Chemical weeding in dry direct seeding fields of single cropping middle-late rice was studied in Huida vegetable farm of Huizhou City in 2012. The main treatment was herbicide（ pretilachlor ＋ bensulfuron-methyl,Yang...Chemical weeding in dry direct seeding fields of single cropping middle-late rice was studied in Huida vegetable farm of Huizhou City in 2012. The main treatment was herbicide（ pretilachlor ＋ bensulfuron-methyl,Yangguo and butachlor）,and the sub-treatment was application method（ soil treatments,seedling treatment and integrated treatment）. The results showed that 80 g pretilachlor ＋ bensulfuron-methyl（ 36% pretilachlor ＋ 4% bensulfuron-methyl） diluted with 50 kg water could be sprayed or 200 g Yangguo（ 23. 9% butachlor ＋ 1. 1% bensulfuron-methyl） mixed with 15 kg sandy soil could be broadcasted per 667 m2 on the sowing day or the second day under moist condition of soil,which could effectively control weeds in dry direct seeding fields of single cropping middle-late rice.展开更多
为解决光线遮蔽、藻萍干扰以及稻叶尖形状相似等复杂环境导致稻田杂草识别效果不理想问题,该研究提出一种基于组合深度学习的杂草识别方法。引入MSRCP(multi-scale retinex with color preservation)对图像进行增强,以提高图像亮度及对...为解决光线遮蔽、藻萍干扰以及稻叶尖形状相似等复杂环境导致稻田杂草识别效果不理想问题,该研究提出一种基于组合深度学习的杂草识别方法。引入MSRCP(multi-scale retinex with color preservation)对图像进行增强,以提高图像亮度及对比度;加入ViT分类网络去除干扰背景,以提高模型在复杂环境下对小目标杂草的识别性能。在YOLOv7模型中主干特征提取网络替换为GhostNet网络,并引入CA注意力机制,以增强主干特征提取网络对杂草特征提取能力及简化模型参数计算量。消融试验表明:改进后的YOLOv7模型平均精度均值为88.2%,较原YOLOv7模型提高了3.3个百分点,参数量减少10.43 M,计算量减少66.54×109次/s。识别前先经过MSRCP图像增强后,与原模型相比,改进YOLOv7模型的平均精度均值提高了2.6个百分点,光线遮蔽、藻萍干扰以及稻叶尖形状相似的复杂环境下平均精度均值分别提高5.3、3.6、3.1个百分点,加入ViT分类网络后,较原模型平均精度均值整体提升了4.4个百分点,光线遮蔽、藻萍干扰一级稻叶尖形状相似的复杂环境下的平均精度均值较原模型整体提升了6.2、6.1、5.7个百分点。ViT-改进YOLOv7模型的平均精度均值为92.6%,相比于YOLOv5s、YOLOXs、MobilenetV3-YOLOv7、YOLOv8和改进YOLOv7分别提高了11.6、10.1、5.0、4.2、4.4个百分点。研究结果可为稻田复杂环境的杂草精准识别提供支撑。展开更多
[Objective] The aim was to explore the prevention effects of pyrazosul-furon-ethyl·quinclorac 50% WP on stem and leaf of dry direct seeded rice. [Method] Totaling 7 drug doses were set in the test and the effect...[Objective] The aim was to explore the prevention effects of pyrazosul-furon-ethyl·quinclorac 50% WP on stem and leaf of dry direct seeded rice. [Method] Totaling 7 drug doses were set in the test and the effects were observed 5, 10, 15 and 30 d after drug application, with the remained weeds surveyed 15 and 30 d after drug application. [Result] The test showed pyrazosulfuron-ethyl·quinclorac 50%WP was safe to rice as treatment agent of stem and leaf, and rice yield increased dramatical y, with the growth rate in 2.1%-49.3%. After the drug application, the prevention effects on Echinochloa crusgal i, Ludwimlia prostrata, Monochoria vamli-nalis, and Zong grass were 92.2%, 96.6%, 93.5%, and 94.8%, and the effects on fresh weight were 94.8%, 97.1%, 93.3% and 94.9%, respectively. [Conclusion] Pyra-zosulfuron-ethyl·quinclorac 50% WP can be taken as treatment agents of stem and leaf of dry direct seeded rice.展开更多
为明确38%敌稗·唑酰草胺·异草松乳油对直播水稻田一年生杂草的防除效果及其对水稻生长的安全性。参照《农药田间药效试验准则(一)》于2023年在水稻3~4叶期、杂草2~4叶期采用茎叶喷雾法开展大田试验。38%敌稗·唑酰草胺...为明确38%敌稗·唑酰草胺·异草松乳油对直播水稻田一年生杂草的防除效果及其对水稻生长的安全性。参照《农药田间药效试验准则(一)》于2023年在水稻3~4叶期、杂草2~4叶期采用茎叶喷雾法开展大田试验。38%敌稗·唑酰草胺·异草松乳油以有效成分用药量684.0~855.0 g a.i./hm^(2)进行茎叶喷雾,可有效防除稗草、千金子、丁香蓼、异型莎草等一年生杂草,药后40 d株防效89.7%~94.5%,鲜重防效92.4%~96.9%。38%敌稗·唑酰草胺·异草松乳油能够有效防除直播水稻田一年生杂草,且对水稻安全,杀草速度较快。展开更多
基金Supported by National Science and Technology Support Program(2007BAD89B14)Special Fund for Agro-scientific Research in the PublicInterest(00803028)+3 种基金Major Technical Research Project of Ministry of Agriculture for Agricultural Structure Adjustment(06-03-07B)Project ofGuangdong Provincial Finance Department(YCY[2005]No.11,YCJ[2006]No.187)Agricultural Research Project of Guangdong ProvincialScience and Technology Department(2005B20101001)Special Fund forAgro-scientific Research in the Public Interest(201103001)
文摘Chemical weeding in dry direct seeding fields of single cropping middle-late rice was studied in Huida vegetable farm of Huizhou City in 2012. The main treatment was herbicide（ pretilachlor ＋ bensulfuron-methyl,Yangguo and butachlor）,and the sub-treatment was application method（ soil treatments,seedling treatment and integrated treatment）. The results showed that 80 g pretilachlor ＋ bensulfuron-methyl（ 36% pretilachlor ＋ 4% bensulfuron-methyl） diluted with 50 kg water could be sprayed or 200 g Yangguo（ 23. 9% butachlor ＋ 1. 1% bensulfuron-methyl） mixed with 15 kg sandy soil could be broadcasted per 667 m2 on the sowing day or the second day under moist condition of soil,which could effectively control weeds in dry direct seeding fields of single cropping middle-late rice.
文摘为解决光线遮蔽、藻萍干扰以及稻叶尖形状相似等复杂环境导致稻田杂草识别效果不理想问题,该研究提出一种基于组合深度学习的杂草识别方法。引入MSRCP(multi-scale retinex with color preservation)对图像进行增强,以提高图像亮度及对比度;加入ViT分类网络去除干扰背景,以提高模型在复杂环境下对小目标杂草的识别性能。在YOLOv7模型中主干特征提取网络替换为GhostNet网络,并引入CA注意力机制,以增强主干特征提取网络对杂草特征提取能力及简化模型参数计算量。消融试验表明:改进后的YOLOv7模型平均精度均值为88.2%,较原YOLOv7模型提高了3.3个百分点,参数量减少10.43 M,计算量减少66.54×109次/s。识别前先经过MSRCP图像增强后,与原模型相比,改进YOLOv7模型的平均精度均值提高了2.6个百分点,光线遮蔽、藻萍干扰以及稻叶尖形状相似的复杂环境下平均精度均值分别提高5.3、3.6、3.1个百分点,加入ViT分类网络后,较原模型平均精度均值整体提升了4.4个百分点,光线遮蔽、藻萍干扰一级稻叶尖形状相似的复杂环境下的平均精度均值较原模型整体提升了6.2、6.1、5.7个百分点。ViT-改进YOLOv7模型的平均精度均值为92.6%,相比于YOLOv5s、YOLOXs、MobilenetV3-YOLOv7、YOLOv8和改进YOLOv7分别提高了11.6、10.1、5.0、4.2、4.4个百分点。研究结果可为稻田复杂环境的杂草精准识别提供支撑。
文摘[Objective] The aim was to explore the prevention effects of pyrazosul-furon-ethyl·quinclorac 50% WP on stem and leaf of dry direct seeded rice. [Method] Totaling 7 drug doses were set in the test and the effects were observed 5, 10, 15 and 30 d after drug application, with the remained weeds surveyed 15 and 30 d after drug application. [Result] The test showed pyrazosulfuron-ethyl·quinclorac 50%WP was safe to rice as treatment agent of stem and leaf, and rice yield increased dramatical y, with the growth rate in 2.1%-49.3%. After the drug application, the prevention effects on Echinochloa crusgal i, Ludwimlia prostrata, Monochoria vamli-nalis, and Zong grass were 92.2%, 96.6%, 93.5%, and 94.8%, and the effects on fresh weight were 94.8%, 97.1%, 93.3% and 94.9%, respectively. [Conclusion] Pyra-zosulfuron-ethyl·quinclorac 50% WP can be taken as treatment agents of stem and leaf of dry direct seeded rice.
文摘为明确38%敌稗·唑酰草胺·异草松乳油对直播水稻田一年生杂草的防除效果及其对水稻生长的安全性。参照《农药田间药效试验准则(一)》于2023年在水稻3~4叶期、杂草2~4叶期采用茎叶喷雾法开展大田试验。38%敌稗·唑酰草胺·异草松乳油以有效成分用药量684.0~855.0 g a.i./hm^(2)进行茎叶喷雾,可有效防除稗草、千金子、丁香蓼、异型莎草等一年生杂草,药后40 d株防效89.7%~94.5%,鲜重防效92.4%~96.9%。38%敌稗·唑酰草胺·异草松乳油能够有效防除直播水稻田一年生杂草,且对水稻安全,杀草速度较快。
文摘为明确18%噁唑酰草胺·噁嗪草酮乳油对直播水稻田一年生禾本科杂草的防除效果及其对水稻的安全性。参照GB/T 17980.40—2000《农药田间药效试验准则(一)》于2023年在水稻2.5~4叶期,禾本科杂草2~4叶期采用茎叶喷雾法开展大田试验。试验结果显示18%噁唑酰草胺·噁嗪草酮乳油有效成分用药量108.0~135.0 g a.i./hm^(2),可有效防除稗草、马唐、千金子等一年生禾本科杂草,药后30 d株和鲜重防效分别为89.07%~92.51%和92.00%~94.13%。18%噁唑酰草胺·噁嗪草酮乳油能够有效防除直播水稻田一年生禾本科杂草,且对水稻生长安全,杀草速度较快。
