Human labeling errors and their impact on ConvNets for satellite image scene classification.
Longkang PengTao WeiXuehong ChenXiaobei ChenRui SunLuoma WanXiaolin ZhuPublished in: CoRR (2023)
Keyphrases
- scene classification
- satellite images
- feed forward
- object recognition
- biologically inspired
- image classification
- natural scenes
- remote sensing
- remotely sensed imagery
- multispectral
- change detection
- image representation
- indoor outdoor
- digital elevation models
- visual words
- bag of features
- image segmentation
- unsupervised learning
- higher order
- artificial neural networks
- neural network
- multi instance multi label learning
- bag of words
- back propagation
- natural images
- co occurrence
- high resolution