Abstract | ||
---|---|---|
•Our DCIA framework utilizes all DenseNet-generated features to explore the “optimal” representation for modeling the instance segmentation.•Our DCIA framework employs FL to obtain reasonable contour segmentation results.•Our DCIA framework uses both morphological methods and ConvCRFs to refine the confidence map to obtain more accurate segmentation results. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1016/j.bspc.2020.101988 | Biomedical Signal Processing and Control |
Keywords | DocType | Volume |
Colon gland instance segmentation,Dense convolutional neural network,Focal loss,Convolutional conditional random fields | Journal | 60 |
ISSN | Citations | PageRank |
1746-8094 | 0 | 0.34 |
References | Authors | |
0 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Liye Mei | 1 | 0 | 0.34 |
Xiaopeng Guo | 2 | 13 | 1.80 |
Xiangsheng Huang | 3 | 124 | 21.78 |
Yueyun Weng | 4 | 0 | 0.34 |
Sheng Liu | 5 | 0 | 0.34 |
Cheng Lei | 6 | 0 | 0.68 |