基于改进U-Net网络的多尺度番茄病害分割算法

doi:10.3778/j.issn.1002-8331.2105-0201

摘要/Abstract

摘要： 针对当前农作物病害分割与识别模型病斑分割精度低、数据集不充分、训练速度过慢等问题，构建了一种基于改进的U-Net网络多尺度番茄叶部病害分割算法。在U-Net网络结构基础上进行改进，减小图像输入尺寸，在编码器中使用非对称Inception多通道卷积替换传统卷积，实现多尺度提取病害特征，提升模型准确度；在解码器中加入注意力模块，关注番茄病害边缘，减小上采样噪声；引入GN加速模型收敛，并将改进U-Net网络用在PlantVillage数据集上进行预训练，提高模型的分割准确度和速度。改进后的方法准确率、召回率和MIoU分别为92.9%、91.1%、93.6%，实验结果表明，该方法能够有效地提高模型对番茄的病害分割性能。

关键词: 病害, 图像分割, 多尺度特征提取, 注意力机制

Abstract: Aiming at the problems of current crop disease segmentation and recognition models with low disease spot segmentation accuracy, insufficient dataset and too slow training speed, a multi-scale tomato leaf disease segmentation algorithm based on improved U-Net network is constructed. This model makes improvements on the basis of the U-Net network structure, reduces the image input size, uses asymmetric Inception multi-channel convolution in the encoder to replace the traditional convolution, which can realize multi-scale extraction of disease features and improve the accuracy of the model. The decoder adds attention module to pay attention to the edge of the tomato disease and reduce the upsampling noise. Moreover, this model introduces GN to accelerate its convergence, and using the improved U-Net network to pretrain on the PlantVillage dataset can improve the segmentation accuracy and speed of the model. The accuracy, recall and MIoU of the proposed method are respectively 92.9%, 91.1% and 93.6%. The experimental results show that this method can effectively improve the model’s performance of tomato disease segmentation.

Key words: disease, image segmentation, multi-scale feature extraction, attention mechanism

赵小虎, 李晓, 叶圣, 李晓, 冯伟, 尤星懿. 基于改进U-Net网络的多尺度番茄病害分割算法[J]. 计算机工程与应用, 2022, 58(10): 216-223.

ZHAO Xiaohu, LI Xiao, YE Sheng, LI Xiao, FENG Wei, YOU Xingyi. Multi-Scale Tomato Disease Segmentation Algorithm Based on Improved U-Net Network[J]. Computer Engineering and Applications, 2022, 58(10): 216-223.

参考文献

[1] 张建华，孔繁涛，吴建寨，等.基于改进VGG卷积神经网络的棉花病害识别模型[J].中国农业大学学报，2018，23（11）：161-171.
ZHANG J H，KONG F T，WU J Z，et al.Cotton disease identification model based on improved VGG convolution neural network[J].Journal of China Agricultural University，2018，23（11）：161-171.
[2] 张武，黄帅，汪京京，等.复杂背景下小麦叶部病害图像分割方法研究[J].计算机工程与科学，2015，37（7）：1349-1354.
ZHANG W，HUANG S，WANG J J，et al.A segmentation method for wheat leaf images with disease in complex background[J].Computer Engineering & Science，2015，37（7）：1349-1354.
[3] MA J C，DU K M，ZHANG L X.A segmentation method for greenhouse vegetable foliar disease spots images using color information and region growing[J].Computers and Electronics in Agriculture，2017，142：110-117.
[4] 孙俊，宋佳，武小红，等.基于改进Otsu算法的生菜叶片图像分割方法[J].江苏大学学报（自然科学版），2018，39（2）：179-184.
SUN J，SONG J，WU X H，et al.Image segmentation method of lettuce leaf based on improved Otsu algorithm[J].Journal of Jiangsu University（Natural Science Edition），2018，39（2）：179-184.
[5] 王昌龙，张远东，缪宏，等.双通道卷积神经网络在南瓜病害识别上的应用[J].计算机工程与应用，2021，57（5）：183-189.
WANG C L，ZHANG Y D，MIAO H，et al.Application of double channel convolutional neural network in pumpkin diseases identification[J].Computer Engineering and Applications，2021，57（5）：183-189.
[6] WANG Z B，WANG K Y，PAN S H，et al.Segmentation of crop disease images with an improved k-means clustering algorithm[J].Applied Engineering in Agriculture，2018，34（2）：277-289.
[7] LONG J，SHELHAMER E，DARRELL T.Fully convolutional networks for semantic segmentation[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition，2015：3431-3440.
[8] RONNEBERGER O，FISCHER P，BROX T.U-Net：convolutional networks for biomedical image segmentation[J].Medical Image Computing and Computer-Assisted Intervention，2015，9531：234-241.
[9] 许慧敏.基于深度学习U-Net模型的高分辨率遥感影像分类方法研究[D].成都：西南交通大学，2018.
XU H M.Method research of high resolution remote sensing image classification based on U-Net model of deep learning[D].Chengdu：Southwest Jiaotong University，2018.
[10] 段凌凤，熊雄，刘谦，等.基于深度全卷积神经网络的大田稻穗分割[J].农业工程学报，2018，34（12）：202-209.
DUAN L F，XIONG X，LIU Q，et al.Field rice panicle segmentation based on deep full convolutional neural network[J].Transactions of the Chinese Society of Agricultural Engineering，2018，34（12）：202-209.
[11] 方晨晨，石繁槐.基于改进深度残差网络的番茄病害图像识别[J].计算机应用，2020，40（S1）：203-208.
FANG C C，SHI F H.Image recognition of tomato diseases based on improved deep residual network[J].Journal of Computer Applications，2020，40（S1）：203-208.
[12] 孔德锋.一种迁移学习算法在番茄病害检测上的应用[J].湖北农业科学，2020，59（7）：199-203.
KONG D F.Application of a migration learning algorithm in tomato disease detection[J].Hubei Agricultural Sciences，2020，59（7）：199-203.
[13] JIANG P，CHEN Y，LIU B，et al.Real-time detection of apple leaf diseases using deep learning approach based on improved convolutional neural networks[J].IEEE Access，2019，7：59069-59080.
[14] 王振，师韵，李玉彬.基于改进全卷积神经网络的玉米叶片病斑分割[J].计算机工程与应用，2019，55（22）：127-132.
WANG Z，SHI Y，LI Y B.Segmentation of corn leaf diseases based on improved fully convolutional neural network[J].Computer Engineering and Applications，2019，55（22）：127-132.
[15] 张善文，王振，王祖良.多尺度融合卷积神经网络的黄瓜病害叶片图像分割方法[J].农业工程学报，2020，36（16）：149-157.
ZHANG S W，WANG Z，WANG Z L.Method for image segmentation of cucumber disease leaves based on multi-scale fusion convolutional neural networks[J].Transactions of the Chinese Society of Agricultural Engineering，2020，36（16）：149-157.
[16] 宋余庆，谢熹，刘哲，等.基于多层EESP深度学习模型的农作物病虫害识别方法[J].农业机械学报，2020，51（8）：196-202.
SONG Y Q，XIE X，LIU Z，et al.Crop pests and diseases recognition method based on multi-level EESP model[J].Transactions of the Chinese Society for Agricultural Machinery，2020，51（8）：196-202.
[17] 余小东，杨孟辑，张海清，等.基于迁移学习的农作物病虫害检测方法研究与应用[J].农业机械学报，2020，51（10）：252-258.
YU X D，YANG M Y，ZHANG H Q，et al.Research and application of crop diseases detection method based on transfer learning[J].Transactions of the Chinese Society for Agricultural Machinery，2020，51（10）：252-258.
[18] 史冰莹，李佳琦，张磊，等.基于CNN的农作物病虫害图像识别模型[J].计算机系统应用，2020，29（6）：89-96.
SHI B Y，LI J Q，ZHANG L，et al.Recognition model of crop pests and diseases image based on CNN[J].Computer Systems & Applications，2020，29（6）：89-96.
[19] 任守纲，贾馥玮，顾兴健，等.反卷积引导的番茄叶部病害识别及病斑分割模型[J].农业工程学报，2020，36（12）：186-195.
REN S G，JIA F W，GU X J，et al.Recognition and segmentation model of tomato leaf diseases based on deconvolution-guiding[J].Transactions of the Chinese Society of Agricultural Engineering，2020，36（12）：186-195.
[20] SZEGEDY C，LIU W，JIA Y Q，et al.Going deeper with convolutions[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition，2015：1-9.
[21] DING X H，GUO Y C，DING G D，et al.ACNet：strengthening the kernel skeletons for powerful CNN via asymmetric convolution blocks[C]//2019 IEEE/CVF International Conference on Computer Vision，2019：1911-1920.
[22] SZEGEDY C，VANHOUCKE V，IOFFE S，et al.Rethinking the inception architecture for computer vision[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition，2016：2818-2826.
[23] 梁礼明，盛校棋，蓝智敏，等.自适应尺度信息的U型视网膜血管分割算法[J].光学学报，2019，39（8）：126-140.
LIANG L M，SHENG X Q，LAN Z M，et al.U-shaped retinal vessel segmentation algorithm based on adaptive scale information[J].Acta Optica Sinica，2019，39（8）：126-140.
[24] WOO S，PARK J，LEE J Y，et al.CBAM：convolutional block attention module[C]//15th European Conference on Computer Vision，2018：3-19.
[25] JIE H，LI S，GANG S，et al.Squeeze-and-excitation networks[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2020，42：2011-2023.
[26] IOFFE S，SZEGEDY C.Batch normalization：accelerating deep network training by reducing internal covariate shift[J].arXiv：1502.03167，2015.
[27] WU Y X，HE K M.Group normalization[J].International Journal of Computer Vision，2018，128（3）：742-755.
[28] SHAMIR O.Making gradient descent optimal for strongly convex stochastic optimization[J].arXiv：1109.5647，2011.
[29] SHELHAMER E，LONG J，DARRELL T.Fully convolutional networks for semantic segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2015，39（4）：640-651.