MetaCNN: A New Hybrid Deep Learning Image-based Approach for Vehicle Classification Using Transformer-like Framework.
Juntian ChenRuikang LuoPublished in: CSSE (2022)
Keyphrases
- deep learning
- image classification
- single image
- image features
- input image
- machine learning
- multiscale
- image representation
- image segmentation
- lighting conditions
- image content
- image regions
- segmentation method
- unsupervised learning
- pattern recognition
- feature extraction
- high resolution
- region of interest
- restricted boltzmann machine
- energy function
- training set
- object recognition
- training data