基于YOLO的小型动物识别系统设计
|
摘要
针对常见动物识别的问题,提出了一种基于YOLO的识别分类系统。通过搭载linux系统的树莓派完成实施检测运行的识别程序。利用YOLO算法进行目标物体检测,其单个神经网络能够在一帧图像中直接预测边界边框(bounding box)和分类概率(classified probability),最后根据输入图像来输出动物物种的识别结果。其识别准确率针对鸟类可达94.61%,狗类为90.60%,牛类为79.03%。
To solve the problem of species identification and population identification of small animals, a recognition and classification system based on YOLO(You Only Look Once) is proposed in this paper. The recognition program is implemented on a Raspberry Pi with Linux system, and YOLO is used as the detection framework. The single neural network can directly predict the bounding box and the classified probability in a frame image, and output the animal recognized according to the input image. The accuracy rate of identification is 94.61% for birds, 90.60% for dogs and 79.03% for cattle.
To solve the problem of species identification and population identification of small animals, a recognition and classification system based on YOLO(You Only Look Once) is proposed in this paper. The recognition program is implemented on a Raspberry Pi with Linux system, and YOLO is used as the detection framework. The single neural network can directly predict the bounding box and the classified probability in a frame image, and output the animal recognized according to the input image. The accuracy rate of identification is 94.61% for birds, 90.60% for dogs and 79.03% for cattle.
引文
[1]Huang Y,Wu R,Sun Y,et al.Vehicle Logo Recognition System Based on Convolutional Neural Networks With a Pretraining Strategy[J].IEEE Transactions on Intel igent Transportation Systems,2015.16(4):1951-1960
[2]王福建,张俊,卢国权等.基于YOLO的车辆信息检测和跟踪系统[J].工业控制计算机,2018.7:89-91
[3]Dlagnekov L,Belongie S.Recognizing cars[J].Ecological Modelling,2005.113(13):71-81
[4]邱道尹,张红涛,刘新宇等.基于机器视觉的大田害虫检测系统[J].农业机械学报,2007.38(1):120-122
[5]Redmon J,Farhadi A.YOLO9000:Better,Faster,Stronger[J],2016:6517-6525
[6]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
[7]魏海坤.神经网络结构设计的理论与方法[M].国防工业出版社,2005.
[8]魏星.基于SVM的山体滑坡灾害图像识别方法[J].电子测量技术,2013.36(8):65-70
[9]李亮,丁万山.基于ARM的纸币号码识别系统[J].电子测量技术,2008.31(10):68-72
[10]程曦,吴云志,张友华等.基于深度卷积神经网络的储粮害虫图像识别[J].中国农学通报,2018.34(1):154-158
[11]王宇宁,庞智恒,袁德明.基于YOLO算法的车辆实时检测[J].武汉理工大学学报,2016.38(10):41-46
[12]严晓波.基于单镜头的车辆检测与跟踪方法研究[D].贵州民族大学,2018.
[13]潘蓉,孙伟.基于预分割和回归的深度学习目标检测[J].光学精密工程,2017.25(10s):221-227
[14]刘尧.基于深度学习的气泡水平尺自动矫正系统设计[D].浙江大学,2018.
[15]李天剑,黄斌,刘江玉等.卷积神经网络物体检测算法在物流仓库中的应用[J].计算机工程,2018.44(6):176-181
[16]张春玉,黄知红.一种有效的基于卷积神经网络的车辆检索算法[J].科技创新与应用,2018.13.
[17]乐国庆.基于车载视觉系统的目标检测优化算法研究[D].北京交通大学,2017.
[2]王福建,张俊,卢国权等.基于YOLO的车辆信息检测和跟踪系统[J].工业控制计算机,2018.7:89-91
[3]Dlagnekov L,Belongie S.Recognizing cars[J].Ecological Modelling,2005.113(13):71-81
[4]邱道尹,张红涛,刘新宇等.基于机器视觉的大田害虫检测系统[J].农业机械学报,2007.38(1):120-122
[5]Redmon J,Farhadi A.YOLO9000:Better,Faster,Stronger[J],2016:6517-6525
[6]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
[7]魏海坤.神经网络结构设计的理论与方法[M].国防工业出版社,2005.
[8]魏星.基于SVM的山体滑坡灾害图像识别方法[J].电子测量技术,2013.36(8):65-70
[9]李亮,丁万山.基于ARM的纸币号码识别系统[J].电子测量技术,2008.31(10):68-72
[10]程曦,吴云志,张友华等.基于深度卷积神经网络的储粮害虫图像识别[J].中国农学通报,2018.34(1):154-158
[11]王宇宁,庞智恒,袁德明.基于YOLO算法的车辆实时检测[J].武汉理工大学学报,2016.38(10):41-46
[12]严晓波.基于单镜头的车辆检测与跟踪方法研究[D].贵州民族大学,2018.
[13]潘蓉,孙伟.基于预分割和回归的深度学习目标检测[J].光学精密工程,2017.25(10s):221-227
[14]刘尧.基于深度学习的气泡水平尺自动矫正系统设计[D].浙江大学,2018.
[15]李天剑,黄斌,刘江玉等.卷积神经网络物体检测算法在物流仓库中的应用[J].计算机工程,2018.44(6):176-181
[16]张春玉,黄知红.一种有效的基于卷积神经网络的车辆检索算法[J].科技创新与应用,2018.13.
[17]乐国庆.基于车载视觉系统的目标检测优化算法研究[D].北京交通大学,2017.