Abstract
Precise segmentation of wheat spikes from a complex background is necessary for obtaining image-based phenotypic information of wheat traits such as yield estimation and disease evaluation. A new instance segmentation method based on a Hybrid Task Cascade model was trained and validated to improve previous attempts of wheat spike detection. In this study, wheat images were collected from fields where the environment varied both spatially and temporally. Res2Net50 was adopted as a backbone network, combined with multi-scale training, deformable convolutional networks, and Generic ROI Extractor for rich feature learning. The proposed methods were trained and validated, and the average precision (AP) obtained for the bounding box and mask was 0.904 and 0.907, respectively, and the accuracy for wheat spike counting was 99.29%. Comprehensive empirical analyses revealed that our method (Wheat-Net) performed well on challenging field-based datasets with mixed qualities, particularly those with various backgrounds and wheat spike adjacence/occlusion. These results provide evidence for dense wheat spike detection capabilities using the latest date deep learning algorithms, which can be useful for improving wheat breeding and disease screening efforts.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Ni C, Wang D, Vinson R, Holmes M, Tao Y (2019) Automatic inspection machine for maize kernels based on deep convo-lutional neural networks. Biosys Eng 178:131–144. https://doi.org/10.1016/j.biosystemseng.2018.11.010
Da Silva LA, Bressan PO, Gonçalves DN, Freitas DM, Machado BB, Gonçalves Wesley N (2019) Estimating soybean leaf defoliation using convolutional neural networks and synthetic images. Comput Electron Agric 156:360–368. https://doi.org/10.1016/j.compag.2018.11.040
Desai SV, Balasubramanian VN, Fukatsu T, Ninomiya S, Guo W (2019) Automatic estimation of heading date of paddy rice using deep learning. Plant Methods 15(1):76. https://doi.org/10.1186/s13007-019-0457-1
Senthilkumar T, Jayas DS, White NDG, Fields PG, Gräfenhan T (2017) Detection of ochratoxin A contamination in stored wheat using nearinfrared hyperspectral imaging. Infrared Phys Techn 81:228–235. https://doi.org/10.1016/j.infred.2017.01.015
Aich S, Josuttes A, Ovsyannikov I, Strueby K, Ahmed I, Duddu HS, Pozniak C, Shirtliffe S, Stavness I (2018) DeepWheat: estimating phenotypic traits from crop images with deep learning. In: 2018 18th IEEE winter conference on applications of computer vision (WACV), pp 323–332. https://doi.org/10.1109/wacv.2018.00042
Momeny M, Jahanbakhshi A, Jafarnezhad K, Zhang YD (2020) Accurate classification of cherry fruit using deep cnn based on hybrid pooling approach. Postharvest Biol Technol 166:111204. https://doi.org/10.1016/j.postharvbio.2020.111204
Suresh G, Gnanaprakash V, Santhiya R (2019) Performance analysis of different CNN architecture with different optimisers for plant disease classification. In: 2019 5th international conference on advanced computing and communication systems (ICACCS), Coimbatore, India. 916–921. https://doi.org/10.1109/ICACCS.2019.8728282
Khaki S, Wang L, Archontoulis SV (2020) A cnn-rnn framework for crop yield prediction. Front Plant Sci. https://doi.org/10.3389/fpls.2019.01750
Sun J, Di L, Sun Z, Shen Y, Lai Z (2019) County-level soybean yield prediction using deep cnn-lstm model. Sensors 19(20):4363. https://doi.org/10.3390/s19204363
Yang Q, Shi L, Lin L (2019) Plot-scale rice grain yield estimation using UAV-based remotely sensed images via CNN with time-invariant deep features decomposition. In: IGARSS 2019–2019 IEEE international geoscience and remote sensing symposium, Yokohama, Japan, 7180–7183. https://doi.org/10.1109/IGARSS.2019.8898061
Luna RG de, Dadios EP, Bandala AA, Vicerra RRP (2020) J Agric Sci 42:24–36. https://doi.org/10.17503/agrivita.v42i1.2499
Zhuang S, Wang P, Jiang B, Li M (2020) Learned features of leaf phenotype to monitor maize water status in the fields. Comput Electron Agric 172. https://doi.org/10.1016/j.compag.2020.105347
Kovalchuk N, Laga H, Cai J, Kumar P, Parent B, Lu Z et al (2017) Phenotyping of plants in competitive but controlled environments: a study of drought response in transgenic wheat. Funct Plant Biol 44(3):290. https://doi.org/10.1071/FP16202
Misra T, Arora A, Marwaha S, Chinnusamy V, Goel S (2020) Spikesegnet-a deep learning approach utilizing encoder-decoder network with hourglass for spike segmentation and counting in wheat plant from visual imaging. Plant Methods. https://doi.org/10.1186/s13007-020-00582-9
Qiongyan L, Cai J, Berger B, Okamoto M, Miklavcic SJ (2017) Detecting spikes of wheat plants using neural networks with laws texture energy. Plant Methods 13(1):83. https://doi.org/10.1186/s13007-017-0231-1
Pound MP, Atkinson JA, Wells DM, Pridmore TP, French AP (2017) Deep learning for multi-task plant phenotyping. In: 2017 IEEE international conference on computer vision workshops (ICCVW). IEEE
Chandra AL, Desai SV, Balasubramanian VN, Ninomiya S, Guo W (2020) Active learning with point supervision for cost-effective panicle detection in cereal crops. Plant Methods 16(1). https://doi.org/10.1186/s13007-020-00575-8
David E, Madec S, Sadeghi-Tehran P, Aasen H, Zheng B, Liu S et al (2020) Global wheat head detection (gwhd) dataset: a large and diverse dataset of high resolution rgb labelled images to develop and benchmark wheat head detection methods. Plant Phenomics. https://doi.org/10.34133/2020/3521852
Hasan MM, Chopin JP, Laga H, Miklavcic SJ (2018) Detection and analysis of wheat spikes using convolutional neural networks. Plant Methods 14(1):100. https://doi.org/10.1186/s13007-018-0366-8
Sadeghi-Tehran P, Virlet N, Ampe EM, Reyns P, Hawkesford MJ (2019) Deepcount: in-feld automatic quantifcation of wheat spikes using simple linear iterative clustering and deep convolutional neural networks. Front Plant Sci 10:1176. https://doi.org/10.3389/fpls.2019.01176
Xu X, Li H, Yin F, Xi L, Ma X (2020) Wheat ear counting using k-means clustering segmentation and convolutional neural network. Plant Methods 16(1). https://doi.org/10.1186/s13007-020-00648-8
Zhang D, Wang D, Gu C, Jin N, Liang D (2019) Using neural network to identify the severity of wheat fusarium head blight in the field environment. Remote Sens 11(20):2375. https://doi.org/10.1016/10.3390/rs11202375
Alkhudaydi T, Reynolds D, Griffiths S, Zhou J, Iglesia BDL (2019) An exploration of deep-learning based phenotypic analysis to detect spike regions in field conditions for uk bread wheat. Plant Phenomics (1, article no. 45):1–17. https://doi.org/10.34133/2019/7368761
Tan C, Zhang P, Zhang Y, Zhou X, Wang Z, Du Y, Mao W, Li W, Wang D, Guo W (2020) Rapid recognition of field-grown wheat spikes based on a Superpixel segmentation algorithm using digital images. Front Plant Sci 11:259. https://doi.org/10.3389/fpls.2020.00259
Ma J, Li Y, Liu H, Du K, Zhang L (2020) Improving segmentation accuracy for ears of winter wheat at flowering stage by semantic segmentation. Comput Electron Agric 176:105662.https://doi.org/10.1016/j.compag.2020.105662
Chen K, Pang J, Wang J, Xiong Y, Li X, Sun S et al (2019) Hybrid task cascade for instance segmentation. https://doi.org/10.1109/CVPR.2019.00511
Li X, Liu Z, Luo P, Loy CC, Tang X (2017) Not all pixels are equal: difficulty-aware semantic segmentation via deep layer cascade.https://doi.org/10.1109/CVPR.2017.684
Qiu R, Yang C, Moghimi A, Zhang M, Steffenson BJ, Hirsch CD (2019) Detection of fusarium head blight in wheat using a deep neural network and color imaging. Remote Sens 11:2658. https://doi.org/10.20944/preprints201910.0056.v1
Cai Z, Vasconcelos N (2019) Cascade r-cnn: high quality object detection and instance segmentation. IEEE Trans Pattern Anal Mach Intell. https://doi.org/10.1109/TPAMI.2019.2956516
Kaiming H, Georgia G, Piotr D, Ross G (2017) Mask r-cnn. In: IEEE international conference on computer vision. https://doi.org/10.1109/TPAMI.2018.2844175
Lin TY, Maire M, Belongie S, Hays J, Perona P, Ramanan D, Dollár P, Zitnick CL (2014) Microsoft COCO: common objects in context. In: Fleet D, Pajdla T, Schiele B, Tuytelaars T (eds) Computer vision–ECCV 2014. Springer International Publishing, Cham, Switzerland, pp 740–755
Russell BC, Torralba A, Murphy KP, Freeman WT (2008) Labelme: a database and web-based tool for image annotation. Int J Comput Vis 77(1–3). https://doi.org/10.1007/s11263-007-0090-8
Gao S, Cheng MM, Zhao K, Zhang XY, Torr PHS (2019) Res2net: a new multi-scale backbone architecture. IEEE Trans Pattern Anal Mach Intell PP(99):1–1. https://doi.org/10.1109/TPAMI.2019.2938758
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778. https://arxiv.org/abs/1512.03385
Xie S, Girshick R, Dollár P, Tu Z, He K (2017) Aggregated Residual Transformations for Deep Neural Networks. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. arXiv:1611.05431
Dai J, Qi H, Xiong Y, Li Y, Zhang G, Hu H, Wei Y (2017) Deformable convolutional networks. https://doi.org/10.1109/ICCV.2017.89
Ren S, He K, Girshick R, Sun J (2015) Faster r-cnn: towards real-time object detection with region proposal networks. IEEE Trans Pattern Anal Mach Intell 39(6). https://doi.org/10.1109/TPAMI.2016.2577031
Rossi L, Karimi A, Prati A (2020) A novel region of interest extraction layer for instance segmentation. arXiv:2004.13665
Pont-Tuset J, Pablo A, Barron JT, Marques F, Malik J (2016) Multiscale combinatorial grouping for image segmentation and object proposal generation. IEEE Trans Pattern Anal Mach Intell 39(1). https://doi.org/10.1186/10.1109/TPAMI.2016.2537320
Caruana R (1998) Multitask learnin. https://doi.org/10.1007/978-1-4615-5529-2_5
Rosenfield A, Thurston M, Lee YH (1971) Edge and curve detection for visual scene analysis. IEEE Trans Comput C 20(5):562–569. https://doi.org/10.1109/T-C.1971.223290
Bodla N, Singh B, Chellappa R, Davis LS (2017) Soft-nms—improving object detection with one line of code. arXiv:1704.04503
He K, Zhang X, Ren S, Sun J (2014) Spatial pyramid pooling in deep convolutional networks for visual recognition. https://doi.org/10.1007/978-3-319-10578-9_23
Deng J, Dong W, Socher R, Li LJ, Li FF (2009) ImageNet: a large-scale hierarchical image database. In: IEEE conference on computer vision & pattern recognition. https://doi.org/10.1109/CVPR.2009.5206848
Su WH, Zhang J, Yang C, Page R, Steffenson BJ (2021) Automatic evaluation of wheat resistance to fusarium head blight using dual mask-rcnn deep learning frameworks in computer vision. Remote Sens 13(1):1–21. https://doi.org/10.3390/rs13010026
Acknowledgements
This work was supported by the USDA-ARS United States Wheat and Barley Scab Initiative (grant numbers 58-5062-8-018), the Lieberman-Okinow Endowment at the University of Minnesota, and the State of Minnesota Small Grains Initiative.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this chapter
Cite this chapter
Zhang, J. et al. (2022). Wheat-Net: An Automatic Dense Wheat Spike Segmentation Method Based on an Optimized Hybrid Task Cascade Model. In: Zhang, Z., Liu, H., Yang, C., Ampatzidis, Y., Zhou, J., Jiang, Y. (eds) Unmanned Aerial Systems in Precision Agriculture. Smart Agriculture, vol 2. Springer, Singapore. https://doi.org/10.1007/978-981-19-2027-1_6
Download citation
DOI: https://doi.org/10.1007/978-981-19-2027-1_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-2026-4
Online ISBN: 978-981-19-2027-1
eBook Packages: Biomedical and Life SciencesBiomedical and Life Sciences (R0)