An object-oriented method based on multi-scale segmentation for classification and mapping from Quick bird images.
Limei ZhouWenbo XuZhaoxian WangWenzhi ZhangJing WangXiaoyu ZhangJinlong FanPublished in: IGARSS (2013)
Keyphrases
- segmentation method
- edge detection
- segmentation algorithm
- fully automatic
- multiscale
- test images
- segmentation errors
- classification method
- adaptive thresholding
- classification accuracy
- thresholding method
- grey level
- region of interest
- similarity measure
- image regions
- feature extraction and selection
- segmentation scheme
- feature extraction
- image segmentation
- image classification
- support vector machine
- pixel classification
- pixel level
- feature set
- image features
- piece wise
- energy function
- feature vectors
- gray level images
- segmentation accuracy
- input image
- shape prior
- region growing
- medical images
- image slices
- image segments
- pigmented skin lesions
- cardiac magnetic resonance
- region segmentation
- gray value
- multiple scales
- segmented images
- image set
- support vector machine svm