Classification of the Type of Harmonic Source Based on Image-Matrix Transformation and Deep Convolutional Neural Network.
Fei MeiHaoyuan ShaPublished in: IEEE Access (2019)
Keyphrases
- convolutional neural network
- image classification
- image data
- image features
- image content
- image feature extraction
- single image
- multiscale
- singular values
- classification accuracy
- image retrieval
- image analysis
- input image
- segmentation method
- edge detection
- test images
- hyperspectral images
- image segmentation
- feature vectors
- support vector machine
- similarity measure
- orthogonal projection
- image representation
- machine learning
- feature points
- segmentation algorithm
- text classification
- face detection
- sparse representation
- pattern recognition
- support vector
- image regions
- image matching
- transformation matrix
- decision trees
- image scrambling
- preprocessing stage
- unsupervised learning
- auto correlation
- lie group
- image transformations
- region of interest
- class labels
- supervised learning
- training set
- feature space
- feature selection