Training Convolutional Neural Networks with Multi-Size Images and Triplet Loss for Remote Sensing Scene Classification.
Jianming ZhangChaoquan LuJin WangXiao-Guang YueSe-Jung LimZafer Al-MakhadmehAmr TolbaPublished in: Sensors (2020)
Keyphrases
- remote sensing
- satellite images
- image analysis
- scene classification
- remote sensing imagery
- high spatial resolution
- image classification
- object recognition
- multispectral
- hyperspectral images
- change detection
- remote sensing data
- image data
- remote sensing images
- digital image analysis
- multi spectral images
- indoor outdoor
- input image
- hyperspectral
- image processing
- image database
- bag of features
- image fusion
- image retrieval
- image features
- image representation
- convolutional neural networks
- high resolution
- image registration
- machine learning
- image sequences
- pattern recognition
- image matching
- image regions
- land cover
- biologically inspired
- natural scenes
- keypoints
- image segmentation
- lighting conditions