基于生成式对抗网络的区分度模型

Discrimination model based on generative adversarial networks

为了提升图像相似度判别的有效性,通过定义相似度差值(或区分度),设计了基于生成式对抗网络的区分度模型,并与传统的区分度模型进行了对比试验.试验结果表明,本文模型提升了图像相似度的有效性和稳定性.

In order to improve the validity of image similarity discrimination,the region diversity model based on generative confrontation network is designed by defining similarity difference(or division degree).Compared with the traditional discriminant model,the experimental results show that the proposed model improves the effectiveness and stability of image similarity.