Journal on Communications ›› 2020, Vol. 41 ›› Issue (3): 33-44.doi: 10.11959/j.issn.1000-436x.2020041

• Papers • Previous Articles     Next Articles

Noise robust chi-square generative adversarial network

Hongjun LI1,2,3,4,Chaobo LI1,Shibing ZHANG1   

  1. 1 School of Information Science and Technology,Nantong University,Nantong 226019,China
    2 State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210093,China
    3 Research Center for Intelligent Information Technology,Nantong University,Nantong 226019,China
    4 TONGKE School of Microelectronics,Nantong University,Nantong 226019,China
  • Revised:2020-01-07 Online:2020-03-25 Published:2020-03-31
  • Supported by:
    The National Natural Science Foundation of China(61871241);Ministry of Education Cooperation in Production and Education(201802302115);Educational Science Research Subject of China Transportation Education Research Association(交教研1802-118);The Science and Technology Program of Nantong(JC2018025);The Science and Technology Program of Nantong(JC2018129);Nantong University-Nantong Joint Research Center for Intelligent Information Technology(KFKT2017B04);Nanjing University State Key Laboratory for Novel Software Technology(KFKT2019B15);Postgraduate Research and Practice Innovation Program of Jiangsu Province(KYCX19_2056)

Abstract:

Aiming at the obvious difference of image quality generated by generative adversarial network under different noises,a chi-square generative adversarial network (CSGAN) was proposed.Combing the advantages of quantification sensitivity and sparse invariance,the chi-square divergence was introduced to calculate the distance between the generated samples and the original samples,which could reduce the influence of different noises on the generated samples and the quality requirement of original samples.Meanwhile,the network architecture was built and the global optimization objective function was constructed to enhance the adversarial performance.Experimental results show that the quality of the images generated by the proposed algorithm has little difference,and the network is more robust to different noises than the state-of-the-art networks.The application of chi-square divergence not only improves the quality of generated images,but also increases the robustness of the network under different noises.

Key words: generative adversarial network, chi-square divergence, noise distribution, image quality

CLC Number: 

No Suggested Reading articles found!