
dc.subjectKeywords: Machine Learning, Satellite Images, Object Detection, Self- Driving cars, Foggy Conditionsen_US
dc.titleObject Detection for Self-Driving Cars in Foggy Conditionsen_US
dc.typeBook Chapteren_US
Appears in Collections:Book Chapters_ SCSET

Files in This Item:
File SizeFormat 
c2.pdf
  Restricted Access
2.02 MBAdobe PDFView/Open Request a copy

Contact admin for Full-Text

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.