Wheat ear counting is a prerequisite for the evaluation of wheat yield.A wheat ear counting method based on frequency domain decomposition is proposed in this study to improve the accuracy of wheat yield estimation.Th...Wheat ear counting is a prerequisite for the evaluation of wheat yield.A wheat ear counting method based on frequency domain decomposition is proposed in this study to improve the accuracy of wheat yield estimation.The frequency domain decomposition of wheat ear image is completed by multiscale support value filter(MSVF)combined with improved sampled contourlet transform(ISCT).Support Vector Machine(SVM)is the classic classification and regression algorithm of machine learning.MSVF based on this has strong frequency domain filtering and generalization ability,which can effectively remove the complex background,while the multi-direction characteristics of ISCT enable it to represent the contour and texture information of wheat ears.In order to improve the level of wheat yield prediction,MSVF-ISCT method is used to decompose the ear image in multiscale and multi direction in frequency domain,reduce the interference of irrelevant information,and generate the sub-band image with more abundant information components of ear feature information.Then,the ear feature is extracted by morphological operation and maximum entropy threshold segmentation,and the skeleton thinning and corner detection algorithms are used to count the results.The number of wheat ears in the image can be accurately counted.Experiments show that compared with the traditional algorithms based on spatial domain,this method significantly improves the accuracy of wheat ear counting,which can provide guidance and application for the field of agricultural precision yield estimation.展开更多
To understand the contribution of ear photosynthesis to grain yield and its response to water supply in the improvement of winter wheat, 15 cultivars released from 1980 to 2012 in North China Plain(NCP) were planted...To understand the contribution of ear photosynthesis to grain yield and its response to water supply in the improvement of winter wheat, 15 cultivars released from 1980 to 2012 in North China Plain(NCP) were planted under rainfed and irrigated conditions from 2011 to 2013, and the ear photosynthesis was tested by ear shading. During the past 30 years, grain yield significantly increased, the flag leaf area slightly increased under irrigated condition but decreased significantly under rainfed condition, the ratio of grain weight:leaf area significantly increased, and the contribution of ear photosynthesis to grain yield changed from 33.6 to 64.5% and from 32.2 to 57.2% under rainfed and irrigated conditions, respectively. Grain yield, yield components, and ratio of grain weight:leaf area were positively related with contribution of ear photosynthesis. The increase in grain yield in winter wheat was related with improvement in ear photosynthesis contribution in NCP, especially under rainfed condition.展开更多
In wheat, the ear is one of the main photosynthetic contributors to grain filling under drought stress conditions. In order to determine the relationship between stomatal characteristics and plant drought resistance, ...In wheat, the ear is one of the main photosynthetic contributors to grain filling under drought stress conditions. In order to determine the relationship between stomatal characteristics and plant drought resistance, photosynthetic and stomatal characteristics and water use efficiency(WUE) were studied in two wheat cultivars: the drought-resistant cultivar ‘Changhan 58' and the drought-sensitive cultivar ‘Xinong 9871'. Plants of both cultivars were grown in pot conditions under well-watered(WW) and water-stressed(WS) conditions. In both water regimes,‘Changhan 58' showed a significantly higher ear photosynthetic rate with a lower rate of variation and a significantly higher percentage variation of transpiration compared to control plants at the heading stage under WS conditions than did ‘Xinong 9871' plants. Moreover,‘Changhan 58' showed lower stomatal density(SD) and higher stomatal area per unit organ area(A) under both water conditions. Water stress decreased SD, A, and stomatal width(SW), and increased stomatal length in flag leaves(upper and lower surfaces) and ear organs(awn, glume,lemma, and palea), with the changes more pronounced in ear organs than in flag leaves.Instantaneous WUE increased slightly, while integral WUE improved significantly in both cultivars. Integral WUE was higher in ‘Changhan 58', and increased by a greater amount, than in‘Xinong 9871'. These results suggest that drought resistance in ‘Changhan 58' is regulated by stomatal characteristics through a decrease in transpiration rate in order to improve integral WUE and photosynthetic performance, and through sustaining a higher ear photosynthetic rate, therefore enhancing overall drought-resistance.展开更多
The activities of RuBPC and C4 photosynthetic enzymes in ear and flag leaf blade were examined in wheat. The results showed that photosynthesis of ear was less sensitive to soil drought than that of flag leaf, and dec...The activities of RuBPC and C4 photosynthetic enzymes in ear and flag leaf blade were examined in wheat. The results showed that photosynthesis of ear was less sensitive to soil drought than that of flag leaf, and decrease of CO2 assimilation in flag leaf blade with water stress was more than that in ear. Compared with flag leaf, ear organs (awn, glume and lemma) had higher C4 enzyme activities and lower RuBPC activity. Under moderate water-stress, the increase of C4 enzyme activities was induced, and the increase was higher in ear than in flag leaf. Under severe water-stress, relatively higher C4 enzyme activities were still maintained in ear, rather than that in flag leaf. It suggests that high activities of C4 enzymes in ear may contribute to its high tolerance of photosynthesis to water-stress.展开更多
小麦麦穗的高效计数对快速、准确掌握小麦产量具有重要意义。无人机由于具有效率高、成本低等特点被广泛应用于大田小麦信息的采集。但已有的用于小麦麦穗计数的深度学习模型结构复杂、参数量大,不能直接部署在存储空间有限的无人机的...小麦麦穗的高效计数对快速、准确掌握小麦产量具有重要意义。无人机由于具有效率高、成本低等特点被广泛应用于大田小麦信息的采集。但已有的用于小麦麦穗计数的深度学习模型结构复杂、参数量大,不能直接部署在存储空间有限的无人机的边缘设备上。针对这一问题,提出了一种融合剪枝策略和知识蒸馏的模型压缩方法,基于YOLOv5s模型构建了一种轻量化模型,并设计了面向无人机边缘计算的小麦麦穗计数轻量化方案。试验结果表明,经过模型剪枝和知识蒸馏轻量化处理的YOLOv5s模型,在小麦计数任务上的计数准确率为93.3%,模型的mAP(mean Average Precision,平均精度均值)达到94.4%,模型大小缩小了约76%,模型参数量减少了79.61%。因此,模型在保持较高的计数准确率的同时将会占用更少的计算资源和存储空间,显著的压缩效果使模型可以部署在无人机的边缘设备上,为小麦麦穗的实时计数提供了可能。展开更多
在无人机上安装光学传感器捕捉农作物图像是一种经济高效的方法,它有助于产量预测、田间管理等。该研究以无人机小麦作物图像为研究对象,针对图像中麦穗分布稠密、重叠现象严重、背景信息复杂等特点,设计了一种基于TPH-YOLO(YOLO with t...在无人机上安装光学传感器捕捉农作物图像是一种经济高效的方法,它有助于产量预测、田间管理等。该研究以无人机小麦作物图像为研究对象,针对图像中麦穗分布稠密、重叠现象严重、背景信息复杂等特点,设计了一种基于TPH-YOLO(YOLO with transformer prediction heads)的麦穗检测模型,提高无人机图像麦穗计数的精度。首先,为了减小光照不均匀对无人机图像质量造成的影响,该研究采用Retinex算法进行图像增强处理。其次,在YOLOv5的骨干网络中添加坐标注意力机制(coordinate attention,CA),使模型细化特征,更加关注麦穗信息,抑制麦秆、麦叶等一些背景因素的干扰。再次,将YOLOv5中原始的预测头转换为Transformer预测头(transformer prediction heads,TPH),该预测头具有多头注意力机制的预测潜力,可以在高密度场景下准确定位到麦穗。最后,为了提高模型的泛化能力和检测精度,采用了迁移学习的训练策略,先使用田间采集的小麦图像数据集对模型进行预训练,接着再使用无人机采集的小麦图像数据集对模型进行参数更新和优化训练,并在无人机采集的小麦图像数据集上进行了试验。结果表明,该研究方法精确率、召回率及平均精确率分别为87.2%、84.1%和88.8%,相较于基础的YOLOv5平均精确率提高4.1个百分点,性能优于SSD、Faster-RCNN、CenterNet、YOLOv5等目标检测模型。此外,该研究利用公开数据集Global Wheat Head Detection(GWHD)在不同目标检测模型上进行对比试验,该数据集的小麦样本是多样的和典型的,与SSD、Faster-RCNN、CenterNet和YOLOv5等模型相比,平均精确率分别提升11.1、5.4、6.9和3.3个百分点,进一步验证了该研究所提方法的可靠性和有效性,研究结果可以为小麦的产量预测提供支撑。展开更多
小麦的最终产量可由单位面积的小麦麦穗数侧面反映,为了快速准确统计小麦麦穗数,该研究给出一种在单幅图像上利用深度卷积神经网络估计田间麦穗密度图并进行麦穗计数的方法。首先对采集的田间小麦图像进行直方图均衡化及阈值分割预处理...小麦的最终产量可由单位面积的小麦麦穗数侧面反映,为了快速准确统计小麦麦穗数,该研究给出一种在单幅图像上利用深度卷积神经网络估计田间麦穗密度图并进行麦穗计数的方法。首先对采集的田间小麦图像进行直方图均衡化及阈值分割预处理,以减少图像中光照及一些复杂背景对计数的影响;然后根据灌浆期田间小麦图像麦穗密集的特点,引入拥挤场景识别网络(Congested Scene Recognition Network,CSRNet)构建麦穗密度图估计模型,并采用迁移学习方法,利用小麦图像公开数据集对模型进行预训练,再用所采集的小麦图像数据集进行模型参数调整和优化;利用得到的模型生成单幅小麦图像的麦穗密度图,根据密度图中所有密度值的总和对图像进行麦穗计数。最后根据对单幅麦穗图像的试验数据,构建田间麦穗计数函数模型,实现田间小麦麦穗数估计。通过对所采集的安农170、苏麦188、乐麦608和宁麦24这4个品种共296幅小麦图像进行试验,平均绝对误差(Mean Absolute Error,MAE)和均方根误差(Root Mean Squared Error,RMSE)分别为16.44和17.89,4个品种小麦的麦穗计数值与真实值的决定系数R^2均在0.9左右,表明该方法对单幅图像小麦麦穗计数精度较高。此外,通过对田间小麦麦穗数进行估计试验,结果表明,随面积的增大麦穗估计的误差越小,研究结果可以为小麦的产量自动估计提供参考。展开更多
基金National Natural Science Foundation of China(61672032)National Key Research and Development Program of China(2016YFD0800904)+1 种基金Anhui Provincial Science and Technology Project(16030701091)The Open Research Fund of National Engineering Research Center for Agro-Ecological Big Data Analysis&Application,Anhui University(AE2018009).
文摘Wheat ear counting is a prerequisite for the evaluation of wheat yield.A wheat ear counting method based on frequency domain decomposition is proposed in this study to improve the accuracy of wheat yield estimation.The frequency domain decomposition of wheat ear image is completed by multiscale support value filter(MSVF)combined with improved sampled contourlet transform(ISCT).Support Vector Machine(SVM)is the classic classification and regression algorithm of machine learning.MSVF based on this has strong frequency domain filtering and generalization ability,which can effectively remove the complex background,while the multi-direction characteristics of ISCT enable it to represent the contour and texture information of wheat ears.In order to improve the level of wheat yield prediction,MSVF-ISCT method is used to decompose the ear image in multiscale and multi direction in frequency domain,reduce the interference of irrelevant information,and generate the sub-band image with more abundant information components of ear feature information.Then,the ear feature is extracted by morphological operation and maximum entropy threshold segmentation,and the skeleton thinning and corner detection algorithms are used to count the results.The number of wheat ears in the image can be accurately counted.Experiments show that compared with the traditional algorithms based on spatial domain,this method significantly improves the accuracy of wheat ear counting,which can provide guidance and application for the field of agricultural precision yield estimation.
基金supported by the National Natural Science Foundation of China (31401297)the National Key Research and Development Program of China (2016YFD0300105)+1 种基金the Chinese Universities Scientific Fund (2016NX002)the Earmarked Fund for Modern Agro-Industry Technology Research System, China (CARS-3)
文摘To understand the contribution of ear photosynthesis to grain yield and its response to water supply in the improvement of winter wheat, 15 cultivars released from 1980 to 2012 in North China Plain(NCP) were planted under rainfed and irrigated conditions from 2011 to 2013, and the ear photosynthesis was tested by ear shading. During the past 30 years, grain yield significantly increased, the flag leaf area slightly increased under irrigated condition but decreased significantly under rainfed condition, the ratio of grain weight:leaf area significantly increased, and the contribution of ear photosynthesis to grain yield changed from 33.6 to 64.5% and from 32.2 to 57.2% under rainfed and irrigated conditions, respectively. Grain yield, yield components, and ratio of grain weight:leaf area were positively related with contribution of ear photosynthesis. The increase in grain yield in winter wheat was related with improvement in ear photosynthesis contribution in NCP, especially under rainfed condition.
基金supported by the National Key Technology R&D Program of China (2015BAD22B01)the Plan 111 of the Ministry of Education (B12007)+1 种基金the National Natural Science Foundation of China (31500320)Special Funds of Scientific Research Programs of State Key Laboratory of Soil Erosion and Dryland Farming on the Loess Plateau (A314021403-C5)
文摘In wheat, the ear is one of the main photosynthetic contributors to grain filling under drought stress conditions. In order to determine the relationship between stomatal characteristics and plant drought resistance, photosynthetic and stomatal characteristics and water use efficiency(WUE) were studied in two wheat cultivars: the drought-resistant cultivar ‘Changhan 58' and the drought-sensitive cultivar ‘Xinong 9871'. Plants of both cultivars were grown in pot conditions under well-watered(WW) and water-stressed(WS) conditions. In both water regimes,‘Changhan 58' showed a significantly higher ear photosynthetic rate with a lower rate of variation and a significantly higher percentage variation of transpiration compared to control plants at the heading stage under WS conditions than did ‘Xinong 9871' plants. Moreover,‘Changhan 58' showed lower stomatal density(SD) and higher stomatal area per unit organ area(A) under both water conditions. Water stress decreased SD, A, and stomatal width(SW), and increased stomatal length in flag leaves(upper and lower surfaces) and ear organs(awn, glume,lemma, and palea), with the changes more pronounced in ear organs than in flag leaves.Instantaneous WUE increased slightly, while integral WUE improved significantly in both cultivars. Integral WUE was higher in ‘Changhan 58', and increased by a greater amount, than in‘Xinong 9871'. These results suggest that drought resistance in ‘Changhan 58' is regulated by stomatal characteristics through a decrease in transpiration rate in order to improve integral WUE and photosynthetic performance, and through sustaining a higher ear photosynthetic rate, therefore enhancing overall drought-resistance.
基金This work was supported by the State Key Basic Research and Development Plan(G1998010100)the State Natural Science Fund(30270780)the State"Tenth Five Year"Project(2001BA507A-09)of China
文摘The activities of RuBPC and C4 photosynthetic enzymes in ear and flag leaf blade were examined in wheat. The results showed that photosynthesis of ear was less sensitive to soil drought than that of flag leaf, and decrease of CO2 assimilation in flag leaf blade with water stress was more than that in ear. Compared with flag leaf, ear organs (awn, glume and lemma) had higher C4 enzyme activities and lower RuBPC activity. Under moderate water-stress, the increase of C4 enzyme activities was induced, and the increase was higher in ear than in flag leaf. Under severe water-stress, relatively higher C4 enzyme activities were still maintained in ear, rather than that in flag leaf. It suggests that high activities of C4 enzymes in ear may contribute to its high tolerance of photosynthesis to water-stress.
文摘小麦麦穗的高效计数对快速、准确掌握小麦产量具有重要意义。无人机由于具有效率高、成本低等特点被广泛应用于大田小麦信息的采集。但已有的用于小麦麦穗计数的深度学习模型结构复杂、参数量大,不能直接部署在存储空间有限的无人机的边缘设备上。针对这一问题,提出了一种融合剪枝策略和知识蒸馏的模型压缩方法,基于YOLOv5s模型构建了一种轻量化模型,并设计了面向无人机边缘计算的小麦麦穗计数轻量化方案。试验结果表明,经过模型剪枝和知识蒸馏轻量化处理的YOLOv5s模型,在小麦计数任务上的计数准确率为93.3%,模型的mAP(mean Average Precision,平均精度均值)达到94.4%,模型大小缩小了约76%,模型参数量减少了79.61%。因此,模型在保持较高的计数准确率的同时将会占用更少的计算资源和存储空间,显著的压缩效果使模型可以部署在无人机的边缘设备上,为小麦麦穗的实时计数提供了可能。
文摘小麦的最终产量可由单位面积的小麦麦穗数侧面反映,为了快速准确统计小麦麦穗数,该研究给出一种在单幅图像上利用深度卷积神经网络估计田间麦穗密度图并进行麦穗计数的方法。首先对采集的田间小麦图像进行直方图均衡化及阈值分割预处理,以减少图像中光照及一些复杂背景对计数的影响;然后根据灌浆期田间小麦图像麦穗密集的特点,引入拥挤场景识别网络(Congested Scene Recognition Network,CSRNet)构建麦穗密度图估计模型,并采用迁移学习方法,利用小麦图像公开数据集对模型进行预训练,再用所采集的小麦图像数据集进行模型参数调整和优化;利用得到的模型生成单幅小麦图像的麦穗密度图,根据密度图中所有密度值的总和对图像进行麦穗计数。最后根据对单幅麦穗图像的试验数据,构建田间麦穗计数函数模型,实现田间小麦麦穗数估计。通过对所采集的安农170、苏麦188、乐麦608和宁麦24这4个品种共296幅小麦图像进行试验,平均绝对误差(Mean Absolute Error,MAE)和均方根误差(Root Mean Squared Error,RMSE)分别为16.44和17.89,4个品种小麦的麦穗计数值与真实值的决定系数R^2均在0.9左右,表明该方法对单幅图像小麦麦穗计数精度较高。此外,通过对田间小麦麦穗数进行估计试验,结果表明,随面积的增大麦穗估计的误差越小,研究结果可以为小麦的产量自动估计提供参考。