基于深度卷积生成对抗网络的图像识别算法

刘恋秋

doi:10.3788/YJYXS20203504.0383

您当前的位置：

首页 >

文章列表页 >

基于深度卷积生成对抗网络的图像识别算法

图像处理 | 更新时间：2020-07-05

- 基于深度卷积生成对抗网络的图像识别算法
- Image recognition algorithms based on deep convolution generative adversarial network
- 液晶与显示 2020年35卷第4期页码：383-388
- 作者机构：
  
  重庆财经职业学院, 重庆 402160
- 作者简介：
  
  [ "刘恋秋(1982-), 男, 重庆市人, 硕士, 讲师, 2007年于重庆邮电大学计算机学院获硕士学位, 主要从事计算机仿真方面的研究。E-mail:liulianqiu0627@hotmail.com" ]
- 基金信息：
- DOI：10.3788/YJYXS20203504.0383
  中图分类号： TP391
- 收稿日期：2019-07-03，
  
  修回日期：2019-10-21，
  
  录用日期：2019-10-21，
  
  纸质出版日期：2020-04-05
- 稿件说明：
移动端阅览
刘恋秋. 基于深度卷积生成对抗网络的图像识别算法[J]. 液晶与显示, 2020,35(4):383-388.

Lian-qiu LIU. Image recognition algorithms based on deep convolution generative adversarial network[J]. Chinese journal of liquid crystals and displays, 2020, 35(4): 383-388.
刘恋秋. 基于深度卷积生成对抗网络的图像识别算法[J]. 液晶与显示, 2020,35(4):383-388. DOI： 10.3788/YJYXS20203504.0383.

Lian-qiu LIU. Image recognition algorithms based on deep convolution generative adversarial network[J]. Chinese journal of liquid crystals and displays, 2020, 35(4): 383-388. DOI： 10.3788/YJYXS20203504.0383.

摘要

针对传统深度卷积生成网络收敛速度慢、稳定性较差的问题，本文在传统深度卷积生成对抗网络的基础上，提出了深度卷积生成对抗网络的优化算法。首先在预处理部分，融合了Canny算子和Prewitt算子的多个方向的卷积核来初始化输入图片参数，同时训练模块。为了减少训练时间，将训练分为3个阶段，每个阶段都采用不同的损失函数，从而提升网络的收敛速度及识别效果。最后再将训练后的判别网络中的卷积神经网络用来提取图像特征。LFW和CIFAR-100的实验证明，本文提出的算法具有很高的可行性和有效性，比传统生成对抗网络、CNN等图像识别具有更高的识别成功率，达到89.5%，为生成对抗网络在计算机视觉领域的应用提供了有益的参考。

Abstract

In view of the slow convergence and poor stability of traditional deep convolutional generation networks

this paper combines the Connaught kernels of multiple directions of Canny operator and Prewitt operator on the basis of traditional deep convolution generation confrontation network. To initialize the input picture parameters and reduce the training time

the training is divided into three stages

each stage adopts a different loss function

thereby improving the convergence speed and recognition effect of the network and then discriminating the volume in the network after training. The neural network is used to extract image features. The experiments of LFW and CIFAR-100 prove that the proposed algorithm is highly feasible and effective

and provides a useful reference for generating anti-network applications in the field of image recognition.

关键词

Keywords

references

I J GOODFELLOW , J POUGET-ABADIE , M MIRZA , 等 . Generative adversarial networks . Advances in Neural Information Processing Systems , 2014 . 3 2672 - 2680 .

A RADFORD , L METZ , S CHINTALA . Unsupervised representation learning with deep convolutional generative adversarial networks . Computer Science , 2015 . 12 ( 36 ): 1280 - 1287 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=aab7df43d42b6a78906b2e140bef7751 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=aab7df43d42b6a78906b2e140bef7751 .

T SALIMANS , I GOODFELLOW , W ZAREMBA , 等 . Improved techniques for training GANs . ArXiv , 2016 . 606 03498v1 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=Arxiv000001327398 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=Arxiv000001327398 .

丁鹏 , 张叶 , 刘让 , 等 . 结合形态学和Canny算法的红外弱小目标检测 . 液晶与显示 , 2016 . 31 ( 8 ): 793 - 800 . http://yjyxs.com/CN/abstract/abstract10063.shtml http://yjyxs.com/CN/abstract/abstract10063.shtml .

P DING , Y ZHANG , R LIU , 等 . Infrared weak target detection based on morphology and Canny algorithm . Chinese Journal of Liquid Crystal and Displays , 2016 . 31 ( 8 ): 793 - 800 . http://yjyxs.com/CN/abstract/abstract10063.shtml http://yjyxs.com/CN/abstract/abstract10063.shtml .

Y KIM . Convolutional neural networks for sentence classification . ArXiv , 2014 . 1408 5882 http://d.old.wanfangdata.com.cn/OAPaper/oai_arXiv.org_1306.3584 http://d.old.wanfangdata.com.cn/OAPaper/oai_arXiv.org_1306.3584 .

金志刚 , 李静昆 . 基于对象性和多层线性模型的协同显著性检测 . 光学精密工程 , 2019 . 27 ( 8 ): 1845 - 1853 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201908020 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201908020 .

Z G JIN , J K LI . Co-saliency detection based onobjectness and multi-layer linear model . Optics and Precision Engineering , 2019 . 27 ( 8 ): 1845 - 1853 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201908020 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201908020 .

潘仙张 , 张石清 , 郭文平 . 多模深度卷积神经网络应用于视频表情识别 . 光学精密工程 , 2019 . 27 ( 4 ): 963 - 970 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201904023 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201904023 .

X Z PAN , S Q ZHANG , W P GUO . Video-based facial expression recognition using multimodal deepconvolutional neural networks . Optics and Precision Engineering , 2019 . 27 ( 4 ): 963 - 970 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201904023 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201904023 .

张坤华 , 谭志恒 , 李斌 . 结合粒子群优化和综合评价的脉冲耦合神经网络图像自动分割 . 光学精密工程 , 2018 . 26 ( 4 ): 962 - 970 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201804026 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201804026 .

K H ZHANG , Z H TAN , B LI . Automated image segmentation based on pulse coupled neural network withpartide swarm optimization and comprehensive evaluation . Optics and Precision Engineering , 2018 . 26 ( 4 ): 962 - 970 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201804026 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201804026 .

吕永标 , 赵建伟 , 曹飞龙 . 基于复合卷积神经网络的图像去噪算法 . 模式识别与人工智能 , 2017 . 30 ( 2 ): 97 - 105 . http://d.old.wanfangdata.com.cn/Periodical/mssbyrgzn201702001 http://d.old.wanfangdata.com.cn/Periodical/mssbyrgzn201702001 .

Y B LYU , J W ZHAO , F L CAO . Image denoising algorithm based on compound convolution neural network . Pattern Recognition and Artificial Intelligence , 2017 . 30 ( 2 ): 97 - 105 . http://d.old.wanfangdata.com.cn/Periodical/mssbyrgzn201702001 http://d.old.wanfangdata.com.cn/Periodical/mssbyrgzn201702001 .

王秀席 , 王茂宁 , 张建伟 , 等 . 基于改进的卷积神经网络lenet-5的车型识别方法 . 计算机应用研究 , 35 ( 7 ): 301 - 304 . http://d.old.wanfangdata.com.cn/Periodical/jsjyyyj201807072 http://d.old.wanfangdata.com.cn/Periodical/jsjyyyj201807072 .

X X WANG , M N WANG , J W ZHANG , 等 . Vehicle recognition method based on improved convolutional neural network lenet-5 . Computer Application Research , 35 ( 7 ): 301 - 304 . http://d.old.wanfangdata.com.cn/Periodical/jsjyyyj201807072 http://d.old.wanfangdata.com.cn/Periodical/jsjyyyj201807072 .

许赟杰 , 徐菲菲 . 基于arcrelu函数的神经网络激活函数优化研究 . 数据采集与处理 , 2019 . 34 ( 3 ): 517 - 529 . http://d.old.wanfangdata.com.cn/Periodical/sjcjycl201903015 http://d.old.wanfangdata.com.cn/Periodical/sjcjycl201903015 .

Y J XU , F F XU . Optimization of activation function in neural network based on arcrelu function . Data Acquisition and Processing , 2019 . 34 ( 3 ): 517 - 529 . http://d.old.wanfangdata.com.cn/Periodical/sjcjycl201903015 http://d.old.wanfangdata.com.cn/Periodical/sjcjycl201903015 .

ZHENG Z, ZHENG L, YANG Y. Unlabeled samples generated by GAN improve the person re-identification baseline in vitro [C]. IEEE International Conference on Computer Vision, USA: Honolulu, IEEE, 2017.

M ABADI , A AGARWAL , P BARHAM , 等 . TensorFlow:large-scale machine learning on heterogeneous distributed systems . ArXiv , 2016 . 1603 04467 http://d.old.wanfangdata.com.cn/Periodical/dlzdhsb201904030 http://d.old.wanfangdata.com.cn/Periodical/dlzdhsb201904030 .

I J GOODFELLOW , D WARDE-FARLEY , M MIRZA , 等 . Maxout networks . Computer Science , 2013 . 1319 - 1327 . http://d.old.wanfangdata.com.cn/Periodical/txxb201707012 http://d.old.wanfangdata.com.cn/Periodical/txxb201707012 .

TAIGMAN Y, MING Y, RANZATO M, et al. DeepFace: closing the gap to human-level performance in face verification[C]. IEEE Conference on Computer Vision & Pattern Recognition, USA: Columbus, Ohio, IEEE, 2014.

浏览量

166

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

RGB深度图像显著性目标检测方法设计

扩散模型与傅里叶滤波器结合的图像去雾方法

基于颜色和光流的多注意力机制微表情识别

NCA-MobileNet：一种轻量化人脸表情识别方法

视觉跟踪技术中孪生网络的研究进展