Joint loss optimization based high similarity identification for milch goats

Cheng Shang; Meili Wang; Jifeng Ning; Qunhui Li; Yu Jiang; Xiaolong Wang

doi:10.11834/jig.200619

Image Analysis and Recognition | Views : 0 下载量: 146 CSCD: 2

PDF
Export
Share
Collection
Album

Joint loss optimization based high similarity identification for milch goats
Vol. 27, Issue 4, Pages: 1137-1147(2022)
Received：26 October 2020，

Revised：2021-2-8，

Accepted：15 February 2021，

Published：16 April 2022
DOI： 10.11834/jig.200619
稿件说明：

移动端阅览

DOI：

Cheng Shang, Meili Wang, Jifeng Ning, Qunhui Li, Yu Jiang, Xiaolong Wang. Joint loss optimization based high similarity identification for milch goats[J]. Journal of Image and Graphics, 2022, 27(4): 1137-1147. DOI： 10.11834/jig.200619.

摘要

目的

动物个体身份识别一直是智慧畜牧业的主要难题之一，由于动物个体本身与人类在图像识别上需要的数据特征不同以及各个特征作为个体属性之间的关系不明确，对动物个体识别领域的研究较少，针对具有高相似度的奶山羊个体身份识别问题，提出了基于深度学习的高相似度的奶山羊识别方法。

方法

采集了26只萨能奶山羊的全身图像，利用SSD(single shot MultiBox detection)网络进行数据集预处理，并随机选取1 040幅图像作为训练集，260幅图像作为测试集。其次采用ResNet18(residual neural network)预训练模型并进行迁移学习，最后联合三元组损失函数与交叉熵损失函数进行参数调整。研究表明，采用联合损失函数并结合Adam优化器算法时，可获得较好的识别效果。此外，在实验部分针对奶山羊的特征选取问题上，对奶山羊的羊脸区域与奶山羊的全身区域分别采用了三元组损失函数与孪生网络，验证了对奶山羊的识别仅靠羊脸区域的特征时准确率较低；此外，针对网络的训练，本文不仅通过YOLOv3(you only look once)以及孪生网络(siamese network)验证了奶山羊本身属于高相似度的数据集，而且针对奶山羊数据集分别采用三元组损失函数与交叉熵损失函数作为唯一的损失函数，并验证了该方法的有效性。

结果

奶山羊识别的最高精准度为93.077%，相较于Triplet-Loss损失函数74.615%的准确率以及CrossEntropy-Loss 89.615%准确率有了较大提升。

结论

本文提出的基于深度学习的高相似度的奶山羊识别方法不仅具有较高的准确率，而且在奶山羊个体身份识别方面具有极大的应用价值，有助于准确识别羊的身份，为相似度高的动物个体身份识别提供了思路。

Abstract

Objective

It is essential for the quick response in tracking information of animals for intelligent agriculture and animal husbandry nowadays. Individual identification of animals has been one of the challenging issues in real-time monitoring. Different from traditional methods with high harmfulness such as imprinting

our deep learning based method is adopted to implement image recognition for the several of animal and human as well as the unclear multi-features relationship.

Method

First

our computer vision method is demonstrated for individual recognition of dairy goat based on deep learning. The 26 goats' pictures-oriented are acquired including the head and other parts. Fancy fancy principal components analysis (PCA) is adopted as the data expansion methods to expand the dataset. A sum of 1 040 goats' images are randomly selected for training

and 260 images are used as independent test sets; single shot MultiBox detection (SSD) network based dataset preprocessing is initial to be required. Our demonstration uses the siamese network for preliminary learning. The network structure and learning rate optimization algorithm are employed to adjust the parameters but not suitable for individual identity classification. It verifies the goat itself in terms of the highly similar data set. The effect of whole goat image is better than single head image. This obtained result has been greatly improved from the training of original head to the whole body in the context of is the solo Triplet-Loss function. The original image input of the Triplet-Loss function is composed of three pictures. Because the dairy goat is proven to have high similarity of individual based on the siamese network

it is not required to conduct the data sets integration derived of the Triplet-Loss function as well as the set of different goats images is complicated based on manual method. Next

Triplet-Loss function in dataset has its potentials compared with the siamese network method. Our joint loss function and transfer learning model residual neural network(ResNet18) obtain the goat information in terms of deep network structure. Finally

the joint loss function takes Adam as the optimizer algorithm

our demonstration can get qualified recognition as the hard batch (difficult triplets) of Triplet-Loss function are not required in the context of Triplet-Loss function and CrossEntropy-Loss function and related parameters. In addition of goats

we use the Triplet-Loss function and siamese network to option the feature for the goat face region and the whole goat region verifies the features of the goat face region recognition is not good in terms of high accuracy rate. Our illustration is not only uses you only look once (YOLOv3) network and siamese network to identify goats

but also uses transfer learning model to learn. The siamese network verifies that the goat itself based on high similarity data set

and the Triplet-Loss function and CrossEntropy-Loss function are used as final loss function to verify the effectiveness of the method.

Result

The SSD network was used to preprocess the dataset. The demonstrated results illustrate that the accuracy can be improved from 86% to 93.077% by combining the joint loss function with Adam algorithm. When the joint loss function is used with Adam optimization algorithm as well as the joint loss function accounts for a certain proportion

other correlation can be realized by adjusting the parameters

the result will be obtain better recognition effect. Just compared with 74.615% of Triplet-Loss function and 89.615% of CrossEntropy-Loss function

the highest recognition accuracy is 93.077%.

Conclusion

Our higher recognition effect of goat analysis is based on the model of deep learning. These goats cannot just get more effective facial features recognition but can obtain higher accuracy derived of the whole goat body. Intelligent research of individual goat should be conducted based on the segmented attribute of each part of the goat body. Furthermore

our research can lower high labor costs issues based on deep learning model in terms of computer vision archives.

关键词

Keywords

references

Chen H L. 2016. Application of RFID technology in large scale dairy farm. Shihezi Science and Technology, (2): 6-8

陈红莉. 2016. 基于RFID技术在规模化奶牛场中的应用. 石河子科技, (2): 6-8 [DOI: 10.3969/j.issn.1008-0899.2016.02.004]

Chopra S, Hadsell R and LeCun Y. 2005. Learning a similarity metric discriminatively, with application to face verification//Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, USA: IEEE: 539-546[ DOI: 10.1109/CVPR.2005.202 http://dx.doi.org/10.1109/CVPR.2005.202 ]

He K M, Zhang X Y, Ren S Q and Sun J. 2016. Deep residual learning for image recognition//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE: 770-778 [ DOI: 10.1109/CVPR.2016.90 http://dx.doi.org/10.1109/CVPR.2016.90 ]

Huang J H and Tian Z. 2019. Cow face recognition based on improved VGG convolutional neural network. Hubei Agricultural Mechanization, (13): #126

黄俊华, 田壮. 2019. 基于改进VGG卷积神经网络的奶牛牛脸识别研究. 湖北农机化, (13): #126 [DOI: 10.3969/j.issn.1009-1440.2019.13.101]

Liang K and Zhang Z C. 2013. Dairy cattle breeding management applications based on UHF RFID system. Journal of Agricultural Mechanization Research, (12): 182-184

梁坤, 张哲纯. 2013. 超高频RFID系统在奶牛养殖管理中的应用方案. 农机化研究, (12): 182-184 [DOI: 10.3969/j.issn.1003-188X.2013.12.046]

Liu D, Zhao K X and He D J. 2016. Real-time target detection for moving cows based on Gaussian mixture model. Transactions of the Chinese Society for Agricultural Machinery, 47(5): 288-294

刘冬, 赵凯旋, 何东健. 2016. 基于混合高斯模型的移动奶牛目标实时检测. 农业机械学报, 47(5): 288-294 [DOI: 10.6041/j.issn.1000-1298.2016.05.039]

Liu W, Wang F, Zhang S N and Guo R. 2020. Pig comfort monitoring based on improved SSD at night. China Feed, (14): 28-31

刘伟, 王芳, 张苏楠, 郭融. 2020. 基于改进SSD的夜间猪群舒适度监测. 中国饲料, (14): 28-31 [DOI: 10.15906/j.cnki.cn11-2975/s.20201407]

Meng H W, Li J B, Li Y P, Kan Z, Xue L Y and Qi J T. 2015. Application of RFID technology in animal husbandry. Xinjiang Agricultural Mechanization, (3): 26-28

蒙贺伟, 李景彬, 李亚萍, 坎杂, 薛令阳, 戚江涛. 2015. RFID技术在畜牧产业中的应用分析, 新疆农机化, (3): 26-28 [DOI: 10.13620/j.cnki.issn1007-7782.2015.03.012]

Qin X and Song G F. 2019. Pig face recognition algorithm based on bilinear convolution neural network. Journal of Hangzhou Dianzi University, 39(2): 12-17

秦兴, 宋各方. 2019. 基于双线性卷积神经网络的猪脸识别算法. 杭州电子科技大学学报, 39(2): 12-17 [DOI: 10.13954/j.cnki.hdu.2019.02.003]

Redmon J and Farhadi A. 2017. YOLOv3: an incremental improvement//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE: 779-788

Schroff F, Kalenichenko D and Philbin J. 2015. FaceNet: a unified embedding for face recognition and clustering//Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, USA: IEEE: 815-823[ DOI: 10.1109/CVPR.2015.7298682 http://dx.doi.org/10.1109/CVPR.2015.7298682 ]

Sun Y, Chen Y H, Wang X G and Tang X O. 2014b. Deep learning face representation by joint identification-verification//Proceedings of the 27th International Conference on Neural Information Processing Systems. Montreal, Canada: NIPS: 1988-1996 [ DOI: 10.5555/2969033.2969049 http://dx.doi.org/10.5555/2969033.2969049 ]

Sun Y, Wang X G and Tang X U. 2014a. Deep learning face representation from predicting 10 000 classes//Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA: IEEE: 1891-1898[ DOI: 10.1109/CVPR.2014.244 http://dx.doi.org/10.1109/CVPR.2014.244 ]

Taigman Y, Yang M, Ranzato M and Wolf L. 2014. DeepFace: closing the gap to human-level performance in face verification//Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA: IEEE: 1701-1708[ DOI: 10.1109/CVPR.2014.220 http://dx.doi.org/10.1109/CVPR.2014.220 ]

Xuan C Z, Ma Y H, Wu P, Zhang L N, Hao M and Zhang X Y. 2016b. Behavior classification and recognition for facility breeding sheep based on acoustic signal weighted feature. Transactions of the Chinese Society of Agricultural Engineering, 32(19): 195-202

宣传忠, 马彦华, 武佩, 张丽娜, 郝敏, 张曦宇. 2016b. 基于声信号特征加权的设施养殖羊行为分类识. 农业工程学报, 32(19): 195-202 [DOI: 10.11975/j.issn.1002-6819.2016.19.027]

Xuan CZ, Wu P, Ma Y H, Zhang L N, Han D and Liu Y Q. 2015. Vocal signal recognition of ewes based on power spectrum and formant analysis method. Transactions of the Chinese Society of Agricultural Engineering, 31(24): 219-224

宣传忠, 武佩, 马彦华, 张丽娜, 韩丁, 刘艳秋. 2015. 基于功率谱和共振峰的母羊发声信号识别. 农业工程学报, 31(24): 219-224 [DOI: 10.11975/j.issn.1002-6819.2015.24.033]

Xuan C Z, Wu P, Zhang L N, Ma Y H, Zhang Y A and Wu J. 2016a. Feature parameters extraction and recognition method of sheep cough sound. Transactions of the Chinese Society for Agricultural Machinery, 47(3): 342-348

宣传忠, 武佩, 张丽娜, 马彦华, 张永安, 邬娟. 2016a. 羊咳嗽声的特征参数提取与识别方法. 农业机械学报, 47(3): 342-348 [DOI: 10.6041/j.issn.1000-1298.2016.03.048]

Zhang H F and Shen Y P. 2012. Application of RFID technology in animal identification and tracking management. Chinese Qinghai Journal of Animal and Veterinary Sciences, 42(3): 36-38

张海峰, 沈媛萍. 2012. RFID技术在动物识别与跟踪管理中的应用. 青海畜牧兽医杂志, 42(3): 36-38 [DOI: 10.3969/j.issn.1003-7950.2012.03.026]

Zhang M T, Mi N, Yu Y, Shan X Y, Yan G and Guo Y C. 2018. Recognition of individual dairy cattle based on feature fusion. Jiangsu Agricultural Sciences, 46(24): 278-281

张满囤, 米娜, 于洋, 单新媛, 阎刚, 郭迎春. 2018. 基于特征融合的奶牛个体识别. 江苏农业科学, 46(24): 278-281 [DOI: 10.15889/j.issn.1002-1302.2018.24.075]

Vinyals O, Blundell C, Lillicrap T, Kavukcuoglu K and Wierstra D. 2016. Matching networks for one shot learning//Advances in Neural Information Processing Systems. Barcelona, Spain: NIPS, 2016: 3637-3645

Alert me when the article has been cited

提交

Correlation subdomain alignment network based cross-domain hyperspectral image classification method

Model distillation for high-level semantic understanding： a survey

Advances and challenges in facial expression analysis

Survey on the applications of deep learning to histopathology

Automatic recognition and classification of diabetic retinopathy images by combining an attention mechanism and an efficient network