Evaluation of Classification Performance of Pole-Like Objects from MMS Images Using Convolutional Neural Network and Image Super Resolution.
Tomohiro MizoguchiPublished in: Int. J. Autom. Technol. (2018)
Keyphrases
- image super resolution
- convolutional neural network
- image classification
- dictionary learning
- image registration
- image regions
- high resolution images
- low resolution images
- input image
- class specific
- bounding box
- super resolution
- high resolution
- image analysis
- sparse coding
- map estimation
- sparse representation
- face detection
- object recognition
- ground truth
- low resolution
- keypoints
- feature extraction
- training set
- image features
- edge detection
- object segmentation
- partial occlusion
- test images
- maximum a posteriori
- support vector machine
- bag of words
- image reconstruction
- neural network
- feature selection