基于VAE-CGAN的牦牛等级评定算法
作者:
基金项目:

青海省科技计划(2020-QY-218); 国家现代农业产业技术体系(CARS-37)


Grade Evaluation Algorithm of Yak Based on VAE-CGAN
Author:
  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献 [23]
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    在牦牛高效养殖过程中, 牦牛等级评定是牦牛育种工作中的重要环节. 为了在牦牛等级评定研究中, 降低数据集分布不平衡对牦牛等级预测结果的影响, 提出一种基于改进条件生成对抗网络模型的牦牛等级评定模型VAE-CGAN. 首先, 为获取高质量生成样本, 模型通过引入变分自编码器取代条件生成对抗网络输入中的随机噪声, 降低了随机变量带来的不确定性. 此外, 模型将牦牛标签作为条件信息输入到生成对抗模型中来获取指定类别的生成样本, 生成样本及训练样本则会被用于训练深度神经网络分类器. 实验结果显示, 模型整体预测准确率达到了97.9%. 而且与生成对抗网络相比较, 在数量较少的特级牦牛等级预测上的精准率、召回率和F1值分别提升了16.7%、16.6%和19.4%. 实验结果表明该模型可以实现高精准度和低误分类率的牦牛等级分类.

    Abstract:

    Yak grade evaluation is an important part of high-efficiency yak breeding. To reduce the influence of imbalanced data set distribution on the prediction results of yak grading in the research, this study proposes a yak grade evaluation model based on an improved conditional generative adversarial network model, called VAE-CGAN. Firstly, to obtain high-quality generated samples, the model reduces the uncertainty from random variables by introducing a variational autoencoder to replace the random noise in the input of the conditional generative adversarial network. In addition, the model inputs the yak label as conditional information into the generative adversarial model to obtain the generated samples of the specified category, and the generated samples and training samples are utilized to train the deep neural network classifier. The experimental results show that the overall prediction accuracy of the model has reached 97.9%. The Precision, Recall, and F1 value on the grade prediction of premium yak have increased by 16.7%, 16.6%, and 19.4% respectively compared with those of the generative adversarial network. The results indicate the model can achieve yak classification with high accuracy and low misclassification rate.

    参考文献
    [1] 袁凯鑫, 王昕. 青海高原大通牦牛的育种进展. 中国牛业科学, 2019, 45(1):28-32.[doi:10.3969/j.issn.1001-9111.2019.01.008
    [2] 张江. 牦牛高效养殖关键技术. 畜牧兽医科技信息, 2020, (4):99.[doi:10.3969/J.ISSN.1671-6027.2020.04.089
    [3] 骆正杰, 马进寿, 保广才, 等. 青海省牦牛种业发展现状、存在问题及应对策略. 中国畜牧杂志, 2021, 57(2):231-234.[doi:10.19556/j.0258-7033.20200429-02
    [4] Yan Q, Ding LM, Wei HY, et al. Body weight estimation of yaks using body measurements from image analysis. Measurement, 2019, 140:76-80.[doi:10.1016/j.measurement.2019.03.021
    [5] 陈争涛, 黄灿, 杨波, 等. 基于迁移学习的并行卷积神经网络牦牛脸识别算法. 计算机应用, 2021, 41(5):1332-1336
    [6] Wakholi C, Kim J, Nabwire S, et al. Deep learning feature extraction for image-based beef carcass yield estimation. Biosystems Engineering, 2022, 218:78-93.[doi:10.1016/j.biosystemseng.2022.04.008
    [7] Douzas G, Bacao F, Last F. Improving imbalanced learning through a heuristic oversampling method based on K-means and SMOTE. Information Sciences, 2018, 465:1-20.[doi:10.1016/j.ins.2018.06.056
    [8] Bennin KE, Keung J, Phannachitta P, et al. MAHAKIL:Diversity based oversampling approach to alleviate the class imbalance issue in software defect prediction. IEEE Transactions on Software Engineering, 2018, 44(6):534-550.[doi:10.1109/TSE.2017.2731766
    [9] Laurikkala J. Improving identification of difficult small classes by balancing class distribution. Proceedings of the 8th Conference on Artificial Intelligence in Medicine in Europe. Cascais:Springer, 2001. 63-66.
    [10] Roy NKS, Rossi B. Cost-sensitive strategies for data imbalance in bug severity classification:Experimental results. Proceedings of the 2017 43rd Euromicro Conference on Software Engineering and Advanced Applications (SEAA). Vienna:IEEE, 2017. 426-429.
    [11] 叶德豪, 王琼. 面向不平衡数据的生成对抗网络研究. 工业控制计算机, 2021, 34(5):95-96.[doi:10.3969/j.issn.1001-182X.2021.05.039
    [12] Gao R, Hou XS, Qin J, et al. Zero-VAE-GAN:Generating unseen features for generalized and transductive zero-shot learning. IEEE Transactions on Image Processing, 2020, 29:3665-3680.[doi:10.1109/TIP.2020.2964429
    [13] Dideriksen BU, Derosche K, Tan ZH. iVAE-GAN:Identifiable VAE-GAN models for latent representation learning. IEEE Access, 2022, 10:48405-48418.[doi:10.1109/ACCESS.2022.3172333
    [14] Haque A. EC-GAN:Low-sample classification using semi-supervised algorithms and GANs. Proceedings of the 35th AAAI Conference on Artificial Intelligence. Online:AAAI, 2021. 15797-15798.
    [15] Kingma DP, Welling M. Auto-encoding variational Bayes. arXiv:1312.6114, 2013.
    [16] Mirza M, Osindero S. Conditional generative adversarial nets. arXiv:1411.1784, 2014.
    [17] 邹秀芳, 朱定局. 生成对抗网络研究综述. 计算机系统应用, 2019, 28(11):1-9.[doi:10.15888/j.cnki.csa.007156
    [18] 于龙泽, 肖白, 孙立国. 风光出力场景生成的条件深度卷积生成对抗网络方法. 东北电力大学学报, 2021, 41(6):90-99.[doi:10.19718/j.issn.1005-2992.2021-06-0090-10
    [19] 李梦磊, 刘新, 赵梦凡, 等. 基于语句结构信息的方面级情感分类. 计算机系统应用, 2020, 29(11):114-120.[doi:10.15888/j.cnki.csa.007681
    [20] 刘丽元, 臧长江, 周靖航, 等. 新疆昌吉地区荷斯坦奶牛生长发育规律分析. 中国畜牧兽医, 2015, 42(8):2036-2041.[doi:10.16431/j.cnki.1671-7236.2015.08.017
    [21] 陈巧红, 王磊, 孙麒, 等. 卷积神经网络的短文本分类方法. 计算机系统应用, 2019, 28(5):137-142.[doi:10.15888/j.cnki.csa.006887
    [22] Shah K, Patel H, Sanghvi D, et al. A comparative analysis of logistic regression, random forest and KNN models for the text classification. Augmented Human Research, 2020, 5(1):12.[doi:10.1007/s41133-020-00032-0
    [23] 尚晖. 基于改进SVM的互联网用户分类. 计算机系统应用, 2021, 30(4):266-270.[doi:10.15888/j.cnki.csa.007914
    相似文献
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

李丹,张玉安,何杰,陈占琦,宋维芳,宋仁德.基于VAE-CGAN的牦牛等级评定算法.计算机系统应用,2023,32(1):249-256

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2022-05-20
  • 最后修改日期:2022-07-01
  • 在线发布日期: 2022-09-14
文章二维码
您是第11369897位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号