计算机科学

计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230600107-8.doi: 10.11896/jsjkx.230600107

• 图像处理&多媒体技术 • 上一篇    下一篇

基于双流卷积神经网络的稻米缺陷分割

吴一博, 郝应光, 王洪玉   

  1. 大连理工大学信息与通信工程学院 辽宁 大连 116024
  • 发布日期:2024-06-06
  • 通讯作者: 郝应光(yghao@dlut.edu.cn)
  • 作者简介:(wyb1302030@163.com)
  • 基金资助:
    中央高校基本科研业务费专项基金(DUT21YG110)

Rice Defect Segmentation Based on Dual-stream Convolutional Neural Networks

WU Yibo, HAO Yingguang, WANG Hongyu   

  1. Department of Information and Communication,Dalian University of Technology,Dalian,Liaoning 116024,China
  • Published:2024-06-06
  • About author:WU Yibo,born in 1998,postgraduate.His main research interests include dense small target segmentation based on deep learning and so on.
    HAO Yingguang,born in 1968,asso-ciate professor.His main research in-terests include modeling complex time-varying systems and image processing algorithm.
  • Supported by:
    Fundamental Research Funds for the Central Universities of Ministry of Education of China(DUT21YG110).

PDF (PC)

265

摘要： 目前水稻质量精细化评估因为没有水稻缺陷精细化检测相关工作而无法实现,传统的水稻质量评估都是基于粗略的缺陷有无分类而实现的。针对水稻缺陷像素级分类问题,提出了一种基于深度学习的水稻缺陷分割模型,该模型使用了一个改进的DoubleU-Net网络作为主要架构,分为NETWORK1和NETWORK2两部分,其中NETWORK1是基于VGG-19修改的U型网络结构,而NETWORK2是基于Swin Transformer修改的U型网络结构,将这两部分串联起来,同时融合CNN局部信息提取和Transformer全局信息提取的优势,可以更好地捕捉图像的上下文信息。同时,使用了多重损失函数,包括加权的二元交叉熵损失、加权的交并比损失和一个无需训练的智能损失网络,在提高模型训练稳定性的同时进一步提高了模型分割的精度。在制作的密集水稻缺陷数据集上进行训练测试,该模型均取得了较其他方法更好的分割性能,具有鲁棒性和较好的泛化能力。

关键词: 稻米质量评估, 语义分割, 深度学习, 卷积神经网络, Transformer

Abstract: Currently,fine-grained assessment of rice quality cannot be achieved due to the lack of related work on fine-grained detection of rice defects.Traditional rice quality assessment is based on rough classification of defect presence or absence.To address the problem of pixel-level classification of rice defects,a deep learning-based rice defect segmentation model is proposed.The model uses an improved DoubleU-Net network as the main architecture,which consists of two parts,NETWORK1 and NETWORK2.NETWORK1 is based on a modified U-shaped network structure of VGG-19,while NETWORK2 is based on a modified U-shaped network structure of Swin Transformer.The two parts are concatenated,and the advantages of CNN local information extraction and Transformer global information extraction are integrated to better capture the contextual information of images.In addition,multiple loss functions are used,including weighted binary cross-entropy loss,weighted intersection-over-union loss,and an intelligent loss network that does not require training,to improve the stability of the model training and further improve the accuracy of model segmentation.The proposed model is trained and tested on a densely annotated rice defect dataset,and achieves better segmentation performance than other methods,with robustness and good generalization ability.

Key words: Rice quality assessment, Semantic segmentation, Deep learning, Convolutional neural network, Transformer

中图分类号: 

  • TP181
[1]KIM M J,LIM J,KWON S W,et al.Geographical origin dis-crimination of white rice based on image pixel size using hyperspectral fluorescence imaging analysis[J].Applied Sciences,2020,10(17):5794.
[2]DING C,LIU Q,LI P,et al.Distribution and quantitative analysis of phenolic compounds in fractions of Japonica andIndica rice[J].Food Chemistry,2019,274:384-391.
[3]LIN Z,ZHANG X,WANG Z,et al.Metabolomic analysis ofpathways related to rice grain chalkiness by a notched-belly mutant with high occurrence of white-belly grains[J].BMC Plant Biology,2017,17:1-15.
[4]HARALICK R M,STERNBERG S R,ZHUANG X.Imageanalysis using mathematical morphology[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1987(4):532-550.
[5]WAN Y N,LIN C M,CHIOU J F.Rice quality classification using an automatic grain quality inspection system[J].Transactions of the ASAE,2002,45(2):379.
[6]PAYMAN S H,BAKHSHIPOUR A,ZAREIFOROUSH H.Development of an expert vision-based system for inspecting rice quality indices[J].Quality Assurance and Safety of Crops & Foods,2018,10(1):103-114.
[7]KAUR H,SINGH B.Classification and grading rice using multi-class SVM[J].International Journal of Scientific and Research Publications,2013,3(4):1-5.
[8]ZAREIFOROUSH H,MINAEI S,ALIZADEH M R,et al.A hybrid intelligent approach based on computer vision and fuzzy logic for quality measurementof milled rice[J].Measurement,2015,66:26-34.
[9]CHEN S,XIONG J,GUO W,et al.Colored rice quality inspection system using machine vision[J].Journal of Cereal Science,2019,88:87-95.
[10]RONNEBERGERO,FISCHER P,BROX T.U-net:Convolu-tional networks for biomedical image segmentation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015:18th International Conference,Munich,Germany,October 5-9,2015,Proceedings,Part III 18.Springer International Publishing,2015:234-241.
[11]ÇIÇEK Ö,ABDULKADIR A,IENKAMP S S,et al.3D U-Net:learning dense volumetric segmentation from sparse annotation[C]//Medical Image Computing and Computer-Assisted Intervention-MICCAI 2016:19th International Conference,Athens,Greece,October 17-21,2016,Proceedings,Part II 19.Springer International Publishing,2016:424-432.
[12]XIAO X,LIAN S,LUO Z,et al.Weighted res-unet for high-quality retina vessel segmentation[C]//2018 9th International Conference on Information Technology in Medicine and Dducation(ITME).IEEE,2018:327-331.
[13]ZHOU Z,RAHMAN SIDDIQUEE M M,TAJBAKHSH N,et al.Unet++:A nested u-net architecture for medical image segmentation[C]//Deep Learning in Medical Image Analysis and Multi modal Learning for Clinical Decision Support:4th International Workshop,DLMIA 2018,and 8th International Workshop,ML-CDS 2018,Held in Conjunction with MICCAI 2018,Granada,Spain,September 20,2018,Proceedings 4.Springer International Publishing,2018:3-11.
[14]HUANG H,LIN L,TONG R,et al.Unet 3+:A full-scale connected unet for medical image segmentation[C]//ICASSP 2020-2020 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).IEEE,2020:1055-1059.
[15]CHEN J,LU Y,YU Q,et al.Transunet:Transformers make strong encoders for medical image segmentation[J].arXiv:2102.04306,2021.
[16]CHEN L C,PAPANDREOU G,KOKKINOS I,et al.Deeplab:Semantic image segmentation with deep convolutional nets,atrous convolution,and fully connected crfs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,40(4):834-848.
[17]SCHLEMPER J,OKTAY O,SCHAAP M,et al.Attention gated networks:Learning to leverage salient regions in medical images[J].Medical Image Analysis,2019,53:197-207.
[18]ZHAO H,SHI J,QI X,et al.Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.2017:2881-2890.
[19]VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[J].arXiv:1706.03762,2017.
[20]CARION N,MASSA F,SYNNAEVE G,et al.End-to-end object detection with transformers[C]//Computer Vision-ECCV 2020:16th European Conference,Glasgow,UK,August 23-28,2020,Proceedings,Part I 16.Springer International Publishing,2020:213-229.
[21]DOSOVITSKIY A,BEYER L,KOLESNIKOV A,et al.An image is worth 16x16 words:Transformers for image recognition at scale[J].arXiv:2010.11929,2020.
[22]TOUVRON H,CORD M,DOUZE M,et al.Training data-effi-cient image transformers & distillation through attention[C]//International Conference on Machine Learning.PMLR,2021:10347-10357.
[23]LIU Z,LIN Y,CAO Y,et al.Swin transformer:Hierarchical vision transformer using shifted windows[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:10012-10022.
[24]VALANARASU J M J,OZA P,HACIHALILOGLU I,et al.Medical transformer:Gated axial-attention for medical image segmentation[C]//Medical Image Computing and Computer Assisted Intervention-MICCAI 2021:24th International Confe-rence,Strasbourg,France,September 27-October 1,2021,Proceedings,Part I 24.Springer International Publishing,2021:36-46.
[25]CAO H,WANG Y,CHEN J,et al.Swin-unet:Unet-like puretransformer for medical image segmentation[C]//Computer Vision-ECCV 2022 Workshops:Tel Aviv,Israel,October 23-27,2022,Proceedings,Part III.Cham:Springer Nature Switzerland,2023:205-218.
[26]JHA D,RIEGLER M A,JOHANSEN D,et al.Doubleu-net:A deep convolutional neural network for medical image segmentation[C]//2020 IEEE 33rd International Symposium on Compu-ter-based Medical Systems(CBMS).IEEE,2020:558-564.
[27]SIMONYAN K,ZISSERMAN A.Very deep convolutional networks for large-scale image recognition[J].arXiv:1409.1556,2014.
[28]DENG J,DONG W,SOCHER R,et al.Imagenet:A large-scale hierarchical image database[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2009:248-255.
[29]CHEN L C,PAPANDREOU G,SCHROFF F,et al.Rethinkingatrous convolution for semantic image segmentation[J].arXiv:1706.05587,2017.
[30]WEI J,WANG S,HUANG Q.F3Net:fusion,feedback and focus for salient object detection[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2020:12321-12328.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发