A Multi-Scale U-Shaped Convolution Auto-Encoder Based on Pyramid Pooling Module for Object Recognition in Synthetic Aperture Radar Images.
Sirui TianYiyu LinWenyun GaoHong ZhangChao WangPublished in: Sensors (2020)
Keyphrases
- synthetic aperture radar images
- multiscale
- object recognition
- image representation
- natural images
- scale space
- image processing
- coarse to fine
- synthetic aperture radar
- spatial pyramid matching
- bit rate
- wavelet transform
- multiple scales
- computer vision
- scene classification
- keypoints
- rate distortion
- feature detection
- local binary pattern
- edge detection
- object detection
- multiresolution
- sar images
- image features
- wavelet coefficients
- image classification
- wavelet domain
- small samples
- input image
- image segmentation
- decision trees
- denoising