It was generally assumed that the accumulation of vegetative storage protein (VSP) in poplar trees and/or temperate hardwoods did not occur in spring. To test this assumption, the accumulation of the 32-kDa VSP and th...It was generally assumed that the accumulation of vegetative storage protein (VSP) in poplar trees and/or temperate hardwoods did not occur in spring. To test this assumption, the accumulation of the 32-kDa VSP and the differential expression of a gene encoding for the protein in poplars were investigated using light and electron microscopy, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and reverse transcription-polymerase chain reaction (RT-PCR). We report, for the first time, that poplar trees initiated VSP accumulation in new shoots during the development of new shoots in spring under conditions of high temperature and long days. The amount of 32-kDa VSP increased gradually in the stem of new shoots and in two-year-old branches, but there were no detectable changes in its abundance in the bark tissues of trunks during new shoot development. Based on the presence of a 286-bp DNA fragment that is identical to the VSP gene bspA, encoding for the 32-kDa VSP in Populus deltoids Bartr. ex Marsh., the differential expression of the 32-kDa VSP gene in P. canadensis Moench was revealed by RT-PCR. The results indicated that the 32-kDa VSP gene was expressed strongly in new shoots, relative weakly in two-year-old branches and was not expressed in the trunk during new shoot development. This pattern of VSP accumulation and VSP gene space-time differential expression may be an important mechanism by which stored nitrogen compounds are used preferentially to exogenously available nitrogen and, in addition, the dynamic pattern may also have a role in the regulation of nitrogen metabolism, especially nitrogen uptake by the roots.展开更多
【目的】采用机器视觉技术开展柑橘梢期的智能感知技术研究,以解决背景与目标颜色相似造成识别精度低的问题,实现柑橘梢期自动监测,探索算法的改进方法。【方法】根据不同卷积层提取特征的特点与不同注意力机制的作用,提出了一种基于多...【目的】采用机器视觉技术开展柑橘梢期的智能感知技术研究,以解决背景与目标颜色相似造成识别精度低的问题,实现柑橘梢期自动监测,探索算法的改进方法。【方法】根据不同卷积层提取特征的特点与不同注意力机制的作用,提出了一种基于多注意力机制改进的YOLOX-Nano智能识别模型,建立多元化果园数据集并进行预训练。【结果】改进的YOLOX-Nano算法使用果园数据集作为预训练数据集后,各类别平均精度的平均值(Mean average precision,mAP)达到88.07%。与YOLOV4-Lite系列模型相比,本文提出的改进模型在使用较少的参数和计算量的情况下,识别精度有显著的提升,mAP分别比YOLOV4-MobileNetV3和YOLOV4-GhostNet提升6.58%和6.03%。【结论】改进后的模型在果园监测终端的轻量化部署方面更具有优势,为农情实时感知和智能监测提供了可行的数据和技术解决方案。展开更多
文摘It was generally assumed that the accumulation of vegetative storage protein (VSP) in poplar trees and/or temperate hardwoods did not occur in spring. To test this assumption, the accumulation of the 32-kDa VSP and the differential expression of a gene encoding for the protein in poplars were investigated using light and electron microscopy, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and reverse transcription-polymerase chain reaction (RT-PCR). We report, for the first time, that poplar trees initiated VSP accumulation in new shoots during the development of new shoots in spring under conditions of high temperature and long days. The amount of 32-kDa VSP increased gradually in the stem of new shoots and in two-year-old branches, but there were no detectable changes in its abundance in the bark tissues of trunks during new shoot development. Based on the presence of a 286-bp DNA fragment that is identical to the VSP gene bspA, encoding for the 32-kDa VSP in Populus deltoids Bartr. ex Marsh., the differential expression of the 32-kDa VSP gene in P. canadensis Moench was revealed by RT-PCR. The results indicated that the 32-kDa VSP gene was expressed strongly in new shoots, relative weakly in two-year-old branches and was not expressed in the trunk during new shoot development. This pattern of VSP accumulation and VSP gene space-time differential expression may be an important mechanism by which stored nitrogen compounds are used preferentially to exogenously available nitrogen and, in addition, the dynamic pattern may also have a role in the regulation of nitrogen metabolism, especially nitrogen uptake by the roots.
文摘【目的】采用机器视觉技术开展柑橘梢期的智能感知技术研究,以解决背景与目标颜色相似造成识别精度低的问题,实现柑橘梢期自动监测,探索算法的改进方法。【方法】根据不同卷积层提取特征的特点与不同注意力机制的作用,提出了一种基于多注意力机制改进的YOLOX-Nano智能识别模型,建立多元化果园数据集并进行预训练。【结果】改进的YOLOX-Nano算法使用果园数据集作为预训练数据集后,各类别平均精度的平均值(Mean average precision,mAP)达到88.07%。与YOLOV4-Lite系列模型相比,本文提出的改进模型在使用较少的参数和计算量的情况下,识别精度有显著的提升,mAP分别比YOLOV4-MobileNetV3和YOLOV4-GhostNet提升6.58%和6.03%。【结论】改进后的模型在果园监测终端的轻量化部署方面更具有优势,为农情实时感知和智能监测提供了可行的数据和技术解决方案。
