液晶与显示, 2020, 35 (4): 389, 网络出版: 2020-05-30   

基于Inception V3的图像状态分类技术

Image classification technology based on inception V3
作者单位
重庆商务职业学院,重庆 401331
摘要
为了实现对物体状态的分类识别, 本文在GoogLeNet的Inception V3模块基础上进行了优化, 使用Tanh作为激活函数并结合RMSprop, SGD优化器提升了模型的准确率。首先采用三次卷积插值, GAN对图像集进行预处理, 再利用Inception对图像进行训练, 最后结合RMSprop和SGD优化器对模型进行优化。用本文提出的模型在20个烹饪对象的图像上进行实验, 结果表明, 本文优化的Inception V3模型能够以71.5%的准确度对这些图像的状态进行分类, 与对比算法相比, 在分类准确度、训练损失上都有明显提升, 可以满足图像分类的可靠性、稳定性等要求。
Abstract
In order to realize the classification and recognition of object state, the Inception V3 module on the basis of GoogLeNet is optimized Using Tanh as the activation function and RMSprop, the SGD optimizer improves the accuracy of the model. Firstly, using cubic convolution interpolation, GAN preprocesses the image set. Then, the Inception is used to train the image. Finally, the SGD optimizer is combined with RMSprop to optimize the model. Experiments on the images of 20 cooking objects using the model proposed in this paper show that the optimized Inception V3 model can classify the state of these images with 71.5% accuracy. Compared with the comparison algorithm, the classification accuracy and training loss are obviously improved, which can meet the requirements of reliability and stability of image classification.
参考文献

[1] 郭敬明, 何昕, 魏仲慧.基于在线支持向量机的Mean Shift彩色图像跟踪[J].液晶与显示, 2014, 29(1): 120-128.

    GUO J M, HE X, WEI Z H. New mean shift tracking for color image based on online support vector machine [J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(1): 120-128. (in Chinese)

[2] 李彦冬, 郝宗波, 雷航.卷积神经网络研究综述[J].计算机应用, 2016, 36(9): 2508-2515.

    LIY D, HAO Z B, LEI H. Survey of convolutional neural network [J]. Journal of Computer Applications, 2016, 36(9): 2508-2515. (in Chinese)

[3] 黄伟国, 顾超, 朱忠奎.用于目标识别的PCA-SC形状匹配算法[J].光学 精密工程, 2013, 21(8): 2103-2110.

    HUANG W G, GU C, ZHU Z K. PCA-SC shape matching for object recognition [J]. Optics and Precision Engineering, 2013, 21(8): 2103-2110. (in Chinese)

[4] 王飞, 李定主.模式识别中贝叶斯决策理论的研究[J].科技情报开发与经济, 2007, 17(7): 165-166.

    WANGF, LI D Z. Research on Bayesian decision-making theory in pattern recognition [J]. Sci-Tech Information Development & Economy, 2007, 17(7): 165-166. (in Chinese)

[5] 刘成颖, 吴昊, 王立平, 等.基于PSO优化LS-SVM的刀具磨损状态识别[J].清华大学学报: 自然科学版, 2017, 57(9): 975-979.

    LIU C Y, WU H, WANG L P, et al. Tool wear state recognition based on LS-SVM with the PSO algorithm [J]. Journal of Tsinghua University: Science and Technology, 2017, 57(9): 975-979. (in Chinese)

[6] 杨昌其, 谭娟, 仇争平.基于BP神经网络的管制员疲劳状态识别研究[J].航空计算技术, 2018, 48(1): 17-20, 25.

    YANG C Q, TAN J, QIU Z P. Controller's fatigue recognition based on BP neural network [J]. Aeronautical Computing Technique, 2018, 48(1): 17-20, 25. (in Chinese)

[7] 刘建伟, 刘媛, 罗雄麟.深度学习研究进展[J].计算机应用研究, 2014, 31(7): 1921-1930, 1942.

    LIU J W, LIU Y, LUO X L. Research and development on deep learning [J]. Application Research of Computers, 2014, 31(7): 1921-1930, 1942. (in Chinese)

[8] WANG X, DING Y, LIU M Y, et al. Efficient implementation of a cubic-convolution based image scaling engine [J]. Journal of Zhejiang University Science C, 2011, 12(9): 743-753.

[9] 李俭川, 秦国军, 温熙森, 等.神经网络学习算法的过拟合问题及解决方法[J].振动、测试与诊断, 2002, 22(4): 260-264.

    LI J C, QIN G J, WEN X S, et al. Over-fitting in neural network learning algorithms and its solving strategies [J]. Journal of Vibration Measurement & Diagnosis, 2002, 22(4): 260-264. (in Chinese)

[10] RADFORD A, METZ L, CHINTALA S. Unsupervised representation learning with deep convolutional generative adversarial networks [J]. ArXiv, 2015,1511: 06434.

[11] CHEN J C, PATEL V M, CHELLAPPA R. Unconstrained face verification using deep CNN features [C]//Proceedings of 2016 IEEE Winter Conference on Applications of Computer Vision. Lake Placid, NY, USA: IEEE, 2016: 1-9.

[12] KRON G. Tensor analysis of networks [J]. Students’ Quarterly Journal, 1996, 36(143): 171.

[13] FAN E G. Extended Tanh-function method and its applications to nonlinear equations [J]. Physics Letters A, 2000, 277(4-5): 212-218.

[14] 杨江涛, 唐军, 王玉波, 等. 用于应力测量的可调谐光栅[J]. 光学 精密工程, 2018, 26(7): 1596-1603.

    YANG J T, TANG J, WANG Y B, et al. Tunable grating for stress measurement [J]. Optics and Precision Engineering, 2018, 26(7): 1596-1603. (in Chinese)

[15] 郁晓晖, 高志山, 袁群. 宽光谱长工作距弱荧光信号检测显微物镜设计[J]. 光学 精密工程, 2018, 26(7): 1588-1595.

    YU X H, GAO Z S, YUAN Q. Design of board spectrum and long work distance microscope objective for weak fluorescence signal detection [J]. Optics and Precision Engineering, 2018, 26(7): 1588-1595. (in Chinese)

[16] 林旭东, 刘欣悦, 王帅, 等. 桌面97单元自适应光学系统性能测试[J]. 光学 精密工程, 2016, 24(6): 1272-1280.

    LIN X D, LIU X Y, WANG S, et al. Performance testing of a desk-top 97-element adaptive optical system [J]. Optics and Precision Engineering, 2016, 24(6): 1272-1280. (in Chinese)

[17] CHEN A J H, WANG B C W, JENG C J H. A preliminary study of ANN implementing image filters [C]//Proceedings of 2013 International Symposium on Next-generation Electronics. Kaohsiung, China: IEEE, 2013: 181-184.

[18] KESKAR N S, SOCHER R. Improving generalization performance by switching from Adam to SGD [J]. arXiv,2017,712: 07628.

王旖旎. 基于Inception V3的图像状态分类技术[J]. 液晶与显示, 2020, 35(4): 389. WANG-Yi ni. Image classification technology based on inception V3[J]. Chinese Journal of Liquid Crystals and Displays, 2020, 35(4): 389.

本文已被 3 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!