DT-CNN: An Energy-Efficient Dilated and Transposed Convolutional Neural Network Processor for Region of Interest Based Image Segmentation.
Dongseok ImDonghyeon HanSungpill ChoiSanghoon KangHoi-Jun YooPublished in: IEEE Trans. Circuits Syst. (2020)
Keyphrases
- convolutional neural network
- region of interest
- image segmentation
- face detection
- medical images
- deformable models
- wireless sensor networks
- energy efficient
- graph cuts
- ct images
- neural network
- markov random field
- active contours
- segmentation algorithm
- ct scans
- blood vessels
- image processing
- multiscale
- computer vision
- reference point
- detection method
- region growing
- palmprint
- sensor networks
- energy consumption
- segmentation method
- segmented images
- region merging
- object recognition