Generic placeholder image

Recent Advances in Computer Science and Communications

Author(s): Huanjun Zhang*

DOI: 10.2174/2666255816666230330153428

DownloadDownload PDF Flyer Cite As
Image Generation Method Based on Improved Generative Adversarial Network

Article ID: e300323215251 Pages: 8

  • * (Excluding Mailing and Handling)

Abstract

Background: The image generation model based on generative adversarial network (GAN) has achieved remarkable achievements. However, traditional GAN has the disadvantage of unstable training, which affects the quality of the generated image.

Objective: This method is to solve the GAN image generation problems of poor image quality, single image category, and slow model convergence.

Methods: An improved image generation method is proposed based on GAN. Firstly, the attention mechanism is introduced into the convolution layer of the generator and discriminator and a batch normalization layer is added after each convolution layer. Secondly, the ReLU and leaky ReLU are used as the active layer of the generator and discriminator, respectively. Thirdly, the transposed convolution is used in the generator while the small step convolution is used in the discriminator, respectively. Fourthly, a new discarding method is applied in the dropout layer.

Results: The experiments were carried out on Caltech 101 dataset. The experimental results showed that the image quality generated by the proposed method is better than that generated by GAN with attention mechanism (AM-GAN) and GAN with stable training strategy (STS-GAN) and the stability was improved.

Conclusion: The proposed method is effectiveness for image generation with high quality.

Graphical Abstract

[1]
M. Kiss, H. Zhang, M.K. Fix II, P. Manser, and L. Xing, "Z-Super resolution CT-image generation with a deep 3D fully convolutional neural network", Int. J. Radiat. Oncol. Biol. Phys., vol. 108, no. 3, p. e784, 2020.
[http://dx.doi.org/10.1016/j.ijrobp.2020.07.249]
[2]
K. Gregor, I. Danihelka, and A. Graves, "DRAW: a recurrent neural network for image generation", Proceedings of the 32nd International Conference on Machine Learning, Lille, France, 2015, pp. 1462-1471.
[3]
J. Chen, "A neuron-MOS-based VLSI implementation of pulse-coupled neural networks for image feature generation", IEEE Trans. Circ. Syst., vol. 57, no. 6, pp. 1143-1153, 2010.
[4]
H.A. Qadir, I. Balasingham, and Y. Shin, "Simple U-net based synthetic polyp image generation: Polyp to negative and negative to polyp", Biomed. Signal Process. Control, vol. 74, pp. 103491-103493, 2022.
[http://dx.doi.org/10.1016/j.bspc.2022.103491]
[5]
I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, "Generative adversarial networks", Commun. ACM, vol. 63, no. 11, pp. 139-144, 2020.
[http://dx.doi.org/10.1145/3422622]
[6]
A. Frühstück, K.K. Singh, and E. Shechtman, "InsetGAN for full-body image generation", Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA, 2022, pp. 7723-7732.
[7]
K. Katsumata, D.M. Vo, and H. Nakayama, "OSSGAN: Open-set semi-supervised image generation", Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, LA, USA, 2022, pp. 11185-11193.
[8]
H. Tan, X. Liu, B. Yin, and X. Li, "DR-GAN: distribution regularization for text-to-Image generation", IEEE Trans. Neural Netw. Learn. Syst., vol. 33, no. 4, pp. 1-15, 2022.
[PMID: 35442894]
[9]
C. Han, H. Hayashi, and L. Rundo, "GAN-based synthetic brain MR image generation", 15th International Symposium on Biomedical Imaging, Washington, DC, USA, 2018, pp. 734-738.
[10]
T. Mahendran, and S. Sharmilan, GAN based photo-realistic image generation from sketch using nested u-net., Advanced Computing and Communication Technologies for High Performance Applications: Cochin, India, 2020, pp. 274-280.
[http://dx.doi.org/10.1109/ACCTHPA49271.2020.9213230]
[11]
R. Huang, S. Zhang, and T. Li, "Beyond face rotation: global and local perception GAN for photorealistic and identity preserving frontal view synthesis", IEEE International Conference on Computer Vision, Venice, Italy, 2017, pp. 2439-2448.
[http://dx.doi.org/10.1109/ICCV.2017.267]
[12]
J. Li, X. Liang, and Y. Wei, "Perceptual generative adversarial networks for small object detection", IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA, 2017, pp. 1951-1959.
[http://dx.doi.org/10.1109/CVPR.2017.211]
[13]
W. Liang, D. Ding, and G. Wei, "An improved DualGAN for near-infrared image colorization", Infrared Phys. Technol., vol. 116, p. 103764, 2021.
[http://dx.doi.org/10.1016/j.infrared.2021.103764]
[14]
Q. Zhou, M. Tan, and H. Xi, "ACGANs-CNN: A novel intrusion detection method", J. Phys. Conf. Ser., vol. 1757, no. 1, p. 012012, 2021.
[http://dx.doi.org/10.1088/1742-6596/1757/1/012012]
[15]
T. Kinnunen, J.K. Kamarainen, and L. Lensu, "Making visual object categorization more challenging: Randomized Caltech-101 Data Set", International Conference on Pattern Recognition, Istanbul, Turkey, 2010, pp. 476-479.
[http://dx.doi.org/10.1109/ICPR.2010.124]
[16]
T. Salimans, I. Goodfellow, and W. Zaremba, "Improved techniques for training GANs", Proceedings of the 30th International Conference on Neural Information Processing Systems, Barcelona, Spain, 2016, pp. 2234-2242.
[17]
M. Heusel, H. Ramsauer, and T. Unterthiner, "GANs trained by a two time-scale update rule converge to a local nash equilibrium", Adv. Neural Inf. Process. Syst., vol. 30, pp. 1-5, 2017.