Automatic CIN Grades Prediction of Sequential Cervigram Image Using LSTM With Multistate CNN Features.
Zijie YueShuai DingWeidong ZhaoHao WangJie MaYoutao ZhangYanchun ZhangPublished in: IEEE J. Biomed. Health Informatics (2020)
Keyphrases
- image features
- low level
- extracting features
- input image
- image analysis
- image description
- image data
- prediction accuracy
- image retrieval
- extracted features
- image representation
- single image
- textural features
- image regions
- test images
- image content
- global features
- image classification
- high resolution
- multiscale
- original images
- grey level
- salient features
- feature matching
- geometric constraints
- image pixels
- spatial distribution
- keypoints
- feature vectors
- feature extraction
- individual features
- image set
- feature values
- feature set
- image segmentation
- invariant features
- image similarity
- feature detection
- matching process
- feature descriptors
- spatial relationships
- spatial information
- segmentation method
- scale space
- edge detection
- similarity measure
- sample images
- convolutional neural network
- cellular neural networks
- color histogram
- target object
- feature subset
- region of interest
- feature points
- feature space