Batch Similarity Based Triplet Loss Assembled into Light-Weighted Convolutional Neural Networks for Medical Image Classification.
Zhiwen HuangQuan ZhouXingxing ZhuXuming ZhangPublished in: Sensors (2021)
Keyphrases
- convolutional neural networks
- image classification
- image representation
- convolutional network
- image features
- feature extraction
- bag of words
- medical information
- medical imaging
- medical diagnosis
- medical data
- medical domain
- sparse representation
- weighted sum
- dynamic programming
- remotely sensed data
- object detection
- class specific
- medical experts
- batch mode
- total weight
- object recognition
- multiscale