Please wait a minute...
Journal of Integrative Agriculture  2023, Vol. 22 Issue (6): 1671-1683    DOI: 10.1016/j.jia.2022.09.021
Special Issue: 玉米耕作栽培Maize Physiology · Biochemistry · Cultivation · Tillage
Crop Science Advanced Online Publication | Current Issue | Archive | Adv Search |
Detection of maize tassels for UAV remote sensing image with an improved YOLOX Model
SONG Chao-yu1, ZHANG Fan1, LI Jian-sheng2, XIE Jin-yi1, YANG Chen1, ZHOU Hang1, ZHANG Jun-xiong1#
1 College of Engineering, China Agricultural University, Beijing 100083, P.R.China

2 College of Agronomy and Biotechnology, China Agricultural University, Beijing 100083, P.R.China

Download:  PDF in ScienceDirect  
Export:  BibTeX | EndNote (RIS)      
摘要  玉米雄穗检测是玉米种植和育种农艺管理中必不可少的技术,可应用于产量估算、生长监测、智能采摘、病害检测等方面。然而,田间的玉米雄穗普遍存在遮挡现象,不同生长阶段的雄穗大小和形态颜色也不尽相同。针对这些问题,本文提出了SEYOLOX-tiny模型,可以更精准、更鲁棒地识别田间的玉米雄穗。通过无人机构建了丰富的玉米雄穗图像数据集,在保证图像质量和图像采集效率的同时兼顾不同时期的玉米雄穗的图像多样性。另外,YOLOX嵌入注意力机制,在关键特征的提取时,能有效抑制不利因素(遮挡、重叠)的噪声,有助于应对农田多变复杂的环境。实验结果显示,改进的识别算法SEYOLOX-tiny平均检测精度达到95.0%;相较于原始模型的mAP@0.5、mAP@0.5-0.95、mAP@0.5-0.95(面积=小)和mAP@0.5-0.95(面积=中)提升1.5、1.8、5.3和1.7个百分点。因此,本文提出的方法可以满足玉米穗检测视觉系统中所需要的精度和鲁棒性。

Abstract  

Maize tassel detection is essential for future agronomic management in maize planting and breeding, with application in yield estimation, growth monitoring, intelligent picking, and disease detection.  However, detecting maize tassels in the field poses prominent challenges as they are often obscured by widespread occlusions and differ in size and morphological color at different growth stages.  This study proposes the SEYOLOX-tiny Model that more accurately and robustly detects maize tassels in the field.  Firstly, the data acquisition method ensures the balance between the image quality and image acquisition efficiency and obtains maize tassel images from different periods to enrich the dataset by unmanned aerial vehicle (UAV).  Moreover, the robust detection network extends YOLOX by embedding an attention mechanism to realize the extraction of critical features and suppressing the noise caused by adverse factors (e.g., occlusions and overlaps), which could be more suitable and robust for operation in complex natural environments.  Experimental results verify the research hypothesis and show a mean average precision (mAP@0.5) of 95.0%.  The mAP@0.5, mAP@0.5–0.95, mAP@0.5–0.95 (area=small), and mAP@0.5–0.95 (area=medium) average values increased by 1.5, 1.8, 5.3, and 1.7%, respectively, compared to the original model.  The proposed method can effectively meet the precision and robustness requirements of the vision system in maize tassel detection.

Keywords:  maize       tassel detection       remote sensing       deep learning       attention mechanism  
Received: 11 April 2022   Accepted: 26 August 2022 Online: 27 September 2022  
Fund: 

This research was supported by the Chinese Universities Scientific Fund (2022TC169). 

About author:  SONG Chao-yu, E-mail: chaoyu@cau.edu.cn; #Correspondence ZHANG Jun-xiong, E-mail: cau2007@cau.edu.cn.

Cite this article: 

SONG Chao-yu, ZHANG Fan, LI Jian-sheng, XIE Jin-yi, YANG Chen, ZHOU Hang, ZHANG Jun-xiong. 2023. Detection of maize tassels for UAV remote sensing image with an improved YOLOX Model. Journal of Integrative Agriculture, 22(6): 1671-1683.

Bendig J, Yu K, Aasen H, Bolten A, Bennertz S, Broscheit J, Gnyp M L, Bareth G. 2015. Combining UAV-based plant height from crop surface models, visible, and near infrared vegetation indices for biomass monitoring in barley. International Journal of Applied Earth Observation and Geoinformation39, 79–87.

Camargo A, Smith J S. 2009. An image-processing based algorithm to automatically identify plant disease visual symptoms. Biosystems Engineering102, 9–21.

Chang L, He S, Huang D, Liu Q, Xiang J. 2018. Quantifying muskmelon fruit attributes with A-TEP-based model and machine vision measurement. Journal of Integrative Agriculture17, 1369–1379.

Gage J L, Miller N D, Spalding E P, Kaeppler S M, de Leon N. 2017. TIPS: A system for automated image-based phenotyping of maize tassels. Plant Methods13, 21.

Ge Z, Liu S, Li Z, Yoshie O, Sun J. 2021. OTA: Optimal transport assignment for object detection. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nashville, TN, USA. pp. 303–312.

Ge Z, Liu S, Wang F, Li Z, Sun J. 2021. YOLOX: Exceeding YOLO series in 2021. arXiv Preprint arXiv:2107.08430.

Hu J, Shen L, Albanie S, Sun G, Wu E. 2020. Squeeze-and-excitation networks. IEEE Transactions on Pattern Analysis and Machine Intelligence42, 2011–2023.

Jannoura R, Brinkmann K, Uteau D, Bruns C, Joergensen R G. 2015. Monitoring of crop biomass using true colour aerial photographs taken from a remote controlled hexacopter. Biosystems Engineering129, 341–351.

Ji M, Yang Y, Zheng Y, Zhu Q, Huang M, Guo Y. 2021. In-field automatic detection of maize tassels using computer vision. Information Processing in Agriculture8, 87–95.

Jin G H, Hyeonjoon M, Jin T K, Syed I H, Minh D, O N L, Han Y P. 2017. Deep convolutional neural network for classifying Fusarium wilt of radish from unmanned aerial vehicles. Journal of Applied Remote Sensing11, 1–14.

Karami A, Quijano K, Crawford M. 2021. Advancing tassel detection and counting: Annotation and algorithms. Remote Sensing13, 2881.

Koh J C O, Hayden M, Daetwyler H, Kant S. 2019. Estimation of crop plant density at early mixed growth stages using UAV imagery. Plant Methods15, 64.

Kurtulmuş F, Kavdir 0. 2014. Detecting corn tassels using computer vision and support vector machines. Expert Systems with Applications41, 7390–7397.

Li L, Sun Q, Tu K L, Wang J H, Yang L M. 2018. Selection for high quality pepper seeds by machine vision and classifiers. Journal of Integrative Agriculture17, 1999–2006.

Lin T, Goyal P, Girshick R, He K, Dollar P. 2020. Focal loss for dense object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence42, 318–327.

Liu Y, Cen C, Che Y, Ke R, Ma Y, Ma Y. 2020. Detection of maize tassels from UAV RGB imagery with faster R-CNN. Remote Sensing12, 338.

Lu H, Cao Z, Xiao Y, Fang Z, Zhu Y, Xian K. 2015. Fine-grained maize tassel trait characterization with multi-view representations. Computers and Electronics in Agriculture118, 143–158.

Osco L P, Marcato J, Ramos A P M, Jorge L A D, Fatholahi S N, Silva J D, Matsubara E T, Pistori H, Gonçalves W N, Li J. 2021. A review on deep learning in UAV remote sensing. International Journal of Applied Earth Observation and Geoinformation102, 102456.

Pan S, Qiao J, Wang R, Yu H, Wang C, Taylor K, Pan H. 2022. Intelligent diagnosis of northern corn leaf blight with deep learning model. Journal of Integrative Agriculture21, 1094–1105.

Ranum P, Peña-Rosas J P, Garcia-Casal M N. 2014. Global maize production, utilization, and consumption. Annals of the New York Academy of Sciences1312, 105–112.

Ren S, He K, Girshick R, Sun A J. 2017. Faster R-CNN: Towards Real-Time object detection with region proposal networks. Ieee Transactions on Pattern Analysis and Machine Intelligence39, 1137–1149.

Su Y, Wu F, Ao Z, Jin S, Qin F, Liu B, Pang S, Liu L, Guo Q. 2019. Evaluating maize phenotype dynamics under drought stress using terrestrial lidar. Plant Methods15, 11.

Wang Q, Wu B, Zhu P, Li P, Zuo W, Hu Q. 2020. ECA-Net: Efficient channel attention for deep convolutional neural networks. In: 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Long Beach, CA, USA. pp. 11531–11539.

Woo S, Park J, Lee J, Kweon I S. 2018. CBAM: Convolutional block attention module. In: The 15th European Conference on Computer Vision (ECCV). Springer, Cham, Munich, Germany. pp. 3–19.

Yang G, Yang Y, He Z, Zhang X, He Y. 2022. A rapid, low-cost deep learning system to classify strawberry disease based on cloud service. Journal of Integrative Agriculture21, 460–473.

Yang S, Liu J, Xu K, Sang X, Ning J, Zhang Z. 2021. Improved CenterNet based maize tassel recognition for UAV remote sensing image. Transactions of the Chinese Society for Agricultural Machinery52, 206–212.

Ye M, Cao Z, Yu Z. 2013. An image-based approach for automatic detecting tasseling stage of maize using spatio-temporal saliency. In: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications. International Society for Optics and Photonics. SPIE, Washington, DC, USA. p. 89210Z.

Zan X, Zhang X, Xing Z, Liu W, Zhang X, Su W, Liu Z, Zhao Y, Li S. 2020. Automatic detection of maize tassels from UAV images by combining random forest classifier and VGG16. Remote Sensing12, 3049.

Zhang C, Kovacs J. 2012. The application of small unmanned aerial systems for precision agriculture: A review. Precision Agriculture13, 693–712.

Zhang H, Wang Y, Dayoub F, Sunderhauf N. 2021. VarifocalNet: An IoU-aware dense object detector. In: 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, Nashville, TN, USA. pp. 8510–8519.

[1] Lichao Zhai, Shijia Song, Lihua Zhang, Jinan Huang, Lihua Lv, Zhiqiang Dong, Yongzeng Cui, Mengjing Zheng, Wanbin Hou, Jingting Zhang, Yanrong Yao, Yanhong Cui, Xiuling Jia. Subsoiling before winter wheat alleviates the kernel position effect of densely grown summer maize by delaying post-silking root–shoot senescence[J]. >Journal of Integrative Agriculture, 2025, 24(9): 3384-3402.
[2] Ling Ai, Ju Qiu, Jiuguang Wang, Mengya Qian, Tingting Liu, Wan Cao, Fangyu Xing, Hameed Gul, Yingyi Zhang, Xiangling Gong, Jing Li, Hong Duan, Qianlin Xiao, Zhizhai Liu. A naturally occurring 31 bp deletion in TEOSINTE BRANCHED1 causes branched ears in maize[J]. >Journal of Integrative Agriculture, 2025, 24(9): 3322-3333.
[3] Zishuai Wang, Wangchang Li, Zhonglin Tang. Enhancing the genomic prediction accuracy of swine agricultural economic traits using an expanded one-hot encoding in CNN models[J]. >Journal of Integrative Agriculture, 2025, 24(9): 3574-3582.
[4] Dan Lü, Jianxin Li, Xuehai Zhang, Ran Zheng, Aoni Zhang, Jingyun Luo, Bo Tong, Hongbing Luo, Jianbing Yan, Min Deng. Genetic analysis of maize crude fat content by multi-locus genome-wide association study[J]. >Journal of Integrative Agriculture, 2025, 24(7): 2475-2491.
[5] Chunxiang Li, Yongfeng Song, Yong Zhu, Mengna Cao, Xiao Han, Jinsheng Fan, Zhichao Lü, Yan Xu, Yu Zhou, Xing Zeng, Lin Zhang, Ling Dong, Dequan Sun, Zhenhua Wang, Hong Di. GWAS analysis reveals candidate genes associated with density tolerance (ear leaf structure) in maize (Zea mays L.)[J]. >Journal of Integrative Agriculture, 2025, 24(6): 2046-2062.
[6] Lihua Xie, Lingling Li, Junhong Xie, Jinbin Wang, Zechariah Effah, Setor Kwami Fudjoe, Muhammad Zahid Mumtaz. A suitable organic fertilizer substitution ratio stabilizes rainfed maize yields and reduces gaseous nitrogen loss in the Loess Plateau, China[J]. >Journal of Integrative Agriculture, 2025, 24(6): 2138-2154.
[7] Huairen Zhang, Tauseef Taj Kiani, Huabang Chen, Juan Liu, Xunji Chen. Genome wide association analysis reveals multiple QTLs controlling root development in maize [J]. >Journal of Integrative Agriculture, 2025, 24(5): 1656-1670.
[8] Lanjie Zheng, Qianlong Zhang, Huiying Liu, Xiaoqing Wang, Xiangge Zhang, Zhiwei Hu, Shi Li, Li Ji, Manchun Ji, Yong Gu, Jiaheng Yang, Yong Shi, Yubi Huang, Xu Zheng. Fine mapping and discovery of MIR172e, a candidate gene required for inflorescence development and lower floret abortion in maize ear[J]. >Journal of Integrative Agriculture, 2025, 24(4): 1372-1389.
[9] Xiaoxia Guo, Wanmao Liu, Yunshan Yang, Guangzhou Liu, Bo Ming, Ruizhi Xie, Keru Wang, Shaokun Li, Peng Hou. Matching the light and nitrogen distributions in the maize canopy to achieve high yield and high radiation use efficiency[J]. >Journal of Integrative Agriculture, 2025, 24(4): 1424-1435.
[10] Yang Wang, Chunhua Mu, Xiangdong Li, Canxing Duan, Jianjun Wang, Xin Lu, Wangshu Li, Zhennan Xu, Shufeng Sun, Ao Zhang, Zhiqiang Zhou, Shenghui Wen, Zhuanfang Hao, Jienan Han, Jianzhou Qu, Wanli Du, Fenghai Li, Jianfeng Weng. A genome-wide association study and transcriptome analysis reveal the genetic basis for the Southern corn rust resistance in maize[J]. >Journal of Integrative Agriculture, 2025, 24(2): 453-466.
[11] Yulong Wang, Aizhong Yu, Pengfei Wang, Yongpan Shang, Feng Wang, Hanqiang Lü, Xiaoneng Pang, Yue Li, Yalong Liu, Bo Yin, Dongling Zhang, Jianzhe Huo, Keqiang Jiang, Qiang Chai. No-tillage with total green manure mulching increases maize yield through improved soil moisture and temperature environment and enhanced maize root structure and photosynthetic capacity[J]. >Journal of Integrative Agriculture, 2025, 24(11): 4211-4224.
[12] Yujiao Wei, Yiyun Chen, Jiaxue Wang, Peiheng Yu, Lu Xu, Chi Zhang, Huanfeng Shen, Yaolin Liu, Ganlin Zhang. Mapping soil organic carbon in fragmented agricultural landscapes: The efficacy and interpretability of multi-category remote sensing variables[J]. >Journal of Integrative Agriculture, 2025, 24(11): 4395-4414.
[13] Hong Ren, Zheng Liu, Xinbing Wang, Wenbin Zhou, Baoyuan Zhou, Ming Zhao, Congfeng Li. Long-term excessive nitrogen application decreases spring maize nitrogen use efficiency via suppressing root physiological characteristics[J]. >Journal of Integrative Agriculture, 2025, 24(11): 4195-4210.
[14] Fei Bao, Ping Zhang, Qiying Yu, Yunfei Cai, Bin Chen, Heping Tan, Hailiang Han, Junfeng Hou, Fucheng Zhao. Response of fresh maize yield to nitrogen application rates and  characteristics of nitrogen-efficient varieties[J]. >Journal of Integrative Agriculture, 2025, 24(10): 3803-3818.
[15] Tianqi Wang, Jihui Tian, Xing Lu, Chang Liu, Junhua Ao, Huafu Mai, Jinglin Tan, Bingbing Zhang, Cuiyue Liang, Jiang Tian. Soybean variety influences the advantages of nutrient uptake and yield in soybean/maize intercropping via regulating root-root interaction and rhizobacterial composition[J]. >Journal of Integrative Agriculture, 2025, 24(10): 4048-4062.
No Suggested Reading articles found!