基于生成对抗式神经网络的红外目标仿真方法
|
摘要
提出了一种应用于红外目标仿真的模型。利用训练后的条件深度卷积生成对抗网络,只需输入随机噪声和类别标签,便能够自动产生预期类别的红外目标仿真图像。在手写数字数据集(MNIST)和红外数据集上分别训练模型参数,再进行自动生成实验,均可以产生高真实度的样本图像;将判别网络提取的特征用于分类实验,并将所提方法合成的图像用于数据增强,以提升分类器性能。研究结果表明,所提方法能够有效模仿红外辐射特征。
A model applied to the simulation of infrared targets is proposed. By the trained conditional deep convolutional generative adversarial networks, only the random noise and category label are necessary for the automatic generation of the simulation images of infrared targets belonging to the expected category. The parameters are trained on the handwritten digital dataset(MNIST) and the infrared dataset, respectively, and subsequently the automatic generation experiment is carried out, which can produce the high trueness sample images. The features extracted by the discrimination network are used in the classification experiments, and the images synthesized by the proposed method are used for data augmentation to improve the classifier performance. The research results show that the proposed method can effectively imitate the infrared radiation characteristics.
A model applied to the simulation of infrared targets is proposed. By the trained conditional deep convolutional generative adversarial networks, only the random noise and category label are necessary for the automatic generation of the simulation images of infrared targets belonging to the expected category. The parameters are trained on the handwritten digital dataset(MNIST) and the infrared dataset, respectively, and subsequently the automatic generation experiment is carried out, which can produce the high trueness sample images. The features extracted by the discrimination network are used in the classification experiments, and the images synthesized by the proposed method are used for data augmentation to improve the classifier performance. The research results show that the proposed method can effectively imitate the infrared radiation characteristics.
引文
[1] Huang X, Zhang J Q, Zhang S Z, et al. GPU-based high-precision real-time radiometric rendering for IR scene generation[J]. Infrared Physics & Technology, 2014, 65: 134-143.
[2] Sun W, Wang B. Calculation and image simulation of aircraft infrared radiation[J]. Laser & Infrared, 2017, 47(6): 728-735. 孙卫, 王彪. 飞机红外辐射计算及图像仿真[J]. 激光与红外, 2017, 47(6): 728-735.
[3] Shao L, Bai T Z, Zheng H J, et al. Imaging analysis of infrared thermal imaging system based on sea-sky model[J]. Laser & Optoelectronics Progress, 2017, 54(11): 111104. 邵龙, 白廷柱, 郑海晶, 等. 基于海空模型的红外热成像系统成像影响分析[J]. 激光与光电子学进展, 2017, 54(11): 111104.
[4] Zhu Q R. Design of dynamic infrared scene simulation system based on augmented reality[D]. Beijing: University of Chinese Academy of Sciences (Shanghai Institute of Technical Physics,Chinese Academy of Sciences), 2017. 祝清瑞. 基于增强现实的动态红外场景仿真系统的设计研究[D]. 北京: 中国科学院大学(中国科学院上海技术物理研究所), 2017.
[5] Yu S Q, Han Z, Tang Y D, et al. Texture synthesis method based on generative adversarial networks[J]. Infrared and Laser Engineering, 2018, 47(2): 34-39. 余思泉, 韩志, 唐延东, 等. 基于对抗生成网络的纹理合成方法[J]. 红外与激光工程, 2018, 47(2): 34-39.
[6] Yousef A M, Omar Y M K, Fakharany E. Deep generative image model using a hybrid system of generative adversarial nets (GANs)[C]. International Conference on Advanced Control Circuits Systems, 2017: 278-285.
[7] Liu X H, Xie C J, Kuang H L, et al. Face aging simulation with deep convolutional generative adversarial networks[C]. International Conference on Measuring Technology and Mechatronics Automation, 2018: 220-224.
[8] Suárez P L, Sappa A D, Vintimilla B X. Infrared image colorization based on a triplet DCGAN architecture[C]. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017: 212-217.
[9] Zhao D C, Weng J K, Liu Y H. Generating traffic scene with deep convolutional generative adversarial networks[C]. Chinese Automation Congress, 2017: 6612-6617.
[10] Zhang H, Patel V M, Riggan B S, et al. Generative adversarial network-based synthesis of visible faces from polarimetrie thermal faces[C]. IEEE International Joint Conference on Biometrics, 2017: 100-107.
[11] Enomoto K, Sakurada K, Wang W M, et al. Filmy cloud removal on satellite imagery with multispectral conditional generative adversarial nets[C]. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017: 1533-1541.
[12] Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[C]. Advances in Neural Information Processing Systems, 2014: 2672-2680.
[13] Mirza M, Osindero S. Conditional generative adversarial nets[EB/OL]. (2014-11-06)[2018-07-28]. https://arxiv.org/abs/1411.1784.
[14] Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[EB/OL]. (2016-01-07)[2018-07-28]. https://arxiv.org/abs/1511.06434
[15] Tang X L, Du Y M, Liu Y W, et al. Image recognition with conditional deep convolutional generative adversarial networks[J]. Acta Automatica Sinica, 2018, 44(5): 855-864. 唐贤伦, 杜一铭, 刘雨微, 等. 基于条件深度卷积生成对抗网络的图像识别方法[J]. 自动化学报, 2018, 44(5): 855-864.
[16] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2015-04-10)[2018-10-16]. https://arxiv.org/abs/1409.1556
[17] Liu X, Liu B, Jin W Q, et al. A new infrared image detail enhancement and dynamic range compression algorithm[J]. Acta Optica Sinica, 2011, 31(s1): s100504. 刘秀, 刘斌, 金伟其, 等. 一种红外图像细节增强和动态范围压缩处理算法[J]. 光学学报, 2011, 31(s1): s100504.
[18] Liu D W, Han L, Han X Y. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4): 0428001. 刘大伟, 韩玲, 韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报, 2016, 36(4): 0428001.
[19] LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[20] Frid-Adar M, Diamant I, Klang E, et al. GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification[EB/OL]. (2018-03-03)[2018-10-16]. https://arxiv.org/abs/1803.01229.
[2] Sun W, Wang B. Calculation and image simulation of aircraft infrared radiation[J]. Laser & Infrared, 2017, 47(6): 728-735. 孙卫, 王彪. 飞机红外辐射计算及图像仿真[J]. 激光与红外, 2017, 47(6): 728-735.
[3] Shao L, Bai T Z, Zheng H J, et al. Imaging analysis of infrared thermal imaging system based on sea-sky model[J]. Laser & Optoelectronics Progress, 2017, 54(11): 111104. 邵龙, 白廷柱, 郑海晶, 等. 基于海空模型的红外热成像系统成像影响分析[J]. 激光与光电子学进展, 2017, 54(11): 111104.
[4] Zhu Q R. Design of dynamic infrared scene simulation system based on augmented reality[D]. Beijing: University of Chinese Academy of Sciences (Shanghai Institute of Technical Physics,Chinese Academy of Sciences), 2017. 祝清瑞. 基于增强现实的动态红外场景仿真系统的设计研究[D]. 北京: 中国科学院大学(中国科学院上海技术物理研究所), 2017.
[5] Yu S Q, Han Z, Tang Y D, et al. Texture synthesis method based on generative adversarial networks[J]. Infrared and Laser Engineering, 2018, 47(2): 34-39. 余思泉, 韩志, 唐延东, 等. 基于对抗生成网络的纹理合成方法[J]. 红外与激光工程, 2018, 47(2): 34-39.
[6] Yousef A M, Omar Y M K, Fakharany E. Deep generative image model using a hybrid system of generative adversarial nets (GANs)[C]. International Conference on Advanced Control Circuits Systems, 2017: 278-285.
[7] Liu X H, Xie C J, Kuang H L, et al. Face aging simulation with deep convolutional generative adversarial networks[C]. International Conference on Measuring Technology and Mechatronics Automation, 2018: 220-224.
[8] Suárez P L, Sappa A D, Vintimilla B X. Infrared image colorization based on a triplet DCGAN architecture[C]. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017: 212-217.
[9] Zhao D C, Weng J K, Liu Y H. Generating traffic scene with deep convolutional generative adversarial networks[C]. Chinese Automation Congress, 2017: 6612-6617.
[10] Zhang H, Patel V M, Riggan B S, et al. Generative adversarial network-based synthesis of visible faces from polarimetrie thermal faces[C]. IEEE International Joint Conference on Biometrics, 2017: 100-107.
[11] Enomoto K, Sakurada K, Wang W M, et al. Filmy cloud removal on satellite imagery with multispectral conditional generative adversarial nets[C]. IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2017: 1533-1541.
[12] Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks[C]. Advances in Neural Information Processing Systems, 2014: 2672-2680.
[13] Mirza M, Osindero S. Conditional generative adversarial nets[EB/OL]. (2014-11-06)[2018-07-28]. https://arxiv.org/abs/1411.1784.
[14] Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks[EB/OL]. (2016-01-07)[2018-07-28]. https://arxiv.org/abs/1511.06434
[15] Tang X L, Du Y M, Liu Y W, et al. Image recognition with conditional deep convolutional generative adversarial networks[J]. Acta Automatica Sinica, 2018, 44(5): 855-864. 唐贤伦, 杜一铭, 刘雨微, 等. 基于条件深度卷积生成对抗网络的图像识别方法[J]. 自动化学报, 2018, 44(5): 855-864.
[16] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[EB/OL]. (2015-04-10)[2018-10-16]. https://arxiv.org/abs/1409.1556
[17] Liu X, Liu B, Jin W Q, et al. A new infrared image detail enhancement and dynamic range compression algorithm[J]. Acta Optica Sinica, 2011, 31(s1): s100504. 刘秀, 刘斌, 金伟其, 等. 一种红外图像细节增强和动态范围压缩处理算法[J]. 光学学报, 2011, 31(s1): s100504.
[18] Liu D W, Han L, Han X Y. High spatial resolution remote sensing image classification based on deep learning[J]. Acta Optica Sinica, 2016, 36(4): 0428001. 刘大伟, 韩玲, 韩晓勇. 基于深度学习的高分辨率遥感影像分类研究[J]. 光学学报, 2016, 36(4): 0428001.
[19] LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.
[20] Frid-Adar M, Diamant I, Klang E, et al. GAN-based synthetic medical image augmentation for increased CNN performance in liver lesion classification[EB/OL]. (2018-03-03)[2018-10-16]. https://arxiv.org/abs/1803.01229.