使用CNN特征在NBI内窥镜检查中进行计算机辅助结直肠肿瘤分类
Computer-Aided Colorectal Tumor Classification in NBI Endoscopy Using CNN Features
摘要:
在本文中,我们报告的结直肠NBI内窥镜图像识别的结果,使用卷积神经网络(CNN)提取的特征。在这项比较研究中,我们从不同的CNN模型中提取不同层的特征,然后训练线性SVM分类器。具有10倍交叉验证的实验结果表明,来自前几个卷积层的特征足以实现类似的性能(即,识别率为95%),具有非CNN局部特征,如视觉袋词,Fisher向量和VLAD。
英文摘要:
In this paper we report results for recognizing colorectal NBI endoscopic images by using features extracted from convolutional neural network (CNN). In this comparative study, we extract features from different layers from different CNN models, and then train linear SVM classifiers. Experimental results with 10-fold cross validations show that features from first few convolution layers are enough to achieve similar performance (i.e., recognition rate of 95%) with non-CNN local features such as Bag-of-Visual words, Fisher vector, and VLAD.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。