一种场景自适应的双分支牛脸高效识别算法
A Scene-Adaptive Dual-Branch Efficient Cattle Facial Recognition Algorithm
- 电子学报 2024年52卷第9期 页码：3251-3261
- 作者机构：
1.内蒙古工业大学电力学院，内蒙古呼和浩特 010051
2.大规模储能技术教育部工程研究中心，内蒙古呼和浩特 010080
3.内蒙古自治区高等学校智慧能源技术与装备工程研究中心，内蒙古呼和浩特 010080
- 作者简介：
[ "焦 杰 男，1998年3月出生于山西省大同市.现为内蒙古工业大学电力学院硕士研究生.基于计算机视觉与深度学习的目标检测与识别. E-mail: 1344148511@qq.com" ]
[ "齐咏生 男，1975年12月出生于内蒙古包头市.现为内蒙古工业大学电力学院教授、副院长.主要研究方向为移动机器人协同控制技术. E-mail: qys@imut.edu.cn" ]
[ "刘利强 男，1975年5月出生于内蒙古包头市.现为内蒙古工业大学电力学院教授、硕士生导师.主要研究方向为计算机视觉、新能源发电技术.E-mail: llqiang@imut.edu.cn" ]
[ "李永亭 女，1975年11月出生于内蒙古乌海市.现为内蒙古工业大学电力学院副教授、硕士生导师.主要研究方向为为复杂工业过程故障监测与诊断. E-mail: liyt@imut.edu.cn" ]
[ "王朝霞 男，1985年12月出生于内蒙古呼和浩特市.现为内蒙古工业大学电力学院讲师.主要研究方向为为3维高度成像技术、卫星组网观测技术. E-mail: wangzx@imut.edu.cn" ]
- 基金信息：国家自然科学基金(62363029;62241309);内蒙古科技计划项目(2020GG0283;2021GG164);内蒙古自然科学基金(2022MS06018;2021MS06018)
DOI：10.12263/DZXB.20230829
中图分类号： TP391.4收稿：2023-09-01，
修回：2024-02-05，
纸质出版：2024-09-25
- 稿件说明：
移动端阅览
焦杰, 齐咏生, 刘利强, 等. 一种场景自适应的双分支牛脸高效识别算法[J]. 电子学报, 2024, 52(09): 3251-3261.
JIAO Jie, QI Yong-sheng, LIU Li-qiang, et al. A Scene-Adaptive Dual-Branch Efficient Cattle Facial Recognition Algorithm[J]. Acta Electronica Sinica, 2024, 52(09): 3251-3261.
焦杰, 齐咏生, 刘利强, 等. 一种场景自适应的双分支牛脸高效识别算法[J]. 电子学报, 2024, 52(09): 3251-3261. DOI：10.12263/DZXB.20230829
JIAO Jie, QI Yong-sheng, LIU Li-qiang, et al. A Scene-Adaptive Dual-Branch Efficient Cattle Facial Recognition Algorithm[J]. Acta Electronica Sinica, 2024, 52(09): 3251-3261. DOI：10.12263/DZXB.20230829
摘要
随着智慧牧业的高速发展，牛脸识别已成为牛场智能化养殖的关键，但牛场养殖环境复杂且动物的自主能动性差，导致牛脸数据采集与识别过程会受到模糊、遮挡和光照等环境因素的严重干扰.针对此问题，提出一种复杂场景自适应选择双分支牛脸高效识别算法.该算法首先设计了基于像素融合的数据增强策略，通过Beta分布计算融合系数，将牛的左右脸图像按融合系数进行像素级整合，在丰富样本特征信息同时，增强网络学习模糊和遮挡下的牛脸特征，提升网络对复杂场景的泛化能力；其次，在主干特征提取网络中引入一种新型注意力机制CDAA（Composite Dual-branch Adaptive Attention），可随着场景信息变换，自适应加强通道与空间注意力分支的权重，提高网络在复杂场景下的特征筛选能力；之后，设计FaceNet与U-LBP（Uniform Local Binary Patterns）结合的双分支特征提取结构，并将提取的特征向量实现自适应加权融合，增加网络在过亮或过暗环境下的鲁棒性；最后，在损失函数中加入改进交叉熵损失（Focal Loss），根据场景信息复杂度动态调控权重系数，实现对难易分类样本自主控制.为检测算法的有效性和实时性，在特定数据集上进行消融试验，与多种典型识别算法进行对比.实结果表明，提出的算法能很好满足实时性要求，在开集测试集上准确率达到87.53%，识别速度达到108帧/s，且在复杂场景下，识别效果均优于对比网络.
Abstract
With the rapid development of intelligent animal husbandry
cattle facial recognition has become a key aspect of intelligent farming in cattle ranches. However
due to the complexity of the ranching environment and the limited autonomy of animals
the process of collecting and identifying cattle facial data is severely affected by environmental factors such as blurriness
occlusion
and lighting. To address this issue
a complex scene-adaptive dual-branch efficient cattle facial recognition algorithm is proposed. This algorithm first designs a data augmentation strategy based on pixel fusion. By calculating fusion coefficients using the beta distribution
the left and right facial images of cattle are integrated at the pixel level
enriching the sample's feature information. Simultaneously
the algorithm enhances the network's ability to learn cattle facial features under blurriness and occlusion
improving its generalization ability to complex scenes. Furthermore
a novel attention mechanism called composite dual-branch adaptive attention (CDAA) is introduced into the main feature extraction network. This mechanism adaptively strengthens the weights of the channel and spatial attention branches as scene information changes
enhancing the network's feature selection ability in complex scenarios. Next
a dual-branch feature extraction structure combining FaceNet and U-LBP (Uniform Local Binary Patterns) is designed. The extracted feature vectors are adaptively weighted and fused to increase the network's robustness in overly bright or dark environments. Finally
an improved cross-entropy loss (Focal Loss) is incorporated into the loss function. Weight coefficients are dynamically adjusted based on the complexity of the scene information to autonomously control the classification of difficult and easy samples. To evaluate the effectiveness and real-time performance of the algorithm
ablation experiments are conducted on a specific dataset
comparing it with various typical recognition algorithms. The experimental results indicate that the proposed algorithm effectively meets real-time requirements
achieving an accuracy of 87.53% on the open test set with a recognition speed of 108 frames per second. Moreover
in complex scenarios
the recognition performance of the proposed algorithm surpasses that of the comparative networks.
关键词
Keywords
references
MORRONE S , DIMAURO C , GAMBELLA F , et al . Industry 4.0 and precision livestock farming (PLF): An up to date overview across animal productions [J ] . Sensors , 2022 , 22 ( 12 ): 4319 .
MANCUSO D , CASTAGNOLO G , PORTO S M C . Cow behavioural activities in extensive farms: Challenges of adopting automatic monitoring systems [J ] . Sensors , 2023 , 23 ( 8 ): 3828 .
MAHMUD M S , ZAHID A , DAS A K , et al . A systematic literature review on deep learning applications for precision cattle farming [J ] . Computers and Electronics in Agriculture , 2021 , 187 : 106313 .
WILLIAMS L R , FOX D R , BISHOP-HURLEY G J , et al . Use of radio frequency identification (RFID) technology to record grazing beef cattle water point use [J ] . Computers and Electronics in Agriculture , 2019 , 156 : 193 - 202 .
WANG M , YIN X F , ZHU Y M , et al . Representation learning and pattern recognition in cognitive biometrics: A survey [J ] . Sensors , 2022 , 22 ( 14 ): 5111 .
SHU X , TANG H , HUANG S C . Face spoofing detection based on chromatic ED-LBP texture feature [J ] . Multimedia Systems , 2021 , 27 ( 2 ): 161 - 176 .
吴靖 , 叶晓晶 , 黄峰 , 等 . 基于深度学习的单帧图像超分辨率重建综述 [J ] . 电子学报 , 2022 , 50 ( 9 ): 2265 - 2294 .
WU J , YE X J , HUANG F , et al . A review of single image super-resolution reconstruction based on deep learning [J ] . Acta Electronica Sinica , 2022 , 50 ( 9 ): 2265 - 2294 . (in Chinese)
BINDAL N , GHUMAAN R S , SOHI P J S , et al . A systematic review of state-of-the-art noise removal techniques in digital images [J ] . Multimedia Tools and Applications , 2022 , 81 ( 22 ): 31529 - 31552 .
KUMAR S , SINGH S K , SINGH A K . Muzzle point pattern based techniques for individual cattle identification [J ] . IET Image Processing , 2017 , 11 ( 10 ): 805 - 814 .
PAUL M , KARSH R K , AHMED TALUKDAR F . Image hashing based on shape context and speeded up robust features (SURF) [C ] // 2019 International Conference on Automation, Computational and Technology Management (ICACTM) . Piscataway : IEEE , 2019 : 464 - 468 .
LIU Y J . Athletes' throwing action recognition method based on PCA-LBP algorithm [J ] . International Journal of Computational Intelligence Studies , 2023 , 12 ( 1/2 ): 130 - 141 .
邵志文 , 周勇 , 谭鑫 , 等 . 基于深度学习的表情动作单元识别综述 [J ] . 电子学报 , 2022 , 50 ( 8 ): 2003 - 2017 .
SHAO Z W , ZHOU Y , TAN X , et al . Survey of expression action unit recognition based on deep learning [J ] . Acta Electronica Sinica , 2022 , 50 ( 8 ): 2003 - 2017 . (in Chinese)
程旭 , 宋晨 , 史金钢 , 等 . 基于深度学习的通用目标检测研究综述 [J ] . 电子学报 , 2021 , 49 ( 7 ): 1428 - 1438 .
CHENG X , SONG C , SHI J G , et al . A survey of generic object detection methods based on deep learning [J ] . Acta Electronica Sinica , 2021 , 49 ( 7 ): 1428 - 1438 . (in Chinese)
汪宇玲 , 陈立 , 黎明 , 等 . 基于迹变换和旋转增量调制特征的模糊人脸识别 [J ] . 电子学报 , 2021 , 49 ( 12 ): 2437 - 2448 .
WANG Y L , CHEN L , LI M , et al . Rotational delta modulation feature and its application in blurry face recognition based on trace transform [J ] . Acta Electronica Sinica , 2021 , 49 ( 12 ): 2437 - 2448 . (in Chinese)
FUENTES S , GONZALEZ VIEJO C , TONGSON E , et al . The livestock farming digital transformation: Implementation of new and emerging technologies using artificial intelligence [J ] . Animal Health Research Reviews , 2022 , 23 ( 1 ): 59 - 71 .
WENG Z , MENG F S , LIU S Q , et al . Cattle face recognition based on a Two-Branch convolutional neural network [J ] . Computers and Electronics in Agriculture , 2022 , 196 : 106871 .
WENG Z , FAN L Z , ZHANG Y , et al . Facial recognition of dairy cattle based on improved convolutional neural network [J ] . IEICE Transactions on Information and Systems , 2022 , E105.D( 6 ): 1234 - 1238 .
GONG H , PAN H H , CHEN L , et al . Facial recognition of cattle based on SK-ResNet [J ] . Scientific Programming , 2022 , 2022 : 5773721 .
MA X X , LI Z W , ZHANG L . An improved ResNet-50 for garbage image classification [J ] . Tehnicki Vjesnik - Technical Gazette , 2022 , 29 ( 5 ): 1552 - 1559 .
LI Z , LEI X M . Cattle face recognition under partial occlusion [J ] . Journal of Intelligent & Fuzzy Systems , 2022 , 43 ( 1 ): 67 - 77 .
CHEN T T , LIN G J , ZHANG H J , et al . Non-uniform illumination face recognition method based on improved MobileNetV1 [J ] . Journal of Computers , 2023 , 34 ( 1 ): 145 - 155 .
杨蜀秦 , 刘杨启航 , 王振 , 等 . 基于融合坐标信息的改进YOLO V4模型识别奶牛面部 [J ] . 农业工程学报 , 2021 , 37 ( 15 ): 129 - 135 .
YANG S Q , LIU Y Q H , WANG Z , et al . Improved YOLO V4 model for face recognition of diary cow by fusing coordinate information [J ] . Transactions of the Chinese Society of Agricultural Engineering , 2021 , 37 ( 15 ): 129 - 135 . (in Chinese)
WANG Y F , HUA C C , DING W L , et al . Real-time detection of flame and smoke using an improved YOLOv4 network [J ] . Signal, Image and Video Processing , 2022 , 16 ( 4 ): 1109 - 1116 .
张艳 , 李星汕 , 孙叶美 , 等 . 基于通道注意力与特征融合的水下目标检测算法 [J ] . 西北工业大学学报 , 2022 , 40 ( 2 ): 433 - 441 .
ZHANG Y , LI X S , SUN Y M , et al . Underwater object detection algorithm based on channel attention and feature fusion [J ] . Journal of Northwestern Polytechnical University , 2022 , 40 ( 2 ): 433 - 441 . (in Chinese)
LIN W H , ZHONG J X , LIU S , et al . ROIMIX: Proposal-fusion among multiple images for underwater object detection [C ] // 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . Piscataway : IEEE , 2020 : 2588 - 2592 .
ZHANG H Y , CISSE M , DAUPHIN Y N , et al . MixUp: Beyond empirical risk minimization [EB/OL ] . ( 2018-04-27 )[ 2023-08-29 ] . https://arxiv.org/abs/1710.09412 https://arxiv.org/abs/1710.09412 .
PENG C , LIU Y K , YUAN X P , et al . Research of image recognition method based on enhanced inception-ResNet-V2 [J ] . Multimedia Tools and Applications , 2022 , 81 ( 24 ): 34345 - 34365 .
PANG Y , YU W B , ZHANG Y A , et al . Sheep face recognition and classification based on an improved MobilenetV2 neural network [J ] . International Journal of Advanced Robotic Systems , 2023 , 20 ( 1 ): 172988062311529 .
张红斌 , 熊其鹏 , 蒋子良 , 等 . 联合SENet异构层特征融合与集成学习的材质图像识别 [J ] . 控制与决策 , 2022 , 37 ( 6 ): 1632 - 1642 .
ZHANG H B , XIONG Q P , JIANG Z L , et al . Material image recognition combining heterogeneous-layer feature fusion of SENet and ensemble learning [J ] . Control and Decision , 2022 , 37 ( 6 ): 1632 - 1642 . (in Chinese)
SCHROFF F , KALENICHENKO D , PHILBIN J . FaceNet: A unified embedding for face recognition and clustering [C ] // 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2015 : 815 - 823 .
张宏鸣 , 周利香 , 李永恒 , 等 . 基于改进 MobileFaceNet 的羊脸识别方法 [J ] . 农业机械学报 , 2022 , 53 ( 5 ): 267 - 274 .
ZHANG H M , ZHOU L X , LI Y H , et al . Sheep face recognition method based on improved MobileFaceNet [J ] . Transactions of the Chinese Society for Agricultural Machinery , 2022 , 53 ( 05 ): 267 - 274 . (in Chinese)
0
浏览量
11
下载量
2
CSCD
- 网络PDF下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies