: Explores both geometric and appearance-based approaches for multi-class and intra-class vehicle classification.
: Introduces a classification scheme for surveillance images using deep learning and data augmentation to handle varying camera resolutions. Feature-Based Approaches :
: Provides a historical account and technical review of how vehicle detection and classification have evolved from basic computer vision to modern high-accuracy neural networks. AI responses may include mistakes. Learn more
Several research papers focus on the classification and recognition of using various computational methods, primarily for intelligent traffic management and autonomous driving. Key research papers and their focus areas include: Deep Learning and Computer Vision :
: Proposes a method using YOLO and ResNet-50 to detect and classify vehicles into four size categories and eight color categories with high accuracy.
: Uses Principal Components Analysis (PCA) to extract features from vehicle fronts for classification, specifically handling day and night conditions separately. Comprehensive Reviews :
: Discusses a model specialized in recognizing cars, SUVs, and vans by combining multi-layer features to improve precision in complex traffic scenarios.