Autonomous Vehicle Detection and Classification using Feature Mapping in Generative Adversarial Network
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Autonomous vehicle (AV) navigation is facilitated using LiDAR and imaging sensors for detecting objects and identifying vehicles.In particular, computer vision (CV) and automated processing are required to maximize navigation through image processing.The problem of vehicle detection and classification is influenced by the extracted feature mapping due to uneven image sizes.This article introduces a feature-mapping-based vehicle boundary detection method for addressing this problem.The proposed method uses a generative adversarial network for mapping extracted features based on contrast.Using multiple mapping instances, the low intensity and high contrast features are used to detect vehicle boundaries.Based on the detected boundaries, the adversarial network mapping rate is trained to improve the detection precision.The proposed method improves accuracy by 9.42%, precision by 10.38%, and reduces the mean square error by 9.28% for the maximum boundaries identified.
Fields of Study & Subjects
Advanced Neural Network Applications
Autonomous Vehicle Technology and Safety
Domain Adaptation and Few-Shot Learning
Publication details
- Journal / source
- Advances in intelligent systems research/Advances in Intelligent Systems Research
- Pages
- 307–318
- Publication date
- 1 April 2026
- Language
- EN
- Document type
- Book chapter