Transgenic insect-resistant cotton is being increasingly planted in Xinjiang cotton-planting regions, where geographical climate conditions and species composition of pests and natural enemies are greatly unique in Ch...Transgenic insect-resistant cotton is being increasingly planted in Xinjiang cotton-planting regions, where geographical climate conditions and species composition of pests and natural enemies are greatly unique in China. Limited studies have been conducted on the ecological impacts of transgenic insect-resistant cotton, especially for transgenic double genes (Bt+CpTI) cotton, in this region. In this study, the potential effects of transgenic Bt+CpTI cotton on the seasonal abundance of non-target pests and predators were assessed from 2009 to 2011 in Korla, Xinjiang. The results showed that species composition and seasonal abundance of 5 groups of pests and 5 groups of predators were not significantly different between transgenic Bt+CpTI cotton and non-transgenic cotton every year. It suggests that transgenic Bt+CpTI cotton per se does not affect the population dynamics of non-target pests and predators on this crop in Xinjiang.展开更多
针对水稻病害检测精度低、易发生漏检和误检等问题,提出了改进YOLO第8常规版(you only look once version 8 normal,YOLOv8n)水稻叶片病害识别检测方法。以原始YOLOv8n算法为基准,在主干网络中使用下采样操作卷积(a convolutional block...针对水稻病害检测精度低、易发生漏检和误检等问题,提出了改进YOLO第8常规版(you only look once version 8 normal,YOLOv8n)水稻叶片病害识别检测方法。以原始YOLOv8n算法为基准,在主干网络中使用下采样操作卷积(a convolutional block for down-sampling,ADown)模块,减少特征信息的丢失,并在主干网络与颈部网络之间引入基于挤压-激励(squeeze-and-excitation,SE)的注意力机制模块,提高网络的特征融合能力;同时,设计出共享参数检测头,增加检测任务的感受野;此外,使用顺序证据加权插值算法的交并比(weighted interpolation of sequential evidence for intersection over union,WIoU)损失函数,进一步提升网络的检测性能。在水稻病害数据集上进行大量实验,结果表明,与原始YOLOv8n算法相比,改进YOLOv8n算法的精确率和平均精确率均值分别提升了6.59%和6.86%;改进YOLOv8n算法能够满足水稻叶片病害识别对速度及检测精度的需求,同时提高了对小目标和密集目标的检测能力,从而减少了漏检和误检的情况。改进YOLOv8n算法与目前主流算法相比在检测速度和精度上具有一定优势,检测速度是YOLO第7版(YOLO version 7,YOLOv7)算法的3.61倍,平均精确率均值提高了8.63%。该研究能够为水稻智能化种植管理提供一定的参考。展开更多
针对YOLOv5(you only look once version five)模型在农作物害虫密集目标上的检测效果无法满足实际需求,以及训练过程中模型收敛速度较慢等问题,该研究提出了融入全局响应归一化(global response normalization,GRN)注意力机制的YOLOv5...针对YOLOv5(you only look once version five)模型在农作物害虫密集目标上的检测效果无法满足实际需求,以及训练过程中模型收敛速度较慢等问题,该研究提出了融入全局响应归一化(global response normalization,GRN)注意力机制的YOLOv5农作物害虫识别模型(YOLOv5-GRNS)。设计了融入GRN注意力机制的编码器(convolution three,C3)模块,提高对密集目标的识别精度;利用形状交并比(shape intersection over union,SIoU)损失函数提高模型收敛速度和识别精度;在公开数据集IP102(insect pests 102)的基础上,筛选出危害陕西省主要农作物的8种害虫类型,构建了新数据集IP8-CW(insect pests eight for corn and wheat)。改进后的模型在新IP8-CW和完整的IP102两种数据集上进行了全面验证。对于IP8-CW,全类别平均准确率(mean average precision,mAP)mAP@.5和mAP@.5:.95分别达到了72.3%和47.0%。该研究还对YOLOv5-GRNS模型进行了类激活图分析,不仅从识别精度,而且从可解释性的角度,验证了对农作物害虫、尤其是密集目标的优秀识别效果。此外,模型还兼具参数量少、运算量低的优势,具有良好的嵌入式设备应用前景。展开更多
基金the 973 Program (2001CB109004and 2007CB109202)the Key Projects for Breeding Genetically Modified Organisms of China (2011ZX0811-002 and 2009ZX08011-008B)
文摘Transgenic insect-resistant cotton is being increasingly planted in Xinjiang cotton-planting regions, where geographical climate conditions and species composition of pests and natural enemies are greatly unique in China. Limited studies have been conducted on the ecological impacts of transgenic insect-resistant cotton, especially for transgenic double genes (Bt+CpTI) cotton, in this region. In this study, the potential effects of transgenic Bt+CpTI cotton on the seasonal abundance of non-target pests and predators were assessed from 2009 to 2011 in Korla, Xinjiang. The results showed that species composition and seasonal abundance of 5 groups of pests and 5 groups of predators were not significantly different between transgenic Bt+CpTI cotton and non-transgenic cotton every year. It suggests that transgenic Bt+CpTI cotton per se does not affect the population dynamics of non-target pests and predators on this crop in Xinjiang.
文摘针对水稻病害检测精度低、易发生漏检和误检等问题,提出了改进YOLO第8常规版(you only look once version 8 normal,YOLOv8n)水稻叶片病害识别检测方法。以原始YOLOv8n算法为基准,在主干网络中使用下采样操作卷积(a convolutional block for down-sampling,ADown)模块,减少特征信息的丢失,并在主干网络与颈部网络之间引入基于挤压-激励(squeeze-and-excitation,SE)的注意力机制模块,提高网络的特征融合能力;同时,设计出共享参数检测头,增加检测任务的感受野;此外,使用顺序证据加权插值算法的交并比(weighted interpolation of sequential evidence for intersection over union,WIoU)损失函数,进一步提升网络的检测性能。在水稻病害数据集上进行大量实验,结果表明,与原始YOLOv8n算法相比,改进YOLOv8n算法的精确率和平均精确率均值分别提升了6.59%和6.86%;改进YOLOv8n算法能够满足水稻叶片病害识别对速度及检测精度的需求,同时提高了对小目标和密集目标的检测能力,从而减少了漏检和误检的情况。改进YOLOv8n算法与目前主流算法相比在检测速度和精度上具有一定优势,检测速度是YOLO第7版(YOLO version 7,YOLOv7)算法的3.61倍,平均精确率均值提高了8.63%。该研究能够为水稻智能化种植管理提供一定的参考。
文摘针对YOLOv5(you only look once version five)模型在农作物害虫密集目标上的检测效果无法满足实际需求,以及训练过程中模型收敛速度较慢等问题,该研究提出了融入全局响应归一化(global response normalization,GRN)注意力机制的YOLOv5农作物害虫识别模型(YOLOv5-GRNS)。设计了融入GRN注意力机制的编码器(convolution three,C3)模块,提高对密集目标的识别精度;利用形状交并比(shape intersection over union,SIoU)损失函数提高模型收敛速度和识别精度;在公开数据集IP102(insect pests 102)的基础上,筛选出危害陕西省主要农作物的8种害虫类型,构建了新数据集IP8-CW(insect pests eight for corn and wheat)。改进后的模型在新IP8-CW和完整的IP102两种数据集上进行了全面验证。对于IP8-CW,全类别平均准确率(mean average precision,mAP)mAP@.5和mAP@.5:.95分别达到了72.3%和47.0%。该研究还对YOLOv5-GRNS模型进行了类激活图分析,不仅从识别精度,而且从可解释性的角度,验证了对农作物害虫、尤其是密集目标的优秀识别效果。此外,模型还兼具参数量少、运算量低的优势,具有良好的嵌入式设备应用前景。