We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
基于多尺度超像素融合网络的脑CT图像分类方法.
- Authors
冀俊忠; 张梦隆; 宋 晓; 张晓丹
- Abstract
In order to identify brain CT image features with variable lesion morphology and location, a novel multi-scale superpixel fusion network (MSFN) was proposed to improve the image classification method. MSFN was able to extract more expressive classification features through image fusion and feature fusion. Firstly, the original brain CT image was enhanced by multi-scale superpixel to obtain the optimized fusion image. Then, the high-level features of the fusion image were combined with the multiscale superpixel low-level features to obtain more discriminative features for the classification of brain CT images. Experimental results validate the effectiveness of the proposed method.
- Subjects
COMPUTED tomography; BRAIN imaging; CLASSIFICATION; MORPHOLOGY
- Publication
China Sciencepaper, 2022, Vol 17, Issue 11, p1173
- ISSN
2095-2783
- Publication type
Article